20 - World Journal of Gastroenterology

World Journal ofGastroenterology

ISSN 1007-9327 (print)ISSN 2219-2840 (online)

World J Gastroenterol 2021 May 28; 27(20): 2434-2663

Published by Baishideng Publishing Group Inc

WJG https://www.wjgnet.com I May 28, 2021 Volume 27 Issue 20

World Journal of

GastroenterologyW J GContents Weekly Volume 27 Number 20 May 28, 2021

REVIEW

Role of modern radiotherapy in managing patients with hepatocellular carcinoma2434

Chen LC, Lin HY, Hung SK, Chiou WY, Lee MS

Open reading frame 3 protein of hepatitis E virus: Multi-function protein with endless potential2458

Yang YL, Nan YC

Breakthroughs and challenges in the management of pediatric viral hepatitis2474

Nicastro E, Norsa L, Di Giorgio A, Indolfi G, D'Antiga L

MINIREVIEWS

Pancreatitis after endoscopic retrograde cholangiopancreatography: A narrative review2495

Ribeiro IB, do Monte Junior ES, Miranda Neto AA, Proença IM, de Moura DTH, Minata MK, Ide E, dos Santos MEL, Luz GO, Matuguma SE, Cheng S, Baracat R, de Moura EGH

RON in hepatobiliary and pancreatic cancers: Pathogenesis and potential therapeutic targets2507

Chen SL, Wang GP, Shi DR, Yao SH, Chen KD, Yao HP

Evolving role of endoscopy in inflammatory bowel disease: Going beyond diagnosis2521

Núñez F P, Krugliak Cleveland N, Quera R, Rubin DT

Deep learning for diagnosis of precancerous lesions in upper gastrointestinal endoscopy: A review2531

Yan T, Wong PK, Qin YY

State of machine and deep learning in histopathological applications in digestive diseases2545

Kobayashi S, Saltz JH, Yang VW

COVID-19 in normal, diseased and transplanted liver2576

Signorello A, Lenci I, Milana M, Grassi G, Baiocchi L

ORIGINAL ARTICLE

Basic Study

Upregulation of long noncoding RNA W42 promotes tumor development by binding with DBN1 in hepatocellular carcinoma

2586

Lei GL, Niu Y, Cheng SJ, Li YY, Bai ZF, Yu LX, Hong ZX, Liu H, Liu HH, Yan J, Gao Y, Zhang SG, Chen Z, Li RS, Yang PH

Retrospective Cohort Study

Understanding celiac disease monitoring patterns and outcomes after diagnosis: A multinational, retrospective chart review study

2603

Lundin KEA, Kelly CP, Sanders DS, Chen K, Kayaniyil S, Wang S, Wani RJ, Barrett C, Yoosuf S, Pettersen ES, Sambrook R, Leffler DA

WJG https://www.wjgnet.com II May 28, 2021 Volume 27 Issue 20

World Journal of GastroenterologyContents

Weekly Volume 27 Number 20 May 28, 2021

Development and validation of a prognostic model for patients with hepatorenal syndrome: A retrospective cohort study

2615

Sheng XY, Lin FY, Wu J, Cao HC

Observational Study

Inflammatory bowel disease in Tuzla Canton, Bosnia-Herzegovina: A prospective 10-year follow-up2630

Tulumović E, Salkić N, Tulumović D

META-ANALYSIS

Association between oral contraceptive use and pancreatic cancer risk: A systematic review and meta-analysis

2643

Ilic M, Milicic B, Ilic I

CASE REPORT

Cyclophosphamide-associated enteritis presenting with severe protein-losing enteropathy in granulomatosis with polyangiitis: A case report

2657

Sato H, Shirai T, Fujii H, Ishii T, Harigae H

WJG https://www.wjgnet.com III May 28, 2021 Volume 27 Issue 20

World Journal of GastroenterologyContents

Weekly Volume 27 Number 20 May 28, 2021

ABOUT COVER

Editorial Board Member of World Journal of Gastroenterology, Fernando J Castro, MD, AGAF, FACG, Gastroenterology Training Program Director, Cleveland Clinic Florida, 2950 Cleveland Clinic Blvd, Weston, FL 33331, United States. [email protected]

AIMS AND SCOPE

The primary aim of World Journal of Gastroenterology (WJG, World J Gastroenterol) is to provide scholars and readers from various fields of gastroenterology and hepatology with a platform to publish high-quality basic and clinical research articles and communicate their research findings online. WJG mainly publishes articles reporting research results and findings obtained in the field of gastroenterology and hepatology and covering a wide range of topics including gastroenterology, hepatology, gastrointestinal endoscopy, gastrointestinal surgery, gastrointestinal oncology, and pediatric gastroenterology.

INDEXING/ABSTRACTING

The WJG is now indexed in Current Contents®/Clinical Medicine, Science Citation Index Expanded (also known as SciSearch®), Journal Citation Reports®, Index Medicus, MEDLINE, PubMed, PubMed Central, and Scopus. The 2020 edition of Journal Citation Report® cites the 2019 impact factor (IF) for WJG as 3.665; IF without journal self cites: 3.534; 5-year IF: 4.048; Ranking: 35 among 88 journals in gastroenterology and hepatology; and Quartile category: Q2. The WJG’s CiteScore for 2019 is 7.1 and Scopus CiteScore rank 2019: Gastroenterology is 17/137.

RESPONSIBLE EDITORS FOR THIS ISSUE

Production Editor: Ji-Hong Liu; Production Department Director: Yun-Xiaojian Wu; Editorial Office Director: Ze-Mao Gong.

NAME OF JOURNAL INSTRUCTIONS TO AUTHORS

World Journal of Gastroenterology https://www.wjgnet.com/bpg/gerinfo/204

ISSN GUIDELINES FOR ETHICS DOCUMENTS

ISSN 1007-9327 (print) ISSN 2219-2840 (online) https://www.wjgnet.com/bpg/GerInfo/287

LAUNCH DATE GUIDELINES FOR NON-NATIVE SPEAKERS OF ENGLISH

October 1, 1995 https://www.wjgnet.com/bpg/gerinfo/240

FREQUENCY PUBLICATION ETHICS

Weekly https://www.wjgnet.com/bpg/GerInfo/288

EDITORS-IN-CHIEF PUBLICATION MISCONDUCT

Andrzej S Tarnawski, Subrata Ghosh https://www.wjgnet.com/bpg/gerinfo/208

EDITORIAL BOARD MEMBERS ARTICLE PROCESSING CHARGE

http://www.wjgnet.com/1007-9327/editorialboard.htm https://www.wjgnet.com/bpg/gerinfo/242

PUBLICATION DATE STEPS FOR SUBMITTING MANUSCRIPTS

May 28, 2021 https://www.wjgnet.com/bpg/GerInfo/239

COPYRIGHT ONLINE SUBMISSION

© 2021 Baishideng Publishing Group Inc https://www.f6publishing.com

© 2021 Baishideng Publishing Group Inc. All rights reserved. 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA

E-mail: [email protected] https://www.wjgnet.com

https://www.wjgnet.com/bpg/gerinfo/204

https://www.wjgnet.com/bpg/GerInfo/287




http://www.wjgnet.com/1007-9327/editorialboard.htm



https://www.f6publishing.com

mailto:[email protected]

https://www.wjgnet.com

WJG https://www.wjgnet.com 2434 May 28, 2021 Volume 27 Issue 20

World Journal of

GastroenterologyW J GSubmit a Manuscript: https://www.f6publishing.com World J Gastroenterol 2021 May 28; 27(20): 2434-2457

DOI: 10.3748/wjg.v27.i20.2434 ISSN 1007-9327 (print) ISSN 2219-2840 (online)

REVIEW

Role of modern radiotherapy in managing patients with hepatocellular carcinoma

Liang-Cheng Chen, Hon-Yi Lin, Shih-Kai Hung, Wen-Yen Chiou, Moon-Sing Lee

ORCID number: Liang-Cheng Chen 0000-0002-0657-5636; Hon-Yi Lin 0000-0001-8866-2997; Shih-Kai Hung 0000-0002-4945-528X; Wen-Yen Chiou 0000-0002-5541-6834; Moon-Sing Lee 0000-0001-6584-8668.

Author contributions: Chen LC, Lin HY, and Hung SK contributed to the present review's conception and design; Chiou WY and Lee MS collected relevant papers and contributed to initial data interpretation; Chen LC, Lin HY and Hung SK wrote the manuscript; all authors brainstormed and discussed bi-weekly in the context of department research meetings.

Supported by Dalin Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation.

Conflict-of-interest statement: The authors declare no conflict of interest.

Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works

Liang-Cheng Chen, Hon-Yi Lin, Shih-Kai Hung, Wen-Yen Chiou, Moon-Sing Lee, Department of Radiation Oncology, Dalin Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, Dalin, Chia-Yi 62247, Taiwan

Hon-Yi Lin, Shih-Kai Hung, Wen-Yen Chiou, Moon-Sing Lee, School of Medicine, Buddhist Tzu Chi University, Hualien 970, Taiwan

Hon-Yi Lin, Institute of Molecular Biology, National Chung Cheng University, Min-Hsiung, Chia-Yi 62102, Taiwan

Corresponding author: Shih-Kai Hung, MD, PhD, Assistant Professor, Chief Doctor, Department of Radiation Oncology, Dalin Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, No. 2 Mingsheng Road, Dalin, Chia-Yi 62247, Taiwan. [email protected]

AbstractHepatocellular carcinoma (HCC) is the most common type of primary liver cancer. Several treatment options are available for managing HCC patients, classified roughly as local, local-regional, and systemic therapies. The high post-monotherapy recurrence rate of HCC urges the need for the use of combined modalities to increase tumor control and patient survival. Different international guidelines offer treatment recommendations based on different points of view and classification systems. Radiotherapy (RT) is a well-known local-regional treatment modality for managing many types of cancers, including HCC. However, only some of these treatment guidelines include RT, and the role of combined modalities is rarely mentioned. Hence, the present study reviewed clinical evidence for the use of different combined modalities in managing HCC, focusing on modern RT's role. Modern RT has an increased utility in managing HCC patients, mainly due to two driving forces. First, technological advancement (e.g., stereotactic body radiotherapy and advanced proton-beam therapy) enables precise delivery of radiation to increase tumor control and reduce side effects in the surrounding normal tissue. Second, the boom in developing target therapies and checkpoint-blockade immunotherapy prolongs overall survival in HCC patients, re-emphasizing the importance of local tumor control. Remarkably, RT combines with systemic therapies to generate the systemic therapy augmented by radiotherapy effect, a benefit now being actively investigated.

Key Words: Hepatocellular carcinoma; Stereotactic body radiotherapies; Radiotherapy;


https://dx.doi.org/10.3748/wjg.v27.i20.2434

http://orcid.org/0000-0002-0657-5636

http://orcid.org/0000-0002-0657-5636

http://orcid.org/0000-0001-8866-2997

http://orcid.org/0000-0001-8866-2997

http://orcid.org/0000-0002-4945-528X

http://orcid.org/0000-0002-4945-528X

http://orcid.org/0000-0002-5541-6834

http://orcid.org/0000-0002-5541-6834

http://orcid.org/0000-0001-6584-8668

http://orcid.org/0000-0001-6584-8668


Chen LC et al. Modern radiotherapy to manage HCC


on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/Licenses/by-nc/4.0/

Manuscript source: Invited manuscript

Specialty type: Gastroenterology and hepatology

Country/Territory of origin: Taiwan

Peer-review report’s scientific quality classificationGrade A (Excellent): 0 Grade B (Very good): B Grade C (Good): 0 Grade D (Fair): 0 Grade E (Poor): 0

Received: January 26, 2021 Peer-review started: January 26, 2021 First decision: March 29, 2021 Revised: April 16, 2021 Accepted: April 26, 2021 Article in press: April 26, 2021 Published online: May 28, 2021

P-Reviewer: Ali Shah SI S-Editor: Gong ZM L-Editor: A P-Editor: Liu JH

Guideline; Combined treatment; Immunotherapy

©The Author(s) 2021. Published by Baishideng Publishing Group Inc. All rights reserved.

Core Tip: Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer and treatments outcome are often unsatisfactory. The high post-monotherapy recurrence rate points to the need to combine modalities, including radiotherapy (RT). The international guidelines from North America, Europe and Asia offer treatment recommendations based on different classification systems. However, not all of these treatment guidelines include RT and the role of combined modalities. Hence, the present study reviewed clinical evidence of different combined modalities in managing HCC, focusing on RT's role and especially modern advanced RT techniques and the interaction with game-changing immunotherapy.

Citation: Chen LC, Lin HY, Hung SK, Chiou WY, Lee MS. Role of modern radiotherapy in managing patients with hepatocellular carcinoma. World J Gastroenterol 2021; 27(20): 2434-2457URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2434.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2434

INTRODUCTIONHepatocellular carcinoma (HCC) is the most common primary malignancy of the liver, ranking as the sixth leading cause of cancer incidence and the fourth most common etiology of cancer death worldwide; overall, around 953000 cases were diagnosed in 2017 globally[1]. The majority of HCC patients are found in the Asia–Pacific region, and around 75%-90% of cases are associated with chronic viral hepatitis (mainly hepatitis B and C) infection[2-4]. Therefore, the management of HCC requires considering detailed information from three main domains together: Cancer character-istics, liver function reserve, and overall patient condition.

Several treatment modalities are available for managing HCC patients; these modalities are generally classified as local, local-regional, and systemic therapies. Local therapy includes surgical resection, ablative therapies, and radiotherapy (RT). Ablative therapies include radiofrequency ablation (RFA)[5-7], percutaneous ethanol injection therapy[8], cryoablation[9], and microwave[10]; these modalities can be performed through percutaneous, laparoscopic, or open approaches. Local-regional or so-called arterially directed therapies include trans-arterial embolization (TAE), trans-arterial chemoembolization (TACE)[5], drug-eluting beads (DEB)-TACE[11-13], trans-arterial radioembolization (TARE) with yttrium 90[14,15], and hepatic arterial infusion chemotherapy (HAIC)[16-18]. Systemic therapy includes target therapy (e.g., sorafenib[19], lenvatinib[20], and regorafenib[21]) and immunotherapy (e.g., nivolu-mab[22]); more notably, the use of several such types of agents has been investigated aggressively.

The preferred curative surgical modalities include liver resection and liver transplantation, generating a 5-year overall survival rate of up to 70%-75%[23,24]. RFA is another frequently used curative modality, with a 5-year overall survival rate of around 40%-70%[24]. However, local control of RFA is decreased in patients with large tumors, especially those larger than 3 cm[25,26]. It is recognized that only 10%-30% of HCC patients are candidates for curative surgical options because most HCCs are diagnosed at an intermediate or advanced stage[27,28]. Even though many treatment options can be chosen for HCC patients and each treatment modality has seen advancement in past decades, 5-year overall survival rates are still unsatisfactory, being less than 20%[29].

A high treatment failure (i.e., recurrence) rate is reported in the heterogeneous group of patients with intermediate- to advanced-stage HCC. Five-year tumor recurrence rates are more than 50% after liver resection and may be up to 80% after RFA[30]. This high recurrence rate includes patients who underwent TACE or some other singular local treatment modality. Of note, intrahepatic recurrence is the most common recurrence pattern. This unsatisfactory tumor control rate suggests that

http://creativecommons.org/Licenses/by-nc/4.0/



https://www.wjgnet.com/1007-9327/full/v27/i20/2434.htm




combining different locoregional therapies may improve treatment outcomes[31-33]. This review focuses mainly on RT's current role in managing HCC, not only as a monotherapy but also as an essential part of combined modalities, especially stereotactic body radiotherapy (SBRT)/stereotactic ablative radiotherapy (SABR).

CURRENT TREATMENT GUIDELINES FOR HCCCurrently, useful treatment guidelines for managing HCC include recommendations from the National Comprehensive Cancer Network (NCCN)[34], the American Association for the Study of Liver Diseases (AASLD)[35], the European Association for the Study of the Liver (EASL)[36], the Japan Society of Hepatology (JSH)-HCC[37], the Korean Liver Cancer Association (KLCA) and National Cancer Center (NCC)[38], the National Health and Family Planning Commission (NHFPC) of the People’s Republic of China HCC guideline[39], and the Taiwan Liver Cancer Association (TLCA)[40].

In practice, these guidelines implemented different classification systems to stratify HCC patients for appropriate management. For example, the NCCN grouped patients as potentially resectable/transplantable vs unresectable/inoperable to guide treatment options. The AASLD and EASL guidelines used the Barcelona clinic liver cancer (BCLC) stage to lead management recommendations. The KLCA and NCC guidelines used a modified Union for International Cancer staging system adapted from the Liver Cancer Study Group of Japan[41,42]. Both the JSH-HCC and TLCA guidelines adopted hepatic functional reserve, extra-hepatic metastasis, vascular invasion, tumor number, and tumor size to guide treatment choice using a step-by-step manner and the Chinese guideline added general health status to the previously-mentioned risk fact-ors[37,39,40]. Due to the variety of classification systems used in these different guidelines, we summarized treatment recommendations according to the BCLC stage in Figure 1.

COMBINED MODALITY FOR HCCNone of the guidelines mentioned above well declare the role of the combined treatment strategies used frequently in daily practice. Clinically, the choice of different combined modalities is based not only on guidelines or evidence, but also on the individual patient’s condition, the liver function preservation, the tumor character-istics, and the treatment perspective including the availability of resources within the facility and the therapeutic ratio. Overview of therapeutic options and the consid-eration behind the combination of different modalities for liver cancer are illustrated in Figure 2.

For very early and early stage HCC Some anatomic locations of HCC, such as tumors adjacent to the gallbladder, liver hilum, bowel, stomach, and other critical organs, may limit the use of RFA as an intervention[43]. On the other hand, RT can be delivered to tumors that arise from any location, so that it can compensate for or combine with RFA.

For intermediate stage HCC Many patients receive TACE as their first local-regional therapy. However, TACE alone seldom achieves satisfactory tumor control[44]. Therefore, several combined modalities have been reported to increase treatment outcomes, and they are subdivided into three main categories, as follows.

Local-regional plus local therapy: TACE combined with RFA achieves a complete response (CR) rate of 55-65% at the time of the first or second post-treatment check[31,32]. TACE combined with conventionally fractionated radiotherapy (CFRT) demonstrates a better 1-year survival rate than TACE alone[45]. Remarkably, TACE combined with SBRT shows promising results[46]. In contrast, the role of HAIC in conjunction with RT is still under investigation[47].

Local-regional plus another local-regional therapy: TACE combined with TARE has been reported as a safe and effective treatment modality for bi-lobar HCC[48].

Local-regional plus systemic therapy: This type of combination includes such options as TACE combined with target therapy[44,49,50], TARE combined with target



Figure 1 Treatment recommendations modified in different guidelines according to the Barcelona clinic liver cancer stage. HCC: Hepatocellular carcinoma; BCLC: Barcelona clinic liver cancer; PS: Performance status; TA(C)E:



Trans-arterial (chemo)embolization; TARE: Trans-arterial radioembolization; DEB-TACE: Trans-arterial chemoembolisation with drug-eluting beads; RT: Radiotherapy; SBRT: Stereotactic body radiation therapy; HAIC: Hepatic arterial infusion chemotherapy; NCCN: National Comprehensive Cancer Network; AASLD: American Association for the Study of Liver Diseases; EASL: European Association for the Study of the Liver; JSH: Japan Society of Hepatology; KLCA: Korean Liver Cancer Association; TLCA: Taiwan Liver Cancer Association; Tx: Treatments; VI+: Positive vascular or bile duct invasion.

therapy[51-53], TACE combined with immunotherapy[54-56], and TARE combined with immunotherapy[57].

For advanced HCC Only limited HCC patients are responsive to immune checkpoint inhibitors, and a combination of these with RT may enhance the immune response; this phenomenon is named the systemic therapy augmented by radiotherapy (STAR) effect[58]. Overall, RT or SBRT combined with other treatment modalities has been applied increasingly in HCC patients.

THE ROLE OF RT IN HCCRT was used as a salvage or palliative treatment in the past, and only a few guidelines mention the role of RT. However, in the modern era, RT is indicated across all stages (i.e., from very early to end-stage HCC)[34,35,37,40]. Notably, RT can be used as a single therapy or as an essential component of a combined modality. Current treatment recommendations based on the BCLC stage and RT's potential roles are summarized in Figure 3.

Different RT techniquesPhoton therapy: The most commonly available treatment beam of RT is photons. In managing patients with HCC, several photon-beam delivery systems of external-beam radiation therapy (EBRT) are clinical available, such as conventional fractionated RT (CFRT), hypo-fractionated RT (HFRT), and SBRT. CFRT is usually delivered with daily fractions of 1.8-2 Gy, and HFRT is characterized by a large daily dose (i.e., > 2 Gy) in the context of precise RT. Clinically, HFRT is a useful strategy for improving dose intensity and then tumor control. Both CFRT and HFRT can be delivered by using three-dimensional conformal radiotherapy (3DCRT), intensity-modulated radiotherapy (IMRT), and Volumetric-modulated Arc Therapy (VMAT).

Remarkably, SBRT, or so-called SABR, is an advanced technique of EBRT that delivers a very high dose of irradiation in a very precise way in a limited number of treatment fractions (i.e., usually 3-6 fractions and > 5 Gy per fraction) over a treatment course of 1-2 wk. For more focused and accurate delivery of SBRT, advancements across the whole RT department should be provided, including imaging, immobil-ization, target delineation, treatment planning, on-board image guidance, and



Figure 2 Overview of therapeutic options and the consideration behind the combination of different combined modalities for liver cancer. RT: Radiotherapy; TA(C)E: Trans-arterial (chemo)embolization; TARE: Trans-arterial radioembolization; DEB-TACE: Trans-arterial chemoembolisation with drug-eluting beads; HAIC: Hepatic arterial infusion chemotherapy.

respiratory motion management (RMM). Only advanced IMRT or VMAT with or without non-coplanar beam designs can be used for delivering SBRT. These advancements result in better dose distribution, deliver a higher dose within the tumor, and generate a rapid dose fall-off outside the target. Thus, SBRT can improve tumor control and reduce the irradiation dose to the surrounding normal tissue, to decrease RT toxicity. Owing to this double benefit of enhancing therapeutic gain, SBRT is highly recommended in managing HCC patients treated with curative intent.

For patients who cannot be treated successfully with SBRT, CFRT combined with two or more advanced irradiation techniques, such as combined VMAT and Simultan-eously Integrated inner-Escalated Boost (SIEB), may be helpful to achieve a better therapeutic gain (i.e., better tumor control with minimal RT toxicity) compared to conventional CFRT, including for elderly HCC patients who have inoperable disease[59].

Proton therapy: Charged particle irradiations, including proton-beam therapy (PBT) and carbon-ion RT, have unique dosimetric characteristics. That is, they eliminate the low-dose bath volume distal to the target area that is associated with photons. This elimination is because the characteristic Bragg peak of charged particles deposits irradiation energy mainly within the targeted tumor area and results in a near-zero dose beyond the end of its path[60]. Therefore, charge particle irradiation represents an excellent option for improving normal liver sparing and minimizing side effects such as radiation-induced liver disease (RILD). It also makes possible dose escalation for curing unresectable huge HCC.

Park et al[61] reported that dose escalation could enhance HCC tumor control. Kim et al[62] further confirmed that proton dose escalation is safe and effective; they suggested an EQD2 ≥ 78 Gy-equivalents (GyE) could be delivered to achieve reasonable tumor control. According to the tumor location, the University of Tsukuba proton team developed different PBT dose protocols[63-65]. Extrapolating on the concept of a lung cancer SBRT “No Fly Zone”[66], peripheral liver tumors located at > 2 cm from the hepatic portal region or gastrointestinal (GI) tract can be treated with



Figure 3 Current treatment recommendations based on the Barcelona clinic liver cancer stage and the potential roles of radiotherapy. HCC: Hepatocellular carcinoma; BCLC: Barcelona clinic liver cancer; PS: Performance status; TACE: trans-arterial chemoembolization; TARE: Trans-arterial radioembolization; HAIC: Hepatic arterial infusion chemotherapy; RT: Radiotherapy; CFRT: Conventionally fractionated radiotherapy; HFRT: Hypo-fractionated radiotherapy; SBRT: Stereotactic body radiation therapy; PVT: Portal vein thrombosis; STAR effect: Systemic therapy augmented by radiotherapy effect.

hypofractionated proton 66 GyE in 10 fractions. On the other hand, for tumors located within 2 cm adjacent to the hepatic portal region, small doses per fraction with 72.6 GyE in 22 fractions should be considered. For tumors located within 2 cm of the GI tract, 77.0 GyE in 35 fractions may be given[63-65].

Several studies report using PBT for localized HCC with excellent local control ranging from 80% to 100%, even for huge unresectable HCCs, due to dose escalation and sparing of more liver function[67-69]. Furthermore, Sanford et al[70] reported that the overall survival (OS) benefit of proton-RT over photon-RT might be due to decreasing the incidence of RILD. Hsieh et al[71] further identified the predictors of RILD in HCC patients treated with PBT beyond the conventional concept of minimizing the mean liver dose. A "volume-response" relationship between unirra-diated liver volume (ULV)/standard liver volume (SLV) and RILD was found: For Child-Pugh A patients, it was < 50%; for Child-Pugh B patients, it was < 30%[71].



Both photon and PBT can achieve high rates of local control with acceptable toxicity. However, PBT has better potential to deliver a higher dose while maximizing the volume of unirradiated liver. Clinically, the reduced normal liver dose achieved by PBT is not critically required for all patients, since some may benefit from a smaller irradiated target volume when normal liver constraints can be met. The 2018 Miami Liver Proton Therapy Conference reached the consensus that the patients who should be strongly prioritized for PBT include those with: At least Child-Pugh B cirrhosis, a high tumor-to-liver ratio (i.e., larger tumor size or smaller uninvolved liver volume), a greater number of tumors, or prior RT to the liver[72]. The dose comparison of proton therapy vs SBRT vs conventional RT for liver tumors are illustrated in Figure 4.

The role of RT in very early and early stage HCCVery early and early stage HCCs include those with BCLC classification 0-A, as follows: Carcinoma in situ; a single tumor of ≤ 2 cm, a single tumor of ≤ 5 cm, or three tumors of < 3 cm; and tumor burden laying within the Milan criteria, with Eastern Cooperative Oncology Group (ECOG) performance status (PS) 0 as well as Child-Pugh classification A-B[73]. For these HCC patients, although the standard of care is still surgery and RFA, definitive SBRT is a potential third curative treatment modality for medically inoperable, technically unresectable, and difficult to RFA conditions; more notably, it can serve as a bridge to liver transplantation[26,74].

The role of RT in intermediate stage HCCsIntermediate stage HCCs include those with BCLC classification B, as follows: Multi-tumors, a single tumor of > 5 cm, good patient condition (i.e., PS 0-1), as well as good liver reserve (i.e., Child-Pugh A-B). Tumor burden can be further subdivided, as follows: (1) Beyond Milan criteria but within the Up-to-7 criteria; and (2) Tumors beyond the Up-to-7 criteria[75]. For these patients, only limited cases can be treated by surgery or RFA. Several combined-modality approaches, including local therapy, local-regional therapy, and systemic therapy, have been reported in conjunction with CFRT and SBRT[76,77].

The role of RT in advanced stage HCCsAdvanced stage HCCs include those with BCLC classification C, with the criteria of portal vein invasion, inferior vena cava/heart invasion or thrombosis, lymph node metastasis, distant metastasis, and Child-Pugh A-B. SBRT or conventional RT may be applied in conjunction with other local, local-regional, and systemic therapies to serve as a potentially curative or palliative treatment[76-78].

The role of RT in terminal stage HCCsTerminal stage HCCs include those with BCLC classification D, with the criteria of Child-Pugh C or ECOG PS 3-4. For these patients, SBRT with careful planning is safe as a bridge to liver transplantation in selected patients with a Child-Pugh score of ≥ 8. Additionally, SBRT or conventional RT can be used to treat symptoms[79-81].

THE ROLE OF SBRT IN MANAGING PATIENTS WITH HCCSABR/SBRTRecently, clinical evidence has rapidly grown for the use of SBRT in managing all stage HCC patients, with curative, potentially curative, or palliative intent[33,77,78,82]. Prospective clinical trials have demonstrated that SBRT effectively treats HCC, resulting in satisfactory local control, ranging from 75% to 100% at 1 year and 65% to 100% at 2 years[33,77,78]. Local control of HCC using SBRT is typically defined as no progression or no recurrent disease within the irradiated field according to the Response Evaluation Criteria in Solid Tumors (RECIST) or its modification (mRECIST)[83-85]. Moreover, SBRT showed a benefit of limited toxicity, with a severe late toxicity rate of < 15%; thus, SBRT is considered a safe modality for treating elderly patients[86]. In the literature, most patients treated with SBRT have Child-Pugh A disease and limited lesions (often < 3 tumors), and delivering SBRT in Child-Pugh B patients increases toxicity rates[80]. However, if dose modification is done to meet the more conservative (i.e., strict) normal tissue constraints, SBRT may be allowed for patients with small HCCs with Child-Pugh B or those with relatively larger tumors with a Child-Pugh score of 7[80].



Figure 4 Proton therapy vs stereotactic body radiation therapy vs conventional radiotherapy for liver tumors. Dose distributions for a proton (left), stereotactic body radiation therapy (middle) and conventional radiotherapy (right) hepatocellular carcinoma radiotherapy plan are illustrated for comparison. RT: Radiotherapy; SBRT: Stereotactic body radiation therapy.

SBRT as an alternative treatment modality to ablation therapy (i.e., RFA)Ablative therapy is curative in treating small tumors (i.e., ≤ 3 cm) that locate in a feasible location, achieving excellent local control rates of around 70-90%; these results are similar to that of surgical resection[87]. Under such conditions, ablative therapy is considered an alternative treatment to surgical resection or liver transplantation.

Recently, SBRT showed comparable outcomes with RFA. Wahl et al[26] compared treatment outcomes between SBRT and RFA; they declared that both are effective for patients with inoperable HCC, with comparable freedom from local progression (FFLP) and comparable OS rates. The 1- and 2-year FFLP rates of SBRT were 97.4% and 83.8% vs 83.6% and 80.2% for RFA, respectively. The 1- and 2-year OS rates were also similar between the two treatment modalities. Remarkably, for larger tumors of ≥ 2 cm, SBRT demonstrated a better FFLP than that of RFA [hazard ratio (HR), 3.35; 95% confidence Interval (CI): 1.17-9.62; P = 0.025][26].

Hara et al[88] used propensity score matching (PSM) to assess the pre-treatment characteristics of BCLC stage, computed tomography (CT) status, and tumor size; they reported comparable 3-year OS rates between RFA and SBRT. Kim et al[89] also applied PSM to compare treatment results of SBRT and RFA. Two-year FFLP rates were 74.9% for the SBRT group and 64.9% for the RFA group. Taking these data together, SBRT demonstrates an emerging role as a curative treatment modality that is an alternative to ablative therapy for managing HCC patients.

SBRT as an alternative or adjuvant therapy to arterially directed therapiesAmong arterially directed therapies, TACE is the most widely used treatment modality for managing HCC patients with an intermediate stage, applied in 50%-60% of patients[90]. However, TACE alone demonstrates unsatisfactory local control. This unacceptably low response rate suggests the use of local therapy, such as SBRT, as an alternative or adjuvant therapy to improve local control.

SBRT as an alternative therapy to TACE: Sapir et al[91] used PSM with inverse treatment weighting probability to compare SBRT and TACE in HCC patients with 1-2 tumors. They found that SBRT demonstrated better 1- and 2-year local control rates when compared with TACE: 97% and 91% vs 47% and 23% (HR, 66.5; 95%CI: 18.99-233.0; P< 0.001), respectively. However, the difference in OS did not significantly differ between the two treatment modalities.

SBRT combined with TACE: Several retrospective studies and reviews have demonstrated increased local control rates by adding conventional RT to TACE[45,92,93]. However, studies using SBRT to replace conventional RT are scarce because SBRT is a new technique.

A pilot study and preliminary results of prospective studies also confirmed the safety and efficacy of SBRT in patients with HCC that failed to respond to TACE[94-96]. Kang et al[46] published a phase-II trial that enrolled 47 patients who



received TACE 1 to 5 times before SBRT. The result showed a good response, with a 2-year local control rate of 94.6%. Moreover, 38.3% of patients achieved a complete response within 6 mo. However, there is still no published data from head-to-head trials that compare TACE plus SBRT with TACE alone. Several clinical trials are ongoing (ClinicalTrials.gov Identifier: NCT02762266, NCT02323360, NCT02323360, NCT02794337, NCT02921139).

SBRT combined with TARE: Previously, there has been much concern over increasing radiation-related toxicity when using SBRT after TARE. Hardy-Abeloos et al[97] recently reported that TARE followed by SBRT has comparable safety and efficacy to TACE followed by SBRT.

SBRT as a bridge to liver transplantationOne of the preferred gold-standard treatments for managing HCC patients is liver transplantation, but only a few patients have a chance to receive a transplant due to an insufficient supply of donor livers[23]. Therefore, several local and local-regional therapies for HCC have been used to bridge care in patients seeking a transplant, to delay tumor progression[24]. Several studies have shown that SBRT may be an excellent alternative to conventional therapies as a bridge to transplantation[81,98]. Sapisochin et al[98] reported 1-, 3-, and 5-year actuarial patient survival rates from the time of listing of 83%, 61%, and 61% in the SBRT group, respectively, rates which were comparable with those of the TACE or RFA groups.

SBRT for macroscopic vascular invasion or portal vein thrombosisIn the past decades, two landmark randomized trials revealed that sorafenib yielded modest survival prolongation in patients with portal vein thrombosis (PVT)[99,100]. However, the response rate was unsatisfactory (only 2%). CFRT has been widely used in advanced HCC with macroscopic vascular invasion or PVT because it can be delivered regardless of tumor location, and major vessels demonstrate high radiation tolerance[24,78,79].

Rim et al[101] published a meta-analysis and systematic review which compared 3DCRT, TARE, and SBRT in patients with PVT. No significant differences in OS were observed among the three treatment modalities, but SBRT demonstrated the highest response rate of around 70%. More notably, toxicities of more than grade 3 were rare in the SBRT group[101]. These data revealed that SBRT could be safely applied in patients with PVT, with a better response rate than CFRT.

SBRT and sorafenibSBRT compared with sorafenib: Bettinger et al[102] used PSM to compare SBRT and sorafenib. SBRT showed a superior survival benefit to that of sorafenib: median overall survival was 17.0 mo (range, 10.8-23.2) vs 9.6 mo (range, 8.6-10.7), respectively, with or without adjusting for different baseline characteristics. A cost-effectiveness study also reported that SBRT had better cost-effectiveness than sorafenib for patients with advanced HCC[103].

SBRT combined with sorafenib: Brade et al[104] initiated a phase-I trial to evaluate the combination of sorafenib and SBRT. Sorafenib was delivered before, during, and after SBRT. The researchers found that concurrent use of SBRT with sorafenib significantly increased side effects, e.g., grade 3+ bowel toxicity and tumor rupture. Thus, they did not recommend using this combination outside of a clinical trial. A clinical trial (RTOG 1112) is ongoing to compare SBRT followed by sorafenib with sorafenib alone in patients with advanced HCC (ClinicalTrials.gov Identifier: NCT01730937).

SBRT combined with immunotherapy SBRT as an immunostimulator: Preclinical data demonstrated that RT could augment the intra-tumor cell surface expression of immunogenicity markers (e.g., dendritic cells) and enhance therapeutic efficacy[105-107]. Recently, immunotherapy combined with RT is an active field in managing HCC and has shown promising resu-lts[106,108,109]. SBRT may stimulate the release of tumor antigens and increase antigen-presenting cells to enhance the immune response to cancer cells provided by the immunotherapy[110]. This effect is also termed STAR[58] or “immunotherapy and stereotactic ablative radiotherapy (ISABR)”[111]. Investigations exploring in detail the underlying mechanisms of these effects are ongoing aggressively. Potential mechanism of SBRT combined with systemic therapy to induce the STAR effect (ISABR) for liver tumors is illustrated in Figure 5.



Figure 5 Potential mechanism of stereotactic body radiation therapy combined with systemic therapy to induce the systemic therapy augmented by radiotherapy effect (also known as immunotherapy and stereotactic ablative radiotherapy) for liver tumors. Stereotactic body radiation therapy (SBRT) induces antigen release and immunogenic cell death, activation of several transcription factors and signal pathways, as well as dendritic cell antigen presentation and maturation, resulting in proliferation of tumor-specific T cells and immune-mediated cytotoxicity. SBRT combined with Immune-checkpoint inhibitors augmented the tumoricidal effect by upregulates major histocompatibility complex and FAS on tumor cells, increasing susceptibility to T-cell-mediated cell death. MHC: Major histocompatibility complex; TCR: T cell receptor; FAS-L: FAS ligand.

SBRT is more immunogenic than conventional RT: SBRT is more immunogenic and has a beneficial outcome than the conventionally fractionated RT[112,113]. By using ablative SBRT in a mouse model, Lee et al[113] observed that a 20-Gy single fraction dose leads to a reduction in the tumor burden in both primary and distant metastases in a CD8+ T-cell–dependent fashion. However, conventionally fractionated irradiation showed a CD8+-depleted condition. In another preclinical study that compared soluble PD-L1 (sPD-L1) between the SBRT and conventional RT groups[114], Kim and colleagues found that the sPD-L1 Level increases persistently for 1 mo in the SBRT group. In comparison, it increased initially after irradiation but decreased after 1 mo in the conventional RT group.

Potential optimal SBRT dose and treatment sequences for induced immunogenicity: The optimal window of radiation immunogenicity is determined by the levels of double-strand DNA (dsDNA) vs Trex1[115,116]. SBRT doses up to 10-12 Gy (e.g., 8 Gy x 3 fractions) can up-regulate dsDNA accumulation in cancer cells via the cGAS/STING pathway, turning on RT-driven immune responses. However, higher doses above 12–18 Gy, such as 20-30 Gy in a single fraction, may induce the exonuclease TREX1, which down-regulates the immunogenicity by degrading cytosolic DNA, turning off RT-driven immune responses[115]. Otherwise, increasing the dose to higher than 10 Gy per fraction can rapidly reduce the tumor blood perfusion, leading to a tumoricidal effect via severe vascular damage[117].

The optimal sequence of SBRT and immunotherapy is still unclear. Both concurrent and sequential combination have been applied[118-120]. In other cancers, Wegner et al[121] found that OS was improved when immunotherapy was given more than 3 wk after initiating SBRT/SRS in patients with stage IV non-small-cell lung cancer. In a study designed by Tang et al[120], SBRT was given either concurrently or sequentially with ipilimumab in patients with advanced solid tumors. However, there are limited



studies specifically focused on HCC. Chiang et al[122] added nivolumab at 2 wk after completed SBRT and continuously applied every 2 wk until disease progression. Currently, the optimal timing of SBRT and immunotherapy remains under active investigation.

Clinical data and ongoing trials for SBRT combined with immunotherapy: Recently, two phase-II studies, CheckMate 040[123] and KEYNOTE-224[124], have shown that the anti-PD-1 immune checkpoint inhibitors of nivolumab and pembrolizumab demonstrate favorable tumor responses (15%-20%) in managing HCC patients. The two agents have been approved by the Food and Drug Administration (FDA) for patients previously treated with sorafenib. A phase-III randomized trial, KEYNOTE-240[125], then compared pembrolizumab and placebo in 413 patients previously treated with sorafenib. A median OS was 13.9 mo for pembrolizumab vs 10.6 mo for placebo (HR, 0.781; 95%CI: 0.611-0.998; P = 0.0238) and the median PFS was 3.0 vs 2.8 mo (HR, 0.718; 95%CI: 0.570-0.904; P = 0.0022). However, neither study endpoint reached statistical significance according to the trial-specified criteria (one-sided significance threshold, P = 0.0174, for the final analysis). Another phase-III trial, CheckMate 459, comparing nivolumab vs sorafenib as first-line treatment in advanced HCC, is ongoing (ClinicalTrials.gov Identifier: NCT02576509). Abstracts of CheckMate 459 in the 44th European Society for Medical Oncology Congress revealed that the primary endpoint of OS was statistically insignificant, but the objective response rate was double in the nivolumab group (15% vs 7%, respectively)[126].

Although the immune checkpoint inhibitor's objective response rate was higher than that of sorafenib, 15%-20% is still relatively insufficient[123-125]. Hence, the combination of immunotherapy with SBRT has been proposed to improve the efficacy of immunotherapy, to combine the direct tumoricidal effects with the immunogenic STAR effect[110,111,127,128]. As a novel combined treatment modality, only one published HCC study is currently available, revealing encouraging results. Chiang et al[122] reported 5 retrospective cases with unresectable HCC treated with SBRT 27.5-35 Gy in 5 fractions, followed by nivolumab. The median follow-up time was 14.9 mo. An objective response rate of 100% was reported, with 2 complete response and 3 partial responses. The 1-year OS and LC rates were both 100%. In addition, several phase-I or -II trials that combine SBRT and immunotherapy are ongoing (Clinical-Trials.gov Identifier: NCT03817736, NCT03316872, NCT03482102, and NCT03203304).

SBRT for metastatic disease SBRT becomes much more important in managing cancer patients with oligometastatic disease[129-134], including those with HCC[76]. Recently, a landmark trial applied SBRT to manage oligometastatic cancers, defined as primary controlled tumors with only 1-5 metastatic lesions, PS ≤ 1, and life expectancy > 6 mo[135]. Median OS of the SBRT group was better than that of the control group (41 vs 28 mo; HR, 0.57; 95%CI: 0.30-1.10; P = 0.090)[135]. Therefore, it is reasonable to treat more aggressively those HCC patients with oligometastasis, oligo-progression, and oligo-recurrence, i.e., treat with potentially curative intent by using SBRT if individual conditions allow.

How to combine SBRT?Clinically, managing HCC patients with a combination of different treatment modalities, including SBRT, remains a challenge. Information from several domains should be judged together carefully, such as the individual patient's condition, the tumor location/characteristics, liver function reservation, facility resources, and the irradiation techniques available. A multidisciplinary evaluation before initiating treatment is crucial[34]. Moreover, to individualize the treatment combination into the best available option, clinicians must discuss options with patients and their families, using shared decision making (SDM)[136].

COMBINED MODERN RT TECHNIQUES FOR PATIENTS IN WHOM SBRT CANNOT BE DONE SAFELYAs mentioned above, for vulnerable patients with unresectable bulky HCCs (i.e., > 5 cm), clinicians may hesitate to use SBRT because of the potentially greater risk of severe non-classic RILD or GI tract toxicity[61,137]. Under such conditions, a combination of modern irradiation techniques, such as VMAT, RMM[138], and modified simultaneously integrated boost (SIB)[139-142], has been reported to be useful.



In the literature, SIB dose prescription has been recommended for dose escalation[143]. The original SIB prescription delivers different homogenous doses to separate target regions in the same fraction numbers[144]. That is, the high-, intermediate-, and low-risk target volumes are simultaneously irradiated with high, intermediate, and low doses, respectively, per fraction (Figure 6A).

Recently, a modified SIB technique simultaneously applied a heterogeneous dose per fraction on peripheral (lower dose per fraction, e.g., traditional 200 centigray [cGy]) and intra-tumor zones (escalated higher dose per fraction, e.g., 240 cGy) (Figure 6B)[139-143]. Modified SIB simultaneously delivers an intra-tumor boost dose to the irradiated tumor's inner region to maximally enhance the possibility of tumor control in treating bulky tumors. This type of planned intra-tumor heterogenous dose distri-bution of modified SIB differs from the original SIB. This modified SIB has shown a potential role in managing bulky tumors, such as bulky pelvic tumor[139], retroperi-toneal mass[140], breast cancer[141], and liver tumor[142,143,145]. Recently, higher doses per fraction, e.g., 2.4 Gy on peripheral and 3 Gy on intra-tumor boost zones, have been prescribed for managing very bulky tumors[145].

To gain better tumor control without the cost of increasing irradiation-associated critical organ toxicity, a type of combined SIB and simultaneously integrated protection (SIP) was developed (Figure 6C)[146,147]. In conjunction with an SIB to the intra-tumor volume, SIP includes an attenuated dose per fraction on the overlapping sites of PTV and the extended critical organ volume, to gain a double benefit from clinical irradiation. Several unresectable bulky cancers have benefitted from this modern RT dose prescription technique, including HCC[142,146,148].

Following this line of dose prescription, a secondary-modified SIB technique (i.e., SIEB; Figure 6D) was developed[59]. SIEB secondly remodeled the modified SIB to further expand the double benefits of SIB in managing unresectable bulky tumors. That is, SIEB simultaneously maintained an intra-tumor escalated boost of the modified SIB to gain maximal tumor control (e.g., 220-240 cGy per fraction). Moreover, SIEB includes prescribing a planned attenuated peri-gross-tumor dose to all adjacent normal tissues (e.g., 120-150 cGy per fraction). Note that this intended protection is not limited only to critical organs/structures[59]. As a result, in theory, SIEB's therapeutic gain is further enhanced even over that of modified SIB (Figure 6B) or combined SIB and SIP (Figure 6C). However, prospective randomized clinical trials should be conducted to confirm the effectiveness of SIEB in managing patients with unresectable HCC.

One topic for further investigation in modified SIB and SIEB is whether the intra-tumor boost volume should be guided by using metabolic images, such as positron emission tomography (PET)[149,150]. For example, F-18-labeled fluoromisonidazole PET/CT images have been used to guide an SIB-escalated boost to the hypoxia tumor region[150]. However, using metabolic images to guide the intra-tumor boost volume has some limitations in daily clinical use, such as resource availability and the dynamic change in the PET-guided increased-uptake region during the treatment course of irradiation. As a result, multiple sessions of PET images and adaptive RT treatment planning may be required. Hence, in practice, it is feasible to deliver an intra-tumor boost to the geometric-central zone to maintain treatment efficacy, enhance tumor control, and minimize irradiation toxicity[59,143,145].

CONCLUSIONCurrently, two driving forces have come together to improve the treatment efficacy of RT in HCC. One is technological advancement, which enables the precise delivery of SBRT, increasing tumor control and reducing the side effects to the surrounding normal tissue. The other is the boom in the development of target therapies and checkpoint-blockade immunotherapy, which prolongs HCC patients' survival and re-emphasizes the importance of local tumor control. Currently, the role of RT in HCC treatment is actively being investigating in combination with systemic therapies to generate the STAR effect.

Remarkably, the development and use of combined modalities to increase liver reserve and patient tolerance may have the best chance to improve treatment outcomes. For HCC patients with any stage disease, RT plays a crucial role, whether delivered alone or in combination with other treatment modalities, because it is not limited by tumor location. Currently, a lack of level-III evidence is the main barrier to recommending SBRT as a standard of care in most international treatment guidelines. In this regard, several clinical trials of SBRT for HCC are ongoing. In the modern era,



Figure 6 Cartoon and case illustrations of simultaneously integrated boost, modified simultaneously integrated boost, combined simultaneously integrated boost and simultaneously integrated protection, and simultaneously integrated inner-escalated boost dose-prescription techniques. A: A simple cartoon figure representing original simultaneously integrated boost (SIB) dose prescription. Original SIB is prescribed in different doses per fraction to different target regions according to the risk levels of recurrence. For example, uniform doses per fraction may be given at planning target volume (PTV) high, intermediate, and low with 240 centigray (cGy) (simultaneously integrated boost), 180-200 cGy (traditional dose per fraction), and 160 cGy (inferior to the traditional dose) in the same fractions, respectively; B: A simple cartoon figure representing a modified SIB dose prescription. Traditionally, radiotherapy is prescribed as a uniform dose per fraction (e.g., 200 cGy) on PTV, which provides a homogeneous dose to cover clinical target volume (CTV) and gross target volume (GTV). Recently, to maximally enhance the possibility of tumor control, a modified SIB technique is applied using a planned non-homologous dose distribution, i.e., escalating a simultaneous intra-tumor boost (e.g., 220-240 cGy per fraction) in addition to a traditional covering dose to the PTV (e.g., 200 cGy per fraction) in the same treatment fractions; C: A simple cartoon figure representing combined modified SIB and simultaneously integrated protection (SIP) dose prescription. To reduce treatment toxicities to adjacent critical organs/tissue, SIP was developed in conjunction with modified SIB. SIP prescribes an inferior-to-traditional dose per fraction, e.g., 150 cGy, to the overlapping region between the PTV and the extended critical organ volume (as shown by the long arrow); D: A simple cartoon figure representing simultaneously integrated inner-escalated boost (SIEB) dose prescription. For further enhanced therapeutic gain (i.e., increased tumor control and decreased treatment toxicity) in managing patients with unresectable liver tumors, we applied a secondary modified SIB (also termed SIEB). SIEB further escalates the intra-tumor boost (e.g., 240-260 cGy per fraction) based on a planned generally attenuated peri-tumor dose (e.g., 120-150 cGy per fraction delivered to the PTV), administered simultaneously. The intra-tumor SIEB boost volume is delineated as a uniform-inner-shrinkage area from the GTV with a margin of 1-10 mm (i.e., a geometrically central zone), depending on the tumor size, the intensity of the dose escalation, the level of liver preservation, the closeness of the gastrointestinal organs to the irradiation targets, and the patient's condition. Note that an additional most-inner SIEB boost volume with the highest dose per fraction (e.g., 260-300 cGy or higher) might be considered for highly selected patients with very bulky tumors. E: A clinical case with SIEB dose prescription. The blue, purple, and red lines show the PTV, CTV, and GTV of the irradiating target, respectively. In this case, based on the physician’s choice and the patient’s condition, a dose per fraction of 120 cGy was prescribed to the PTV, 150 cGy to the CTV, and 200 cGy to the GTV. Finally, in the yellow-outlined region, 280 cGy per fraction was simultaneously delivered to the SIEB boost volume. A total of 30 fractions were given, generating total doses levels of 3600 cGy, 4500 cGy, 6000 cGy, and 7400 cGy to the PTC, CTV, GTV, and SIEB boost volume, respectively. Note that the most peripheral dose per fraction of 120 cGy was chosen mainly due to a very close distance between the PTV and an adjacent critical organ, i.e., the duodenum. This short-distance closeness could easily lead irradiation to harm the duodenum under the context of daily organ motion. SIB: Simultaneously integrated boost; SIP: Simultaneously integrated protection; SIEB: Simultaneously integrated inner-escalated boost; PTV: Planning target volume; CTV: Clinical target volume; GTV: Gross target volume.



the role of photon- and proton-based precise RT in managing patients with HCC is shifting continuously from palliative to curative intent.

ACKNOWLEDGEMENTSWe gratefully acknowledge all RT staff members of the Buddhist Dalin Tzu Chi Hospital and Tzu Chi Medication Foundation. We also thank the contribution of all RT members, including Buddhist Hualien Tzu Chi Hospital, Far Eastern Memorial Hospital and China Medical University Hospital who involved in inter-hospital HCC research program TASABR trial registered prospectively on ClinicalTrials.gov (Trial Identifier: NCT02921139).

REFERENCESGlobal Burden of Disease Cancer Collaboration, Fitzmaurice C, Abate D, Abbasi N, Abbastabar H, Abd-Allah F, Abdel-Rahman O, Abdelalim A, Abdoli A, Abdollahpour I, Abdulle ASM, Abebe ND, Abraha HN, Abu-Raddad LJ, Abualhasan A, Adedeji IA, Advani SM, Afarideh M, Afshari M, Aghaali M, Agius D, Agrawal S, Ahmadi A, Ahmadian E, Ahmadpour E, Ahmed MB, Akbari ME, Akinyemiju T, Al-Aly Z, AlAbdulKader AM, Alahdab F, Alam T, Alamene GM, Alemnew BTT, Alene KA, Alinia C, Alipour V, Aljunid SM, Bakeshei FA, Almadi MAH, Almasi-Hashiani A, Alsharif U, Alsowaidi S, Alvis-Guzman N, Amini E, Amini S, Amoako YA, Anbari Z, Anber NH, Andrei CL, Anjomshoa M, Ansari F, Ansariadi A, Appiah SCY, Arab-Zozani M, Arabloo J, Arefi Z, Aremu O, Areri HA, Artaman A, Asayesh H, Asfaw ET, Ashagre AF, Assadi R, Ataeinia B, Atalay HT, Ataro Z, Atique S, Ausloos M, Avila-Burgos L, Avokpaho EFGA, Awasthi A, Awoke N, Ayala Quintanilla BP, Ayanore MA, Ayele HT, Babaee E, Bacha U, Badawi A, Bagherzadeh M, Bagli E, Balakrishnan S, Balouchi A, Bärnighausen TW, Battista RJ, Behzadifar M, Bekele BB, Belay YB, Belayneh YM, Berfield KKS, Berhane A, Bernabe E, Beuran M, Bhakta N, Bhattacharyya K, Biadgo B, Bijani A, Bin Sayeed MS, Birungi C, Bisignano C, Bitew H, Bjørge T, Bleyer A, Bogale KA, Bojia HA, Borzì AM, Bosetti C, Bou-Orm IR, Brenner H, Brewer JD, Briko AN, Briko NI, Bustamante-Teixeira MT, Butt ZA, Carreras G, Carrero JJ, Carvalho F, Castro C, Castro F, Catalá-López F, Cerin E, Chaiah Y, Chanie WF, Chattu VK, Chaturvedi P, Chauhan NS, Chehrazi M, Chiang PP, Chichiabellu TY, Chido-Amajuoyi OG, Chimed-Ochir O, Choi JJ, Christopher DJ, Chu DT, Constantin MM, Costa VM, Crocetti E, Crowe CS, Curado MP, Dahlawi SMA, Damiani G, Darwish AH, Daryani A, das Neves J, Demeke FM, Demis AB, Demissie BW, Demoz GT, Denova-Gutiérrez E, Derakhshani A, Deribe KS, Desai R, Desalegn BB, Desta M, Dey S, Dharmaratne SD, Dhimal M, Diaz D, Dinberu MTT, Djalalinia S, Doku DT, Drake TM, Dubey M, Dubljanin E, Duken EE, Ebrahimi H, Effiong A, Eftekhari A, El Sayed I, Zaki MES, El-Jaafary SI, El-Khatib Z, Elemineh DA, Elkout H, Ellenbogen RG, Elsharkawy A, Emamian MH, Endalew DA, Endries AY, Eshrati B, Fadhil I, Fallah Omrani V, Faramarzi M, Farhangi MA, Farioli A, Farzadfar F, Fentahun N, Fernandes E, Feyissa GT, Filip I, Fischer F, Fisher JL, Force LM, Foroutan M, Freitas M, Fukumoto T, Futran ND, Gallus S, Gankpe FG, Gayesa RT, Gebrehiwot TT, Gebremeskel GG, Gedefaw GA, Gelaw BK, Geta B, Getachew S, Gezae KE, Ghafourifard M, Ghajar A, Ghashghaee A, Gholamian A, Gill PS, Ginindza TTG, Girmay A, Gizaw M, Gomez RS, Gopalani SV, Gorini G, Goulart BNG, Grada A, Ribeiro Guerra M, Guimaraes ALS, Gupta PC, Gupta R, Hadkhale K, Haj-Mirzaian A, Hamadeh RR, Hamidi S, Hanfore LK, Haro JM, Hasankhani M, Hasanzadeh A, Hassen HY, Hay RJ, Hay SI, Henok A, Henry NJ, Herteliu C, Hidru HD, Hoang CL, Hole MK, Hoogar P, Horita N, Hosgood HD, Hosseini M, Hosseinzadeh M, Hostiuc M, Hostiuc S, Househ M, Hussen MM, Ileanu B, Ilic MD, Innos K, Irvani SSN, Iseh KR, Islam SMS, Islami F, Jafari Balalami N, Jafarinia M, Jahangiry L, Jahani MA, Jahanmehr N, Jakovljevic M, James SL, Javanbakht M, Jayaraman S, Jee SH, Jenabi E, Jha RP, Jonas JB, Jonnagaddala J, Joo T, Jungari SB, Jürisson M, Kabir A, Kamangar F, Karch A, Karimi N, Karimian A, Kasaeian A, Kasahun GG, Kassa B, Kassa TD, Kassaw MW, Kaul A, Keiyoro PN, Kelbore AG, Kerbo AA, Khader YS, Khalilarjmandi M, Khan EA, Khan G, Khang YH, Khatab K, Khater A, Khayamzadeh M, Khazaee-Pool M, Khazaei S, Khoja AT, Khosravi MH, Khubchandani J, Kianipour N, Kim D, Kim YJ, Kisa A, Kisa S, Kissimova-Skarbek K, Komaki H, Koyanagi A, Krohn KJ, Bicer BK, Kugbey N, Kumar V, Kuupiel D, La Vecchia C, Lad DP, Lake EA, Lakew AM, Lal DK, Lami FH, Lan Q, Lasrado S, Lauriola P, Lazarus JV, Leigh J, Leshargie CT, Liao Y, Limenih MA, Listl S, Lopez AD, Lopukhov PD, Lunevicius R, Madadin M, Magdeldin S, El Razek HMA, Majeed A, Maleki A, Malekzadeh R, Manafi A, Manafi N, Manamo WA, Mansourian M, Mansournia MA, Mantovani LG, Maroufizadeh S, Martini SMS, Mashamba-Thompson TP, Massenburg BB, Maswabi MT, Mathur MR, McAlinden C, McKee M, Meheretu HAA, Mehrotra R, Mehta V, Meier T, Melaku YA, Meles GG, Meles HG, Melese A, Melku M, Memiah PTN, Mendoza W, Menezes RG, Merat S, Meretoja TJ, Mestrovic T, Miazgowski B, Miazgowski T, Mihretie KMM, Miller TR, Mills EJ, Mir SM, Mirzaei H, Mirzaei HR, Mishra R, Moazen B, Mohammad DK, Mohammad KA, Mohammad Y, Darwesh AM, Mohammadbeigi A, Mohammadi H, Mohammadi M, Mohammadian M, Mohammadian-Hafshejani

1



A, Mohammadoo-Khorasani M, Mohammadpourhodki R, Mohammed AS, Mohammed JA, Mohammed S, Mohebi F, Mokdad AH, Monasta L, Moodley Y, Moosazadeh M, Moossavi M, Moradi G, Moradi-Joo M, Moradi-Lakeh M, Moradpour F, Morawska L, Morgado-da-Costa J, Morisaki N, Morrison SD, Mosapour A, Mousavi SM, Muche AA, Muhammed OSS, Musa J, Nabhan AF, Naderi M, Nagarajan AJ, Nagel G, Nahvijou A, Naik G, Najafi F, Naldi L, Nam HS, Nasiri N, Nazari J, Negoi I, Neupane S, Newcomb PA, Nggada HA, Ngunjiri JW, Nguyen CT, Nikniaz L, Ningrum DNA, Nirayo YL, Nixon MR, Nnaji CA, Nojomi M, Nosratnejad S, Shiadeh MN, Obsa MS, Ofori-Asenso R, Ogbo FA, Oh IH, Olagunju AT, Olagunju TO, Oluwasanu MM, Omonisi AE, Onwujekwe OE, Oommen AM, Oren E, Ortega-Altamirano DDV, Ota E, Otstavnov SS, Owolabi MO, P A M, Padubidri JR, Pakhale S, Pakpour AH, Pana A, Park EK, Parsian H, Pashaei T, Patel S, Patil ST, Pennini A, Pereira DM, Piccinelli C, Pillay JD, Pirestani M, Pishgar F, Postma MJ, Pourjafar H, Pourmalek F, Pourshams A, Prakash S, Prasad N, Qorbani M, Rabiee M, Rabiee N, Radfar A, Rafiei A, Rahim F, Rahimi M, Rahman MA, Rajati F, Rana SM, Raoofi S, Rath GK, Rawaf DL, Rawaf S, Reiner RC, Renzaho AMN, Rezaei N, Rezapour A, Ribeiro AI, Ribeiro D, Ronfani L, Roro EM, Roshandel G, Rostami A, Saad RS, Sabbagh P, Sabour S, Saddik B, Safiri S, Sahebkar A, Salahshoor MR, Salehi F, Salem H, Salem MR, Salimzadeh H, Salomon JA, Samy AM, Sanabria J, Santric Milicevic MM, Sartorius B, Sarveazad A, Sathian B, Satpathy M, Savic M, Sawhney M, Sayyah M, Schneider IJC, Schöttker B, Sekerija M, Sepanlou SG, Sepehrimanesh M, Seyedmousavi S, Shaahmadi F, Shabaninejad H, Shahbaz M, Shaikh MA, Shamshirian A, Shamsizadeh M, Sharafi H, Sharafi Z, Sharif M, Sharifi A, Sharifi H, Sharma R, Sheikh A, Shirkoohi R, Shukla SR, Si S, Siabani S, Silva DAS, Silveira DGA, Singh A, Singh JA, Sisay S, Sitas F, Sobngwi E, Soofi M, Soriano JB, Stathopoulou V, Sufiyan MB, Tabarés-Seisdedos R, Tabuchi T, Takahashi K, Tamtaji OR, Tarawneh MR, Tassew SG, Taymoori P, Tehrani-Banihashemi A, Temsah MH, Temsah O, Tesfay BE, Tesfay FH, Teshale MY, Tessema GA, Thapa S, Tlaye KG, Topor-Madry R, Tovani-Palone MR, Traini E, Tran BX, Tran KB, Tsadik AG, Ullah I, Uthman OA, Vacante M, Vaezi M, Varona Pérez P, Veisani Y, Vidale S, Violante FS, Vlassov V, Vollset SE, Vos T, Vosoughi K, Vu GT, Vujcic IS, Wabinga H, Wachamo TM, Wagnew FS, Waheed Y, Weldegebreal F, Weldesamuel GT, Wijeratne T, Wondafrash DZ, Wonde TE, Wondmieneh AB, Workie HM, Yadav R, Yadegar A, Yadollahpour A, Yaseri M, Yazdi-Feyzabadi V, Yeshaneh A, Yimam MA, Yimer EM, Yisma E, Yonemoto N, Younis MZ, Yousefi B, Yousefifard M, Yu C, Zabeh E, Zadnik V, Moghadam TZ, Zaidi Z, Zamani M, Zandian H, Zangeneh A, Zaki L, Zendehdel K, Zenebe ZM, Zewale TA, Ziapour A, Zodpey S, Murray CJL. Global, Regional, and National Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life-Years for 29 Cancer Groups, 1990 to 2017: A Systematic Analysis for the Global Burden of Disease Study. JAMA Oncol 2019; 5: 1749-1768 [PMID: 31560378 DOI: 10.1001/jamaoncol.2019.2996]Chiang CJ, Yang YW, Chen JD, You SL, Yang HI, Lee MH, Lai MS, Chen CJ. Significant reduction in end-stage liver diseases burden through the national viral hepatitis therapy program in Taiwan. Hepatology 2015; 61: 1154-1162 [PMID: 25476749 DOI: 10.1002/hep.27630]

2

Omata M, Cheng AL, Kokudo N, Kudo M, Lee JM, Jia J, Tateishi R, Han KH, Chawla YK, Shiina S, Jafri W, Payawal DA, Ohki T, Ogasawara S, Chen PJ, Lesmana CRA, Lesmana LA, Gani RA, Obi S, Dokmeci AK, Sarin SK. Asia-Pacific clinical practice guidelines on the management of hepatocellular carcinoma: a 2017 update. Hepatol Int 2017; 11: 317-370 [PMID: 28620797 DOI: 10.1007/s12072-017-9799-9]

3

Yuen MF, Hou JL, Chutaputti A; Asia Pacific Working Party on Prevention of Hepatocellular Carcinoma. Hepatocellular carcinoma in the Asia pacific region. J Gastroenterol Hepatol 2009; 24: 346-353 [PMID: 19220670 DOI: 10.1111/j.1440-1746.2009.05784.x]

4

Yun BY, Lee HW, Min IK, Kim SU, Park JY, Kim DY, Ahn SH, Kim BK. Prognosis of Early-Stage Hepatocellular Carcinoma: Comparison between Trans-Arterial Chemoembolization and Radiofrequency Ablation. Cancers (Basel) 2020; 12 [PMID: 32899584 DOI: 10.3390/cancers12092527]

5

Lu LH, Mei J, Kan A, Ling YH, Li SH, Wei W, Chen MS, Zhang YF, Guo RP. Treatment optimization for recurrent hepatocellular carcinoma: Repeat hepatic resection versus radiofrequency ablation. Cancer Med 2020; 9: 2997-3005 [PMID: 32108433 DOI: 10.1002/cam4.2951]

6

Huang Y, Song J, Zheng J, Jiang L, Yan L, Yang J, Zeng Y, Wu H. Comparison of Hepatic Resection Combined with Intraoperative Radiofrequency Ablation, or Hepatic Resection Alone, for Hepatocellular Carcinoma Patients with Multifocal Tumors Meeting the University of California San Francisco (UCSF) Criteria: A Propensity Score-Matched Analysis. Ann Surg Oncol 2020; 27: 2334-2345 [PMID: 32016632 DOI: 10.1245/s10434-020-08231-0]

7

Yu SJ, Yoon JH, Lee JM, Lee JY, Kim SH, Cho YY, Yoo JJ, Lee M, Lee DH, Cho Y, Cho EJ, Lee JH, Kim YJ, Kim CY. Percutaneous ethanol injection therapy is comparable to radiofrequency ablation in hepatocellular carcinoma smaller than 1.5 cm: A matched case-control comparative analysis. Medicine (Baltimore) 2016; 95: e4551 [PMID: 27583865 DOI: 10.1097/MD.0000000000004551]

8

Qi C, Gao H, Zhao Q, Zhang L. Computed Tomography-Guided Percutaneous Cryoablation for Subcardiac Hepatocellular Carcinoma: Safety, Efficacy, Therapeutic Results and Risk Factors for Survival Outcomes. Cancer Manag Res 2020; 12: 3333-3342 [PMID: 32494197 DOI: 10.2147/CMAR.S250652]

9

Qi C, Li S, Zhang L. Development and Validation of a Clinicopathological-Based Nomogram to 10

http://www.ncbi.nlm.nih.gov/pubmed/31560378

https://dx.doi.org/10.1001/jamaoncol.2019.2996


https://dx.doi.org/10.1002/hep.27630


https://dx.doi.org/10.1007/s12072-017-9799-9


https://dx.doi.org/10.1111/j.1440-1746.2009.05784.x


https://dx.doi.org/10.3390/cancers12092527


https://dx.doi.org/10.1002/cam4.2951


https://dx.doi.org/10.1245/s10434-020-08231-0


https://dx.doi.org/10.1097/MD.0000000000004551


https://dx.doi.org/10.2147/CMAR.S250652



Predict the Survival Outcome of Patients with Recurrent Hepatocellular Carcinoma After Hepatectomy Who Underwent Microwave Ablation. Cancer Manag Res 2020; 12: 7589-7600 [PMID: 32904581 DOI: 10.2147/CMAR.S266052]Peisen F, Maurer M, Grosse U, Nikolaou K, Syha R, Ketelsen D, Artzner C, Bitzer M, Horger M, Grözinger G. Predictive performance of the mHAP-II score in a real-life western cohort with hepatocellular carcinoma following trans-arterial chemoembolisation with drug-eluting beads (DEB-TACE). Eur Radiol 2020; 30: 3782-3792 [PMID: 32125515 DOI: 10.1007/s00330-020-06734-8]

11

Nakano MM, Yamamoto A, Nishida N, Hamuro M, Hamamoto S, Jogo A, Sohgawa E, Kageyama K, Minami T, Miki Y. Risk factors for local recurrence of hepatocellular carcinoma after transcatheter arterial chemoembolization with drug-eluting beads (DEB-TACE). Jpn J Radiol 2019; 37: 543-548 [PMID: 31055710 DOI: 10.1007/s11604-019-00840-4]

12

Vesselle G, Quirier-Leleu C, Velasco S, Charier F, Silvain C, Boucebci S, Ingrand P, Tasu JP. Predictive factors for complete response of chemoembolization with drug-eluting beads (DEB-TACE) for hepatocellular carcinoma. Eur Radiol 2016; 26: 1640-1648 [PMID: 26455721 DOI: 10.1007/s00330-015-3982-y]

13

Adcock CS, Puneky LV, Campbell GS. Favorable Response of Metastatic Hepatocellular Carcinoma to Treatment with Trans-arterial Radioembolization Followed by Sorafenib and Nivolumab. Cureus 2019; 11: e4083 [PMID: 31019861 DOI: 10.7759/cureus.4083]

14

Rognoni C, Ciani O, Sommariva S, Bargellini I, Bhoori S, Cioni R, Facciorusso A, Golfieri R, Gramenzi A, Mazzaferro V, Mosconi C, Ponziani F, Sacco R, Trevisani F, Tarricone R. Trans-arterial radioembolization for intermediate-advanced hepatocellular carcinoma: a budget impact analysis. BMC Cancer 2018; 18: 715 [PMID: 29976149 DOI: 10.1186/s12885-018-4636-7]

15

Fornaro L, Vivaldi C, Lorenzoni G, Masi G, Bargellini I. Moving beyond sorafenib alone in advanced hepatocellular carcinoma: is hepatic arterial infusion chemotherapy the best option? Ann Oncol 2017; 28: 667 [PMID: 27993799 DOI: 10.1093/annonc/mdw664]

16

Ikeda M, Shimizu S, Sato T, Morimoto M, Kojima Y, Inaba Y, Hagihara A, Kudo M, Nakamori S, Kaneko S, Sugimoto R, Tahara T, Ohmura T, Yasui K, Sato K, Ishii H, Furuse J, Okusaka T. Sorafenib plus hepatic arterial infusion chemotherapy with cisplatin vs sorafenib for advanced hepatocellular carcinoma: randomized phase II trial. Ann Oncol 2016; 27: 2090-2096 [PMID: 27573564 DOI: 10.1093/annonc/mdw323]

17

Ishikawa T, Kubota T, Abe S, Watanabe Y, Sugano T, Inoue R, Iwanaga A, Seki K, Honma T, Yoshida T. Hepatic arterial infusion chemotherapy with cisplatin before radical local treatment of early hepatocellular carcinoma (JIS score 0/1) improves survival. Ann Oncol 2014; 25: 1379-1384 [PMID: 24737779 DOI: 10.1093/annonc/mdu155]

18

Labeur TA, Hofsink Q, Takkenberg RB, van Delden OM, Mathôt RAA, Schinner R, Malfertheiner P, Amthauer H, Schütte K, Basu B, Kuhl C, Mayerle J, Ricke J, Klümpen HJ. The value of sorafenib trough levels in patients with advanced hepatocellular carcinoma - a substudy of the SORAMIC trial. Acta Oncol 2020; 59: 1028-1035 [PMID: 32366155 DOI: 10.1080/0284186X.2020.1759826]

19

Briggs A, Daniele B, Dick K, Evans TRJ, Galle PR, Hubner RA, Lopez C, Siebert U, Tremblay G. Covariate-adjusted analysis of the Phase 3 REFLECT study of lenvatinib versus sorafenib in the treatment of unresectable hepatocellular carcinoma. Br J Cancer 2020; 122: 1754-1759 [PMID: 32265508 DOI: 10.1038/s41416-020-0817-7]

20

Fung AS, Tam VC, Meyers DE, Sim HW, Knox JJ, Zaborska V, Davies J, Ko YJ, Batuyong E, Samawi H, Cheung WY, Lee-Ying R. Second-line treatment of hepatocellular carcinoma after sorafenib: Characterizing treatments used over the past 10 years and real-world eligibility for cabozantinib, regorafenib, and ramucirumab. Cancer Med 2020; 9: 4640-4647 [PMID: 32378799 DOI: 10.1002/cam4.3116]

21

Choi WM, Choi J, Lee D, Shim JH, Lim YS, Lee HC, Chung YH, Lee YS, Park SR, Ryu MH, Ryoo BY, Lee SJ, Kim KM. Regorafenib Versus Nivolumab After Sorafenib Failure: Real-World Data in Patients With Hepatocellular Carcinoma. Hepatol Commun 2020; 4: 1073-1086 [PMID: 32626838 DOI: 10.1002/hep4.1523]

22

Duffy JP, Vardanian A, Benjamin E, Watson M, Farmer DG, Ghobrial RM, Lipshutz G, Yersiz H, Lu DS, Lassman C, Tong MJ, Hiatt JR, Busuttil RW. Liver transplantation criteria for hepatocellular carcinoma should be expanded: a 22-year experience with 467 patients at UCLA. Ann Surg 2007; 246: 502-509; discussion 509 [PMID: 17717454 DOI: 10.1097/SLA.0b013e318148c704]

23

Omata M, Lesmana LA, Tateishi R, Chen PJ, Lin SM, Yoshida H, Kudo M, Lee JM, Choi BI, Poon RT, Shiina S, Cheng AL, Jia JD, Obi S, Han KH, Jafri W, Chow P, Lim SG, Chawla YK, Budihusodo U, Gani RA, Lesmana CR, Putranto TA, Liaw YF, Sarin SK. Asian Pacific Association for the Study of the Liver consensus recommendations on hepatocellular carcinoma. Hepatol Int 2010; 4: 439-474 [PMID: 20827404 DOI: 10.1007/s12072-010-9165-7]

24

Shiina S, Tateishi R, Arano T, Uchino K, Enooku K, Nakagawa H, Asaoka Y, Sato T, Masuzaki R, Kondo Y, Goto T, Yoshida H, Omata M, Koike K. Radiofrequency ablation for hepatocellular carcinoma: 10-year outcome and prognostic factors. Am J Gastroenterol 2012; 107: 569-77; quiz 578 [PMID: 22158026 DOI: 10.1038/ajg.2011.425]

25

Wahl DR, Stenmark MH, Tao Y, Pollom EL, Caoili EM, Lawrence TS, Schipper MJ, Feng M. Outcomes After Stereotactic Body Radiotherapy or Radiofrequency Ablation for Hepatocellular Carcinoma. J Clin Oncol 2016; 34: 452-459 [PMID: 26628466 DOI: 10.1200/JCO.2015.61.4925]

26

Delis SG, Dervenis C. Selection criteria for liver resection in patients with hepatocellular carcinoma and chronic liver disease. World J Gastroenterol 2008; 14: 3452-3460 [PMID: 18567070 DOI:

27


https://dx.doi.org/10.2147/CMAR.S266052


https://dx.doi.org/10.1007/s00330-020-06734-8


https://dx.doi.org/10.1007/s11604-019-00840-4


https://dx.doi.org/10.1007/s00330-015-3982-y


https://dx.doi.org/10.7759/cureus.4083


https://dx.doi.org/10.1186/s12885-018-4636-7


https://dx.doi.org/10.1093/annonc/mdw664


https://dx.doi.org/10.1093/annonc/mdw323


https://dx.doi.org/10.1093/annonc/mdu155


https://dx.doi.org/10.1080/0284186X.2020.1759826


https://dx.doi.org/10.1038/s41416-020-0817-7




https://dx.doi.org/10.1002/hep4.1523


https://dx.doi.org/10.1097/SLA.0b013e318148c704


https://dx.doi.org/10.1007/s12072-010-9165-7


https://dx.doi.org/10.1038/ajg.2011.425


https://dx.doi.org/10.1200/JCO.2015.61.4925




10.3748/wjg.14.3452]Bruix J, Sherman M, Llovet JM, Beaugrand M, Lencioni R, Burroughs AK, Christensen E, Pagliaro L, Colombo M, Rodés J; EASL Panel of Experts on HCC. Clinical management of hepatocellular carcinoma. Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver. J Hepatol 2001; 35: 421-430 [PMID: 11592607 DOI: 10.1016/s0168-8278(01)00130-1]

28

McMasters KM, Vauthey JN. Hepatocellular Carcinoma: Targeted Therapy and Multidisciplinary Care. New York: Springer, 2011

29

Chan AC, Chan SC, Chok KS, Cheung TT, Chiu DW, Poon RT, Fan ST, Lo CM. Treatment strategy for recurrent hepatocellular carcinoma: salvage transplantation, repeated resection, or radiofrequency ablation? Liver Transpl 2013; 19: 411-419 [PMID: 23447460 DOI: 10.1002/lt.23605]

30

Veltri A, Moretto P, Doriguzzi A, Pagano E, Carrara G, Gandini G. Radiofrequency thermal ablation (RFA) after transarterial chemoembolization (TACE) as a combined therapy for unresectable non-early hepatocellular carcinoma (HCC). Eur Radiol 2006; 16: 661-669 [PMID: 16228211 DOI: 10.1007/s00330-005-0029-9]

31

Guo W, He X, Li Z, Li Y. Combination of Transarterial Chemoembolization (TACE) and Radiofrequency Ablation (RFA) vs Surgical Resection (SR) on Survival Outcome of Early Hepatocellular Carcinoma: A Meta-Analysis. Hepatogastroenterology 2015; 62: 710-714 [PMID: 26897959]

32

Doi H, Beppu N, Kitajima K, Kuribayashi K. Stereotactic Body Radiation Therapy for Liver Tumors: Current Status and Perspectives. Anticancer Res 2018; 38: 591-599 [PMID: 29374681 DOI: 10.21873/anticanres.12263]

33

Benson AB, D'Angelica MI, Abbott DE, Abrams TA, Alberts SR, Anaya DA, Anders R, Are C, Brown D, Chang DT, Cloyd J, Covey AM, Hawkins W, Iyer R, Jacob R, Karachristos A, Kelley RK, Kim R, Palta M, Park JO, Sahai V, Schefter T, Sicklick JK, Singh G, Sohal D, Stein S, Tian GG, Vauthey JN, Venook AP, Hammond LJ, Darlow SD. Guidelines Insights: Hepatobiliary Cancers, Version 2.2019. J Natl Compr Canc Netw 2019; 17: 302-310 [PMID: 30959462 DOI: 10.6004/jnccn.2019.0019]

34

Marrero JA, Kulik LM, Sirlin CB, Zhu AX, Finn RS, Abecassis MM, Roberts LR, Heimbach JK. Diagnosis, Staging, and Management of Hepatocellular Carcinoma: 2018 Practice Guidance by the American Association for the Study of Liver Diseases. Hepatology 2018; 68: 723-750 [PMID: 29624699 DOI: 10.1002/hep.29913]

35

European Association for the Study of the Liver. EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma. J Hepatol 2018; 69: 182-236 [PMID: 29628281 DOI: 10.1016/j.jhep.2018.03.019]

36

Kokudo N, Takemura N, Hasegawa K, Takayama T, Kubo S, Shimada M, Nagano H, Hatano E, Izumi N, Kaneko S, Kudo M, Iijima H, Genda T, Tateishi R, Torimura T, Igaki H, Kobayashi S, Sakurai H, Murakami T, Watadani T, Matsuyama Y. Clinical practice guidelines for hepatocellular carcinoma: The Japan Society of Hepatology 2017 (4th JSH-HCC guidelines) 2019 update. Hepatol Res 2019; 49: 1109-1113 [PMID: 31336394 DOI: 10.1111/hepr.13411]

37

Korean Liver Cancer Association; National Cancer Center. 2018 Korean Liver Cancer Association-National Cancer Center Korea Practice Guidelines for the Management of Hepatocellular Carcinoma. Gut Liver 2019; 13: 227-299 [PMID: 31060120 DOI: 10.5009/gnl19024]

38

Zhou J, Sun HC, Wang Z, Cong WM, Wang JH, Zeng MS, Yang JM, Bie P, Liu LX, Wen TF, Han GH, Wang MQ, Liu RB, Lu LG, Ren ZG, Chen MS, Zeng ZC, Liang P, Liang CH, Chen M, Yan FH, Wang WP, Ji Y, Cheng WW, Dai CL, Jia WD, Li YM, Li YX, Liang J, Liu TS, Lv GY, Mao YL, Ren WX, Shi HC, Wang WT, Wang XY, Xing BC, Xu JM, Yang JY, Yang YF, Ye SL, Yin ZY, Zhang BH, Zhang SJ, Zhou WP, Zhu JY, Liu R, Shi YH, Xiao YS, Dai Z, Teng GJ, Cai JQ, Wang WL, Dong JH, Li Q, Shen F, Qin SK, Fan J. Guidelines for Diagnosis and Treatment of Primary Liver Cancer in China (2017 Edition). Liver Cancer 2018; 7: 235-260 [PMID: 30319983 DOI: 10.1159/000488035]

39

Surveillance group. ; Diagnosis group; Staging group; Surgery group; Local ablation group; TACE/TARE/HAI group; Target therapy/systemic therapy group; Radiotherapy group; Prevention group; Drafting group. Management consensus guideline for hepatocellular carcinoma: 2016 updated by the Taiwan Liver Cancer Association and the Gastroenterological Society of Taiwan. J Formos Med Assoc 2018; 117: 381-403 [PMID: 29074347 DOI: 10.1016/j.jfma.2017.09.007]

40

Ueno S, Tanabe G, Nuruki K, Hamanoue M, Komorizono Y, Oketani M, Hokotate H, Inoue H, Baba Y, Imamura Y, Aikou T. Prognostic performance of the new classification of primary liver cancer of Japan (4th edition) for patients with hepatocellular carcinoma: a validation analysis. Hepatol Res 2002; 24: 395-403 [PMID: 12479938 DOI: 10.1016/s1386-6346(02)00144-4]

41

Kudo M, Kitano M, Sakurai T, Nishida N. General Rules for the Clinical and Pathological Study of Primary Liver Cancer, Nationwide Follow-Up Survey and Clinical Practice Guidelines: The Outstanding Achievements of the Liver Cancer Study Group of Japan. Dig Dis 2015; 33: 765-770 [PMID: 26488173 DOI: 10.1159/000439101]

42

Künzli BM, Abitabile P, Maurer CA. Radiofrequency ablation of liver tumors: Actual limitations and potential solutions in the future. World J Hepatol 2011; 3: 8-14 [PMID: 21307982 DOI: 10.4254/wjh.v3.i1.8]

43

https://dx.doi.org/10.3748/wjg.14.3452


https://dx.doi.org/10.1016/s0168-8278(01)00130-1


https://dx.doi.org/10.1002/lt.23605


https://dx.doi.org/10.1007/s00330-005-0029-9



https://dx.doi.org/10.21873/anticanres.12263


https://dx.doi.org/10.6004/jnccn.2019.0019




https://dx.doi.org/10.1016/j.jhep.2018.03.019


https://dx.doi.org/10.1111/hepr.13411


https://dx.doi.org/10.5009/gnl19024


https://dx.doi.org/10.1159/000488035


https://dx.doi.org/10.1016/j.jfma.2017.09.007


https://dx.doi.org/10.1016/s1386-6346(02)00144-4


https://dx.doi.org/10.1159/000439101


https://dx.doi.org/10.4254/wjh.v3.i1.8



Zhang X, Wang K, Wang M, Yang G, Ye X, Wu M, Cheng S. Transarterial chemoembolization (TACE) combined with sorafenib versus TACE for hepatocellular carcinoma with portal vein tumor thrombus: a systematic review and meta-analysis. Oncotarget 2017; 8: 29416-29427 [PMID: 28177886 DOI: 10.18632/oncotarget.15075]

44

Huo YR, Eslick GD. Transcatheter Arterial Chemoembolization Plus Radiotherapy Compared With Chemoembolization Alone for Hepatocellular Carcinoma: A Systematic Review and Meta-analysis. JAMA Oncol 2015; 1: 756-765 [PMID: 26182200 DOI: 10.1001/jamaoncol.2015.2189]

45

Kang JK, Kim MS, Cho CK, Yang KM, Yoo HJ, Kim JH, Bae SH, Jung DH, Kim KB, Lee DH, Han CJ, Kim J, Park SC, Kim YH. Stereotactic body radiation therapy for inoperable hepatocellular carcinoma as a local salvage treatment after incomplete transarterial chemoembolization. Cancer 2012; 118: 5424-5431 [PMID: 22570179 DOI: 10.1002/cncr.27533]

46

Li MF, Leung HW, Chan AL, Wang SY. Network meta-analysis of treatment regimens for inoperable advanced hepatocellular carcinoma with portal vein invasion. Ther Clin Risk Manag 2018; 14: 1157-1168 [PMID: 30013351 DOI: 10.2147/TCRM.S162898]

47

Kwon JH, Kim GM, Han K, Won JY, Kim MD, Lee DY, Lee J, Choi W, Kim YS, Kim DY, Han KH. Safety and Efficacy of Transarterial Radioembolization Combined with Chemoembolization for Bilobar Hepatocellular Carcinoma: A Single-Center Retrospective Study. Cardiovasc Intervent Radiol 2018; 41: 459-465 [PMID: 29067511 DOI: 10.1007/s00270-017-1826-7]

48

Qu XD, Chen CS, Wang JH, Yan ZP, Chen JM, Gong GQ, Liu QX, Luo JJ, Liu LX, Liu R, Qian S. The efficacy of TACE combined sorafenib in advanced stages hepatocellullar carcinoma. BMC Cancer 2012; 12: 263 [PMID: 22721173 DOI: 10.1186/1471-2407-12-263]

49

Pan T, Li XS, Xie QK, Wang JP, Li W, Wu PH, Zhao M. Safety and efficacy of transarterial chemoembolization plus sorafenib for hepatocellular carcinoma with portal venous tumour thrombus. Clin Radiol 2014; 69: e553-e561 [PMID: 25304928 DOI: 10.1016/j.crad.2014.09.007]

50

Salman A, Simoneau E, Hassanain M, Chaudhury P, Boucher LM, Valenti D, Cabrera T, Nudo C, Metrakos P. Combined sorafenib and yttrium-90 radioembolization for the treatment of advanced hepatocellular carcinoma. Curr Oncol 2016; 23: e472-e480 [PMID: 27803608 DOI: 10.3747/co.23.2827]

51

Ricke J, Bulla K, Kolligs F, Peck-Radosavljevic M, Reimer P, Sangro B, Schott E, Schütte K, Verslype C, Walecki J, Malfertheiner P; SORAMIC study group. Safety and toxicity of radioembolization plus Sorafenib in advanced hepatocellular carcinoma: analysis of the European multicentre trial SORAMIC. Liver Int 2015; 35: 620-626 [PMID: 24930619 DOI: 10.1111/liv.12622]

52

Lorenzin D, Pravisani R, Leo CA, Bugiantella W, Soardo G, Carnelutti A, Umberto B, Risaliti A. Complete Remission of Unresectable Hepatocellular Carcinoma After Combined Sorafenib and Adjuvant Yttrium-90 Radioembolization. Cancer Biother Radiopharm 2016; 31: 65-69 [PMID: 26986817 DOI: 10.1089/cbr.2015.1905]

53

Floudas CS, Brar G, Greten TF. Immunotherapy: Current Status and Future Perspectives. Dig Dis Sci 2019; 64: 1030-1040 [PMID: 30830521 DOI: 10.1007/s10620-019-05516-7]

54

Okusaka T, Ikeda M. Immunotherapy for hepatocellular carcinoma: current status and future perspectives. ESMO Open 2018; 3: e000455 [PMID: 30622744 DOI: 10.1136/esmoopen-2018-000455]

55

Ikeda M, Morizane C, Ueno M, Okusaka T, Ishii H, Furuse J. Chemotherapy for hepatocellular carcinoma: current status and future perspectives. Jpn J Clin Oncol 2018; 48: 103-114 [PMID: 29253194 DOI: 10.1093/jjco/hyx180]

56

Wehrenberg-Klee E, Goyal L, Dugan M, Zhu AX, Ganguli S. Y-90 Radioembolization Combined with a PD-1 Inhibitor for Advanced Hepatocellular Carcinoma. Cardiovasc Intervent Radiol 2018; 41: 1799-1802 [PMID: 29845347 DOI: 10.1007/s00270-018-1993-1]

57

Torok JA, Salama JK. Combining immunotherapy and radiotherapy for the STAR treatment. Nat Rev Clin Oncol 2019; 16: 666-667 [PMID: 31541199 DOI: 10.1038/s41571-019-0277-2]

58

Lin YH, Hung SK, Chiou WY, Lee MS, Shen BJ, Chen LC, Liu DW, Tsai WT, Lin PH, Shih YT, Hsu FC, Tsai SJ, Chan MWY, Lin HY. Significant symptoms alleviation and tumor volume reduction after combined simultaneously integrated inner-escalated boost and volumetric-modulated arc radiotherapy in a patient with unresectable bulky hepatocellular carcinoma: A care-compliant case report. Medicine (Baltimore) 2016; 95: e4717 [PMID: 27559982 DOI: 10.1097/MD.0000000000004717]

59

Frank SJ, Zhu XR. Proton Therapy: Indications, Techniques, and Outcomes. Elsevier Health Sciences, 2020

60

Park HC, Seong J, Han KH, Chon CY, Moon YM, Suh CO. Dose-response relationship in local radiotherapy for hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2002; 54: 150-155 [PMID: 12182985 DOI: 10.1016/s0360-3016(02)02864-x]

61

Kim TH, Park JW, Kim YJ, Kim BH, Woo SM, Moon SH, Kim SS, Koh YH, Lee WJ, Park SJ, Kim JY, Kim DY, Kim CM. Phase I dose-escalation study of proton beam therapy for inoperable hepatocellular carcinoma. Cancer Res Treat 2015; 47: 34-45 [PMID: 25381830 DOI: 10.4143/crt.2013.218]

62

Mizumoto M, Tokuuye K, Sugahara S, Nakayama H, Fukumitsu N, Ohara K, Abei M, Shoda J, Tohno E, Minami M. Proton beam therapy for hepatocellular carcinoma adjacent to the porta hepatis. Int J Radiat Oncol Biol Phys 2008; 71: 462-467 [PMID: 18243571 DOI: 10.1016/j.ijrobp.2007.09.056]

63


https://dx.doi.org/10.18632/oncotarget.15075


https://dx.doi.org/10.1001/jamaoncol.2015.2189


https://dx.doi.org/10.1002/cncr.27533


https://dx.doi.org/10.2147/TCRM.S162898


https://dx.doi.org/10.1007/s00270-017-1826-7


https://dx.doi.org/10.1186/1471-2407-12-263


https://dx.doi.org/10.1016/j.crad.2014.09.007


https://dx.doi.org/10.3747/co.23.2827


https://dx.doi.org/10.1111/liv.12622


https://dx.doi.org/10.1089/cbr.2015.1905


https://dx.doi.org/10.1007/s10620-019-05516-7


https://dx.doi.org/10.1136/esmoopen-2018-000455


https://dx.doi.org/10.1093/jjco/hyx180


https://dx.doi.org/10.1007/s00270-018-1993-1


https://dx.doi.org/10.1038/s41571-019-0277-2




https://dx.doi.org/10.1016/s0360-3016(02)02864-x


https://dx.doi.org/10.4143/crt.2013.218


https://dx.doi.org/10.1016/j.ijrobp.2007.09.056



Fukumitsu N, Sugahara S, Nakayama H, Fukuda K, Mizumoto M, Abei M, Shoda J, Thono E, Tsuboi K, Tokuuye K. A prospective study of hypofractionated proton beam therapy for patients with hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2009; 74: 831-836 [PMID: 19304408 DOI: 10.1016/j.ijrobp.2008.10.073]

64

Nakayama H, Sugahara S, Fukuda K, Abei M, Shoda J, Sakurai H, Tsuboi K, Matsuzaki Y, Tokuuye K. Proton beam therapy for hepatocellular carcinoma located adjacent to the alimentary tract. Int J Radiat Oncol Biol Phys 2011; 80: 992-995 [PMID: 21543162 DOI: 10.1016/j.ijrobp.2010.03.015]

65

Chang JY, Li QQ, Xu QY, Allen PK, Rebueno N, Gomez DR, Balter P, Komaki R, Mehran R, Swisher SG, Roth JA. Stereotactic ablative radiation therapy for centrally located early stage or isolated parenchymal recurrences of non-small cell lung cancer: how to fly in a "no fly zone". Int J Radiat Oncol Biol Phys 2014; 88: 1120-1128 [PMID: 24661665 DOI: 10.1016/j.ijrobp.2014.01.022]

66

Hong TS, Wo JY, Yeap BY, Ben-Josef E, McDonnell EI, Blaszkowsky LS, Kwak EL, Allen JN, Clark JW, Goyal L, Murphy JE, Javle MM, Wolfgang JA, Drapek LC, Arellano RS, Mamon HJ, Mullen JT, Yoon SS, Tanabe KK, Ferrone CR, Ryan DP, DeLaney TF, Crane CH, Zhu AX. Multi-Institutional Phase II Study of High-Dose Hypofractionated Proton Beam Therapy in Patients With Localized, Unresectable Hepatocellular Carcinoma and Intrahepatic Cholangiocarcinoma. J Clin Oncol 2016; 34: 460-468 [PMID: 26668346 DOI: 10.1200/JCO.2015.64.2710]

67

Nakayama H, Sugahara S, Tokita M, Fukuda K, Mizumoto M, Abei M, Shoda J, Sakurai H, Tsuboi K, Tokuuye K. Proton beam therapy for hepatocellular carcinoma: the University of Tsukuba experience. Cancer 2009; 115: 5499-5506 [PMID: 19645024 DOI: 10.1002/cncr.24619]

68

Kawashima M, Furuse J, Nishio T, Konishi M, Ishii H, Kinoshita T, Nagase M, Nihei K, Ogino T. Phase II study of radiotherapy employing proton beam for hepatocellular carcinoma. J Clin Oncol 2005; 23: 1839-1846 [PMID: 15774777 DOI: 10.1200/JCO.2005.00.620]

69

Sanford NN, Pursley J, Noe B, Yeap BY, Goyal L, Clark JW, Allen JN, Blaszkowsky LS, Ryan DP, Ferrone CR, Tanabe KK, Qadan M, Crane CH, Koay EJ, Eyler C, DeLaney TF, Zhu AX, Wo JY, Grassberger C, Hong TS. Protons vs Photons for Unresectable Hepatocellular Carcinoma: Liver Decompensation and Overall Survival. Int J Radiat Oncol Biol Phys 2019; 105: 64-72 [PMID: 30684667 DOI: 10.1016/j.ijrobp.2019.01.076]

70

Hsieh CE, Venkatesulu BP, Lee CH, Hung SP, Wong PF, Aithala SP, Kim BK, Rao A, Tung-Chieh Chang J, Tsang NM, Wang CC, Lee CC, Lin CC, Tseng JH, Chou WC, Wang YC, Krishnan S, Hong JH. Predictors of Radiation-Induced Liver Disease in Eastern and Western Patients With Hepatocellular Carcinoma Undergoing Proton Beam Therapy. Int J Radiat Oncol Biol Phys 2019; 105: 73-86 [PMID: 30797890 DOI: 10.1016/j.ijrobp.2019.02.032]

71

Chuong MD, Kaiser A, Khan F, Parikh P, Ben-Josef E, Crane C, Brunner T, Okumura T, Schreuder N, Bentzen SM, Gutierrez A, Mendez Romero A, Yoon SM, Sharma N, Kim TH, Kishi K, Moeslein F, Hoffe S, Schefter T, Hanish S, Scorsetti M, Apisarnthanarax S. Consensus Report From the Miami Liver Proton Therapy Conference. Front Oncol 2019; 9: 457 [PMID: 31214502 DOI: 10.3389/fonc.2019.00457]

72

Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F, Montalto F, Ammatuna M, Morabito A, Gennari L. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996; 334: 693-699 [PMID: 8594428 DOI: 10.1056/NEJM199603143341104]

73

Rajyaguru DJ, Borgert AJ, Smith AL, Thomes RM, Conway PD, Halfdanarson TR, Truty MJ, Kurup AN, Go RS. Radiofrequency Ablation Versus Stereotactic Body Radiotherapy for Localized Hepatocellular Carcinoma in Nonsurgically Managed Patients: Analysis of the National Cancer Database. J Clin Oncol 2018; 36: 600-608 [PMID: 29328861 DOI: 10.1200/JCO.2017.75.3228]

74

Mazzaferro V, Llovet JM, Miceli R, Bhoori S, Schiavo M, Mariani L, Camerini T, Roayaie S, Schwartz ME, Grazi GL, Adam R, Neuhaus P, Salizzoni M, Bruix J, Forner A, De Carlis L, Cillo U, Burroughs AK, Troisi R, Rossi M, Gerunda GE, Lerut J, Belghiti J, Boin I, Gugenheim J, Rochling F, Van Hoek B, Majno P; Metroticket Investigator Study Group. Predicting survival after liver transplantation in patients with hepatocellular carcinoma beyond the Milan criteria: a retrospective, exploratory analysis. Lancet Oncol 2009; 10: 35-43 [PMID: 19058754 DOI: 10.1016/S1470-2045(08)70284-5]

75

Burkoň P, Slavik M, Kazda T, Pospíšil P, Prochazka T, Vrzal M, Šlampa P. Stereotactic Body Radiotherapy - Current Indications. Klin Onkol 2019; 32: 10-24 [PMID: 30894002 DOI: 10.14735/amko201910]

76

Murray LJ, Dawson LA. Advances in Stereotactic Body Radiation Therapy for Hepatocellular Carcinoma. Semin Radiat Oncol 2017; 27: 247-255 [PMID: 28577832 DOI: 10.1016/j.semradonc.2017.02.002]

77

Ohri N, Dawson LA, Krishnan S, Seong J, Cheng JC, Sarin SK, Kinkhabwala M, Ahmed MM, Vikram B, Coleman CN, Guha C. Radiotherapy for Hepatocellular Carcinoma: New Indications and Directions for Future Study. J Natl Cancer Inst 2016; 108 [PMID: 27377923 DOI: 10.1093/jnci/djw133]

78

Chen CP. Role of Radiotherapy in the Treatment of Hepatocellular Carcinoma. J Clin Transl Hepatol 2019; 7: 183-190 [PMID: 31293919 DOI: 10.14218/JCTH.2018.00060]

79

Culleton S, Jiang H, Haddad CR, Kim J, Brierley J, Brade A, Ringash J, Dawson LA. Outcomes following definitive stereotactic body radiotherapy for patients with Child-Pugh B or C

80


















https://dx.doi.org/10.3389/fonc.2019.00457


https://dx.doi.org/10.1056/NEJM199603143341104




https://dx.doi.org/10.1016/S1470-2045(08)70284-5


https://dx.doi.org/10.14735/amko201910


https://dx.doi.org/10.1016/j.semradonc.2017.02.002


https://dx.doi.org/10.1093/jnci/djw133


https://dx.doi.org/10.14218/JCTH.2018.00060



hepatocellular carcinoma. Radiother Oncol 2014; 111: 412-417 [PMID: 24906626 DOI: 10.1016/j.radonc.2014.05.002]Gresswell S, Tobillo R, Hasan S, Uemura T, Machado L, Thai N, Kirichenko A. Stereotactic body radiotherapy used as a bridge to liver transplant in patients with hepatocellular carcinoma and Child-Pugh score ≥8 cirrhosis. J Radiosurg SBRT 2018; 5: 261-267 [PMID: 30538886]

81

National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: Hepatobiliary Cancers. [cited 7 July 2020]. Available from: http://www.nccn.org/professionals/physician_gls

82

Ronot M, Bouattour M, Wassermann J, Bruno O, Dreyer C, Larroque B, Castera L, Vilgrain V, Belghiti J, Raymond E, Faivre S. Alternative Response Criteria (Choi, European association for the study of the liver, and modified Response Evaluation Criteria in Solid Tumors [RECIST]) Versus RECIST 1.1 in patients with advanced hepatocellular carcinoma treated with sorafenib. Oncologist 2014; 19: 394-402 [PMID: 24652387 DOI: 10.1634/theoncologist.2013-0114]

83

Liu M, Lin MX, Lu MD, Xu ZF, Zheng KG, Wang W, Kuang M, Zhuang WQ, Xie XY. Comparison of contrast-enhanced ultrasound and contrast-enhanced computed tomography in evaluating the treatment response to transcatheter arterial chemoembolization of hepatocellular carcinoma using modified RECIST. Eur Radiol 2015; 25: 2502-2511 [PMID: 25702094 DOI: 10.1007/s00330-015-3611-9]

84

Edeline J, Boucher E, Rolland Y, Vauléon E, Pracht M, Perrin C, Le Roux C, Raoul JL. Comparison of tumor response by Response Evaluation Criteria in Solid Tumors (RECIST) and modified RECIST in patients treated with sorafenib for hepatocellular carcinoma. Cancer 2012; 118: 147-156 [PMID: 21713764 DOI: 10.1002/cncr.26255]

85

Teraoka Y, Kimura T, Aikata H, Daijo K, Osawa M, Honda F, Nakamura Y, Morio K, Morio R, Hatooka M, Kobayashi T, Nakahara T, Murakami E, Nagaoki Y, Kawaoka T, Tsuge M, Hiramatsu A, Imamura M, Kawakami Y, Nagata Y, Chayama K. Clinical outcomes of stereotactic body radiotherapy for elderly patients with hepatocellular carcinoma. Hepatol Res 2018; 48: 193-204 [PMID: 28544062 DOI: 10.1111/hepr.12916]

86

Chen MS, Li JQ, Zheng Y, Guo RP, Liang HH, Zhang YQ, Lin XJ, Lau WY. A prospective randomized trial comparing percutaneous local ablative therapy and partial hepatectomy for small hepatocellular carcinoma. Ann Surg 2006; 243: 321-328 [PMID: 16495695 DOI: 10.1097/01.sla.0000201480.65519.b8]

87

Hara K, Takeda A, Tsurugai Y, Saigusa Y, Sanuki N, Eriguchi T, Maeda S, Tanaka K, Numata K. Radiotherapy for Hepatocellular Carcinoma Results in Comparable Survival to Radiofrequency Ablation: A Propensity Score Analysis. Hepatology 2019; 69: 2533-2545 [PMID: 30805950 DOI: 10.1002/hep.30591]

88

Kim N, Kim HJ, Won JY, Kim DY, Han KH, Jung I, Seong J. Retrospective analysis of stereotactic body radiation therapy efficacy over radiofrequency ablation for hepatocellular carcinoma. Radiother Oncol 2019; 131: 81-87 [PMID: 30773192 DOI: 10.1016/j.radonc.2018.12.013]

89

Elshaarawy O, Gomaa A, Omar H, Rewisha E, Waked I. Intermediate stage hepatocellular carcinoma: a summary review. J Hepatocell Carcinoma 2019; 6: 105-117 [PMID: 31372364 DOI: 10.2147/JHC.S168682]

90

Sapir E, Tao Y, Schipper MJ, Bazzi L, Novelli PM, Devlin P, Owen D, Cuneo KC, Lawrence TS, Parikh ND, Feng M. Stereotactic Body Radiation Therapy as an Alternative to Transarterial Chemoembolization for Hepatocellular Carcinoma. Int J Radiat Oncol Biol Phys 2018; 100: 122-130 [PMID: 29066120 DOI: 10.1016/j.ijrobp.2017.09.001]

91

Lu L, Zeng J, Wen Z, Tang C, Xu N. Transcatheter arterial chemoembolisation followed by three-dimensional conformal radiotherapy versus transcatheter arterial chemoembolisation alone for primary hepatocellular carcinoma in adults. Cochrane Database Syst Rev 2019; 2: CD012244 [PMID: 30776082 DOI: 10.1002/14651858.CD012244.pub2]

92

Xie H, Yu H, Tian S, Yang X, Wang X, Yang Z, Wang H, Guo Z. What is the best combination treatment with transarterial chemoembolization of unresectable hepatocellular carcinoma? Oncotarget 2017; 8: 100508-100523 [PMID: 29245997 DOI: 10.18632/oncotarget.20119]

93

Paik EK, Kim MS, Jang WI, Seo YS, Cho CK, Yoo HJ, Han CJ, Park SC, Kim SB, Kim YH. Benefits of stereotactic ablative radiotherapy combined with incomplete transcatheter arterial chemoembolization in hepatocellular carcinoma. Radiat Oncol 2016; 11: 22 [PMID: 26896371 DOI: 10.1186/s13014-016-0597-7]

94

Jacob R, Turley F, Redden DT, Saddekni S, Aal AK, Keene K, Yang E, Zarzour J, Bolus D, Smith JK, Gray S, White J, Eckhoff DE, DuBay DA. Adjuvant stereotactic body radiotherapy following transarterial chemoembolization in patients with non-resectable hepatocellular carcinoma tumours of ≥ 3 cm. HPB (Oxford) 2015; 17: 140-149 [PMID: 25186290 DOI: 10.1111/hpb.12331]

95

Lo CH, Huang WY, Lee MS, Lin KT, Lin TP, Chang PY, Fan CY, Jen YM. Stereotactic ablative radiotherapy for unresectable hepatocellular carcinoma patients who failed or were unsuitable for transarterial chemoembolization. Eur J Gastroenterol Hepatol 2014; 26: 345-352 [PMID: 24384685 DOI: 10.1097/MEG.0000000000000032]

96

Hardy-Abeloos C, Lazarev S, Ru M, Kim E, Fischman A, Moshier E, Rosenzweig K, Buckstein M. Safety and Efficacy of Liver Stereotactic Body Radiation Therapy for Hepatocellular Carcinoma After Segmental Transarterial Radioembolization. Int J Radiat Oncol Biol Phys 2019; 105: 968-976 [PMID: 31536781 DOI: 10.1016/j.ijrobp.2019.09.006]

97

Sapisochin G, Barry A, Doherty M, Fischer S, Goldaracena N, Rosales R, Russo M, Beecroft R, 98


https://dx.doi.org/10.1016/j.radonc.2014.05.002


http://www.nccn.org/professionals/physician_gls


https://dx.doi.org/10.1634/theoncologist.2013-0114


https://dx.doi.org/10.1007/s00330-015-3611-9






https://dx.doi.org/10.1097/01.sla.0000201480.65519.b8






https://dx.doi.org/10.2147/JHC.S168682




https://dx.doi.org/10.1002/14651858.CD012244.pub2




https://dx.doi.org/10.1186/s13014-016-0597-7


https://dx.doi.org/10.1111/hpb.12331


https://dx.doi.org/10.1097/MEG.0000000000000032





Ghanekar A, Bhat M, Brierley J, Greig PD, Knox JJ, Dawson LA, Grant DR. Stereotactic body radiotherapy vs TACE or RFA as a bridge to transplant in patients with hepatocellular carcinoma. An intention-to-treat analysis. J Hepatol 2017; 67: 92-99 [PMID: 28257902 DOI: 10.1016/j.jhep.2017.02.022]Llovet JM, Ricci S, Mazzaferro V, Hilgard P, Gane E, Blanc JF, de Oliveira AC, Santoro A, Raoul JL, Forner A, Schwartz M, Porta C, Zeuzem S, Bolondi L, Greten TF, Galle PR, Seitz JF, Borbath I, Häussinger D, Giannaris T, Shan M, Moscovici M, Voliotis D, Bruix J; SHARP Investigators Study Group. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 2008; 359: 378-390 [PMID: 18650514 DOI: 10.1056/NEJMoa0708857]

99

Cheng AL, Kang YK, Chen Z, Tsao CJ, Qin S, Kim JS, Luo R, Feng J, Ye S, Yang TS, Xu J, Sun Y, Liang H, Liu J, Wang J, Tak WY, Pan H, Burock K, Zou J, Voliotis D, Guan Z. Efficacy and safety of sorafenib in patients in the Asia-Pacific region with advanced hepatocellular carcinoma: a phase III randomised, double-blind, placebo-controlled trial. Lancet Oncol 2009; 10: 25-34 [PMID: 19095497 DOI: 10.1016/S1470-2045(08)70285-7]

100

Rim CH, Kim CY, Yang DS, Yoon WS. Comparison of radiation therapy modalities for hepatocellular carcinoma with portal vein thrombosis: A meta-analysis and systematic review. Radiother Oncol 2018; 129: 112-122 [PMID: 29233562 DOI: 10.1016/j.radonc.2017.11.013]

101

Bettinger D, Pinato DJ, Schultheiss M, Sharma R, Rimassa L, Pressiani T, Burlone ME, Pirisi M, Kudo M, Park JW, Buettner N, Neumann-Haefelin C, Boettler T, Abbasi-Senger N, Alheit H, Baus W, Blanck O, Gerum S, Guckenberger M, Habermehl D, Ostheimer C, Riesterer O, Tamihardja J, Grosu AL, Thimme R, Brunner TB, Gkika E. Stereotactic Body Radiation Therapy as an Alternative Treatment for Patients with Hepatocellular Carcinoma Compared to Sorafenib: A Propensity Score Analysis. Liver Cancer 2019; 8: 281-294 [PMID: 31602371 DOI: 10.1159/000490260]

102

Leung HW, Liu CF, Chan AL. Cost-effectiveness of sorafenib versus SBRT for unresectable advanced hepatocellular carcinoma. Radiat Oncol 2016; 11: 69 [PMID: 27193904 DOI: 10.1186/s13014-016-0644-4]

103

Brade AM, Ng S, Brierley J, Kim J, Dinniwell R, Ringash J, Wong RR, Cho C, Knox J, Dawson LA. Phase 1 Trial of Sorafenib and Stereotactic Body Radiation Therapy for Hepatocellular Carcinoma. Int J Radiat Oncol Biol Phys 2016; 94: 580-587 [PMID: 26867886 DOI: 10.1016/j.ijrobp.2015.11.048]

104

Teitz-Tennenbaum S, Li Q, Rynkiewicz S, Ito F, Davis MA, McGinn CJ, Chang AE. Radiotherapy potentiates the therapeutic efficacy of intratumoral dendritic cell administration. Cancer Res 2003; 63: 8466-8475 [PMID: 14679011]

105

Chi KH, Liu SJ, Li CP, Kuo HP, Wang YS, Chao Y, Hsieh SL. Combination of conformal radiotherapy and intratumoral injection of adoptive dendritic cell immunotherapy in refractory hepatoma. J Immunother 2005; 28: 129-135 [PMID: 15725956 DOI: 10.1097/01.cji.0000154248.74383.5e]

106

Teitz-Tennenbaum S, Li Q, Okuyama R, Davis MA, Sun R, Whitfield J, Knibbs RN, Stoolman LM, Chang AE. Mechanisms involved in radiation enhancement of intratumoral dendritic cell therapy. J Immunother 2008; 31: 345-358 [PMID: 18391761 DOI: 10.1097/CJI.0b013e318163628c]

107

Rai V, Abdo J, Alsuwaidan AN, Agrawal S, Sharma P, Agrawal DK. Cellular and molecular targets for the immunotherapy of hepatocellular carcinoma. Mol Cell Biochem 2018; 437: 13-36 [PMID: 28593566 DOI: 10.1007/s11010-017-3092-z]

108

Choi C, Yoo GS, Cho WK, Park HC. Optimizing radiotherapy with immune checkpoint blockade in hepatocellular carcinoma. World J Gastroenterol 2019; 25: 2416-2429 [PMID: 31171886 DOI: 10.3748/wjg.v25.i20.2416]

109

Kreidieh M, Zeidan YH, Shamseddine A. The Combination of Stereotactic Body Radiation Therapy and Immunotherapy in Primary Liver Tumors. J Oncol 2019; 2019: 4304817 [PMID: 31182960 DOI: 10.1155/2019/4304817]

110

Bernstein MB, Krishnan S, Hodge JW, Chang JY. Immunotherapy and stereotactic ablative radiotherapy (ISABR): a curative approach? Nat Rev Clin Oncol 2016; 13: 516-524 [PMID: 26951040 DOI: 10.1038/nrclinonc.2016.30]

111

Sharabi AB, Nirschl CJ, Kochel CM, Nirschl TR, Francica BJ, Velarde E, Deweese TL, Drake CG. Stereotactic Radiation Therapy Augments Antigen-Specific PD-1-Mediated Antitumor Immune Responses via Cross-Presentation of Tumor Antigen. Cancer Immunol Res 2015; 3: 345-355 [PMID: 25527358 DOI: 10.1158/2326-6066.CIR-14-0196]

112

Lee Y, Auh SL, Wang Y, Burnette B, Meng Y, Beckett M, Sharma R, Chin R, Tu T, Weichselbaum RR, Fu YX. Therapeutic effects of ablative radiation on local tumor require CD8+ T cells: changing strategies for cancer treatment. Blood 2009; 114: 589-595 [PMID: 19349616 DOI: 10.1182/blood-2009-02-206870]

113

Kim HJ, Park S, Kim KJ, Seong J. Clinical significance of soluble programmed cell death ligand-1 (sPD-L1) in hepatocellular carcinoma patients treated with radiotherapy. Radiother Oncol 2018; 129: 130-135 [PMID: 29366520 DOI: 10.1016/j.radonc.2017.11.027]

114

Vanpouille-Box C, Formenti SC, Demaria S. TREX1 dictates the immune fate of irradiated cancer cells. Oncoimmunology 2017; 6: e1339857 [PMID: 28932642 DOI: 10.1080/2162402X.2017.1339857]

115

Vanpouille-Box C, Alard A, Aryankalayil MJ, Sarfraz Y, Diamond JM, Schneider RJ, Inghirami G, Coleman CN, Formenti SC, Demaria S. DNA exonuclease Trex1 regulates radiotherapy-induced

116




https://dx.doi.org/10.1056/NEJMoa0708857


https://dx.doi.org/10.1016/S1470-2045(08)70285-7




https://dx.doi.org/10.1159/000490260


https://dx.doi.org/10.1186/s13014-016-0644-4





https://dx.doi.org/10.1097/01.cji.0000154248.74383.5e


https://dx.doi.org/10.1097/CJI.0b013e318163628c


https://dx.doi.org/10.1007/s11010-017-3092-z




https://dx.doi.org/10.1155/2019/4304817


https://dx.doi.org/10.1038/nrclinonc.2016.30


https://dx.doi.org/10.1158/2326-6066.CIR-14-0196


https://dx.doi.org/10.1182/blood-2009-02-206870




https://dx.doi.org/10.1080/2162402X.2017.1339857



tumour immunogenicity. Nat Commun 2017; 8: 15618 [PMID: 28598415 DOI: 10.1038/ncomms15618]Park HJ, Griffin RJ, Hui S, Levitt SH, Song CW. Radiation-induced vascular damage in tumors: implications of vascular damage in ablative hypofractionated radiotherapy (SBRT and SRS). Radiat Res 2012; 177: 311-327 [PMID: 22229487 DOI: 10.1667/rr2773.1]

117

Dovedi SJ, Adlard AL, Lipowska-Bhalla G, McKenna C, Jones S, Cheadle EJ, Stratford IJ, Poon E, Morrow M, Stewart R, Jones H, Wilkinson RW, Honeychurch J, Illidge TM. Acquired resistance to fractionated radiotherapy can be overcome by concurrent PD-L1 blockade. Cancer Res 2014; 74: 5458-5468 [PMID: 25274032 DOI: 10.1158/0008-5472.CAN-14-1258]

118

Tian S, Switchenko JM, Buchwald ZS, Patel PR, Shelton JW, Kahn SE, Pillai RN, Steuer CE, Owonikoko TK, Behera M, Curran WJ, Higgins KA. Lung Stereotactic Body Radiation Therapy and Concurrent Immunotherapy: A Multicenter Safety and Toxicity Analysis. Int J Radiat Oncol Biol Phys 2020; 108: 304-313 [PMID: 31982496 DOI: 10.1016/j.ijrobp.2019.12.030]

119

Tang C, Welsh JW, de Groot P, Massarelli E, Chang JY, Hess KR, Basu S, Curran MA, Cabanillas ME, Subbiah V, Fu S, Tsimberidou AM, Karp D, Gomez DR, Diab A, Komaki R, Heymach JV, Sharma P, Naing A, Hong DS. Ipilimumab with Stereotactic Ablative Radiation Therapy: Phase I Results and Immunologic Correlates from Peripheral T Cells. Clin Cancer Res 2017; 23: 1388-1396 [PMID: 27649551 DOI: 10.1158/1078-0432.CCR-16-1432]

120

Wegner RE, Abel S, Hasan S, White RJ, Finley G, Monga D, Colonias A, Verma V. Time from stereotactic radiotherapy to immunotherapy is a predictor for outcome in stage IV non-small cell lung cancer. J Immunol Sci 2019; 3: 6-13 [DOI: 10.29245/2578-3009/2019/2.1171]

121

Chiang CL, Chan ACY, Chiu KWH, Kong FS. Combined Stereotactic Body Radiotherapy and Checkpoint Inhibition in Unresectable Hepatocellular Carcinoma: A Potential Synergistic Treatment Strategy. Front Oncol 2019; 9: 1157 [PMID: 31799176 DOI: 10.3389/fonc.2019.01157]

122

El-Khoueiry AB, Sangro B, Yau T, Crocenzi TS, Kudo M, Hsu C, Kim TY, Choo SP, Trojan J, Welling TH Rd, Meyer T, Kang YK, Yeo W, Chopra A, Anderson J, Dela Cruz C, Lang L, Neely J, Tang H, Dastani HB, Melero I. Nivolumab in patients with advanced hepatocellular carcinoma (CheckMate 040): an open-label, non-comparative, phase 1/2 dose escalation and expansion trial. Lancet 2017; 389: 2492-2502 [PMID: 28434648 DOI: 10.1016/S0140-6736(17)31046-2]

123

Zhu AX, Finn RS, Edeline J, Cattan S, Ogasawara S, Palmer D, Verslype C, Zagonel V, Fartoux L, Vogel A, Sarker D, Verset G, Chan SL, Knox J, Daniele B, Webber AL, Ebbinghaus SW, Ma J, Siegel AB, Cheng AL, Kudo M; KEYNOTE-224 investigators. Pembrolizumab in patients with advanced hepatocellular carcinoma previously treated with sorafenib (KEYNOTE-224): a non-randomised, open-label phase 2 trial. Lancet Oncol 2018; 19: 940-952 [PMID: 29875066 DOI: 10.1016/S1470-2045(18)30351-6]

124

Finn RS, Ryoo BY, Merle P, Kudo M, Bouattour M, Lim HY, Breder V, Edeline J, Chao Y, Ogasawara S, Yau T, Garrido M, Chan SL, Knox J, Daniele B, Ebbinghaus SW, Chen E, Siegel AB, Zhu AX, Cheng AL; KEYNOTE-240 investigators. Pembrolizumab As Second-Line Therapy in Patients With Advanced Hepatocellular Carcinoma in KEYNOTE-240: A Randomized, Double-Blind, Phase III Trial. J Clin Oncol 2020; 38: 193-202 [PMID: 31790344 DOI: 10.1200/JCO.19.01307]

125

Yau T, Park JW, Finn RS, Cheng A-L, Mathurin P, Edeline J, Kudo M, Han K-H, Harding JJ, Merle P, Rosmorduc O, Wyrwicz L, Schott E, Choo SP, Kelley RK, Begic D, Chen G, Neely J, Anderson J, Sangro B. LBA38_PR - CheckMate 459: A randomized, multi-center phase III study of nivolumab (NIVO) vs sorafenib (SOR) as first-line (1L) treatment in patients (pts) with advanced hepatocellular carcinoma (aHCC). Ann Oncol 2019; 30: v874-v875 [DOI: 10.1093/annonc/mdz394.029]

126

Waidmann O. Recent developments with immunotherapy for hepatocellular carcinoma. Expert Opin Biol Ther 2018; 18: 905-910 [PMID: 29995439 DOI: 10.1080/14712598.2018.1499722]

127

Choi SH, Seong J. Stereotactic Body Radiotherapy: Does It Have a Role in Management of Hepatocellular Carcinoma? Yonsei Med J 2018; 59: 912-922 [PMID: 30187697 DOI: 10.3349/ymj.2018.59.8.912]

128

Salama JK, Milano MT. Radical irradiation of extracranial oligometastases. J Clin Oncol 2014; 32: 2902-2912 [PMID: 25113765 DOI: 10.1200/JCO.2014.55.9567]

129

Kissel M, Martel-Lafay I, Lequesne J, Faivre JC, Le Péchoux C, Stefan D, Barraux V, Loiseau C, Grellard JM, Danhier S, Lerouge D, Chouaid C, Gervais R, Thariat J. Stereotactic ablative radiotherapy and systemic treatments for extracerebral oligometastases, oligorecurrence, oligopersistence and oligoprogression from lung cancer. BMC Cancer 2019; 19: 1237 [PMID: 31856742 DOI: 10.1186/s12885-019-6449-8]

130

Kam TY, Chan OSH, Hung AWM, Yeung RMW. Utilization of stereotactic ablative radiotherapy in oligometastatic & oligoprogressive skeletal metastases: Results and pattern of failure. Asia Pac J Clin Oncol 2019; 15 Suppl 2: 14-19 [PMID: 30859749 DOI: 10.1111/ajco.13115]

131

Cheung P. Stereotactic body radiotherapy for oligoprogressive cancer. Br J Radiol 2016; 89: 20160251 [PMID: 27556349 DOI: 10.1259/bjr.20160251]

132

Berghen C, Joniau S, Ost P, Poels K, Everaerts W, Decaestecker K, Haustermans K, Devos G, De Meerleer G. Progression-directed Therapy for Oligoprogression in Castration-refractory Prostate Cancer. Eur Urol Oncol 2021; 4: 305-309 [PMID: 31558422 DOI: 10.1016/j.euo.2019.08.012]

133

Wray J, Hawamdeh RF, Hasija N, Dagan R, Yeung AR, Lightsey JL, Okunieff P, Daily KC, George TJ, Zlotecki RA, Trevino J, Dang LH. Stereotactic body radiation therapy for oligoprogression of metastatic disease from gastrointestinal cancers: A novel approach to extend

134


https://dx.doi.org/10.1038/ncomms15618


https://dx.doi.org/10.1667/rr2773.1


https://dx.doi.org/10.1158/0008-5472.CAN-14-1258




https://dx.doi.org/10.1158/1078-0432.CCR-16-1432

https://dx.doi.org/10.29245/2578-3009/2019/2.1171




https://dx.doi.org/10.1016/S0140-6736(17)31046-2


https://dx.doi.org/10.1016/S1470-2045(18)30351-6


https://dx.doi.org/10.1200/JCO.19.01307

https://dx.doi.org/10.1093/annonc/mdz394.029


https://dx.doi.org/10.1080/14712598.2018.1499722


https://dx.doi.org/10.3349/ymj.2018.59.8.912




https://dx.doi.org/10.1186/s12885-019-6449-8


https://dx.doi.org/10.1111/ajco.13115


https://dx.doi.org/10.1259/bjr.20160251


https://dx.doi.org/10.1016/j.euo.2019.08.012



chemotherapy efficacy. Oncol Lett 2017; 13: 1087-1094 [PMID: 28454218 DOI: 10.3892/ol.2016.5540]Palma DA, Olson R, Harrow S, Gaede S, Louie AV, Haasbeek C, Mulroy L, Lock M, Rodrigues GB, Yaremko BP, Schellenberg D, Ahmad B, Griffioen G, Senthi S, Swaminath A, Kopek N, Liu M, Moore K, Currie S, Bauman GS, Warner A, Senan S. Stereotactic ablative radiotherapy versus standard of care palliative treatment in patients with oligometastatic cancers (SABR-COMET): a randomised, phase 2, open-label trial. Lancet 2019; 393: 2051-2058 [PMID: 30982687 DOI: 10.1016/S0140-6736(18)32487-5]

135

Molinari M, De Coutere S, Krahn M, Helton S, Urbach DR. Patients' preferences and trade-offs for the treatment of early stage hepatocellular carcinoma. J Surg Res 2014; 189: 57-67 [PMID: 24650457 DOI: 10.1016/j.jss.2014.02.015]

136

Liang SX, Zhu XD, Xu ZY, Zhu J, Zhao JD, Lu HJ, Yang YL, Chen L, Wang AY, Fu XL, Jiang GL. Radiation-induced liver disease in three-dimensional conformal radiation therapy for primary liver carcinoma: the risk factors and hepatic radiation tolerance. Int J Radiat Oncol Biol Phys 2006; 65: 426-434 [PMID: 16690430 DOI: 10.1016/j.ijrobp.2005.12.031]

137

Gargett M, Haddad C, Kneebone A, Booth JT, Hardcastle N. Clinical impact of removing respiratory motion during liver SABR. Radiat Oncol 2019; 14: 93 [PMID: 31159840 DOI: 10.1186/s13014-019-1300-6]

138

Nomiya T, Akamatsu H, Harada M, Ota I, Hagiwara Y, Ichikawa M, Miwa M, Kawashiro S, Hagiwara M, Chin M, Hashizume E, Nemoto K. Modified simultaneous integrated boost radiotherapy for an unresectable huge refractory pelvic tumor diagnosed as a rectal adenocarcinoma. World J Gastroenterol 2014; 20: 18480-18486 [PMID: 25561820 DOI: 10.3748/wjg.v20.i48.18480]

139

Nomiya T, Akamatsu H, Harada M, Ota I, Hagiwara Y, Ichikawa M, Miwa M, Mizutani M, Kato T, Nagaoka A, Tomita Y, Nemoto K. Modified simultaneous integrated boost radiotherapy for large retroperitoneal malignant tumor: A case report. Oncol Lett 2015; 9: 2520-2524 [PMID: 26137100 DOI: 10.3892/ol.2015.3095]

140

Nomiya T, Akamatsu H, Harada M, Ota I, Hagiwara Y, Ichikawa M, Miwa M, Suzuki A, Nemoto K. Modified simultaneous integrated boost radiotherapy for unresectable locally advanced breast cancer: preliminary results of a prospective clinical trial. Clin Breast Cancer 2015; 15: 161-167 [PMID: 25459068 DOI: 10.1016/j.clbc.2014.09.011]

141

Crane CH, Koay EJ. Solutions that enable ablative radiotherapy for large liver tumors: Fractionated dose painting, simultaneous integrated protection, motion management, and computed tomography image guidance. Cancer 2016; 122: 1974-1986 [PMID: 26950735 DOI: 10.1002/cncr.29878]

142

Byun HK, Kim HJ, Im YR, Kim DY, Han KH, Seong J. Dose escalation in radiotherapy for incomplete transarterial chemoembolization of hepatocellular carcinoma. Strahlenther Onkol 2020; 196: 132-141 [PMID: 31286149 DOI: 10.1007/s00066-019-01488-9]

143

Mohan R, Wu Q, Manning M, Schmidt-Ullrich R. Radiobiological considerations in the design of fractionation strategies for intensity-modulated radiation therapy of head and neck cancers. Int J Radiat Oncol Biol Phys 2000; 46: 619-630 [PMID: 10701741 DOI: 10.1016/s0360-3016(99)00438-1]

144

Byun HK, Kim HJ, Im YR, Kim DY, Han KH, Seong J. Dose escalation by intensity modulated radiotherapy in liver-directed concurrent chemoradiotherapy for locally advanced BCLC stage C hepatocellular carcinoma. Radiother Oncol 2019; 133: 1-8 [PMID: 30935563 DOI: 10.1016/j.radonc.2018.12.025]

145

Simoni N, Micera R, Paiella S, Guariglia S, Zivelonghi E, Malleo G, Rossi G, Addari L, Giuliani T, Pollini T, Cavedon C, Salvia R, Milella M, Bassi C, Mazzarotto R. Hypofractionated Stereotactic Body Radiation Therapy With Simultaneous Integrated Boost and Simultaneous Integrated Protection in Pancreatic Ductal Adenocarcinoma. Clin Oncol (R Coll Radiol) 2021; 33: e31-e38 [PMID: 32682686 DOI: 10.1016/j.clon.2020.06.019]

146

Oehlke O, Wucherpfennig D, Fels F, Frings L, Egger K, Weyerbrock A, Prokic V, Nieder C, Grosu AL. Whole brain irradiation with hippocampal sparing and dose escalation on multiple brain metastases: Local tumour control and survival. Strahlenther Onkol 2015; 191: 461-469 [PMID: 25592907 DOI: 10.1007/s00066-014-0808-9]

147

Mazzola R, Ruggieri R, Figlia V, Rigo M, Giaj Levra N, Ricchetti F, Nicosia L, Corradini S, Alongi F. Stereotactic body radiotherapy of central lung malignancies using a simultaneous integrated protection approach : A prospective observational study. Strahlenther Onkol 2019; 195: 719-724 [PMID: 30689026 DOI: 10.1007/s00066-018-01419-0]

148

Li QS, Liang N, Ouyang WW, Su SF, Ma Z, Geng YC, Yang WG, Hu YX, Li HQ, Lu B. Simultaneous integrated boost of intensity-modulated radiation therapy to Stage II-III non-small cell lung cancer with metastatic lymph nodes. Cancer Med 2020; 9: 8364-8372 [PMID: 32902184 DOI: 10.1002/cam4.3446]

149

Qiu J, Lv B, Fu M, Wang X, Zheng X, Zhuo W. 18 F-Fluoromisonidazole positron emission tomography/CT-guided volumetric-modulated arc therapy-based dose escalation for hypoxic subvolume in nasopharyngeal carcinomas: A feasibility study. Head Neck 2017; 39: 2519-2527 [PMID: 28963789 DOI: 10.1002/hed.24925]

150


https://dx.doi.org/10.3892/ol.2016.5540


https://dx.doi.org/10.1016/S0140-6736(18)32487-5


https://dx.doi.org/10.1016/j.jss.2014.02.015




https://dx.doi.org/10.1186/s13014-019-1300-6






https://dx.doi.org/10.1016/j.clbc.2014.09.011




https://dx.doi.org/10.1007/s00066-019-01488-9


https://dx.doi.org/10.1016/s0360-3016(99)00438-1




https://dx.doi.org/10.1016/j.clon.2020.06.019


https://dx.doi.org/10.1007/s00066-014-0808-9


https://dx.doi.org/10.1007/s00066-018-01419-0




https://dx.doi.org/10.1002/hed.24925


World Journal of



REVIEW

Open reading frame 3 protein of hepatitis E virus: Multi-function protein with endless potential

Yong-Lin Yang, Yu-Chen Nan

ORCID number: Yong-Lin Yang 0000-0003-3162-4070; Yu-Chen Nan 0000-0002-4442-8004.

Author contributions: Yang YL and Nan YC prepared the main body of this manuscript; Nan YC reviewed and revised the manuscript, and designed and prepared the figures; all authors approved the manuscript for publication.

Supported by National Natural Science Foundation of China, No. 31672534; Key Project supported by Medical Science and Technology Development Foundation of Nanjing Department of Health, No. ZKX19026.

Conflict-of-interest statement: The authors declare no conflict of interests for this article.

Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: htt

Yong-Lin Yang, Department of Infectious Diseases, Taizhou People's Hospital, The Fifth Affiliated Hospital of Nantong University, Taizhou 225300, Jiangsu Province, China

Yong-Lin Yang, Department of General Practice, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, Jiangsu Province, China

Yu-Chen Nan, Department of Preventive Veterinary Medicine, Northwest A&F University, Yangling 712100, Shaanxi Province, China

Corresponding author: Yu-Chen Nan, PhD, Associate Professor, Department of Preventive Veterinary Medicine, Northwest A&F University, No. 3 Taicheng Road, Yangling Demonstration Zone, Yangling 712100, Shaanxi Province, China. [email protected]

AbstractHepatitis E virus (HEV), a fecal-orally transmitted foodborne viral pathogen, causes acute hepatitis in humans and is responsible for hepatitis E outbreaks worldwide. Since the identification of HEV as a zoonotic agent, this virus has been isolated from a variety of hosts with an ever-expanding host range. HEV-open reading frame (ORF) 3, the smallest ORF in HEV genomes, initially had been perceived as an unremarkable HEV accessory protein. However, as novel HEV-ORF3 function has been discovered that is related to the existence of a putative third virion structural form, referred to as “quasi-enveloped” HEV particles, HEV is challenging the conventional virion structure-based classification scheme, which assigns all viruses to two groups, “enveloped” or “non-enveloped”. In this review, we systematically describe recent progress that has identified multiple pathogenic roles of HEV-ORF3, including roles in HEV virion release, biogenesis of quasi-enveloped virus, regulation of the host innate immune response, and interference with host signaling pathways. In addition, implications of HEV-ORF3-associated quasi-enveloped virions are discussed to guide future development of improved vaccines against zoonotic HEV infection.

Key Words: Hepatitis E virus; Zoonosis; Quasi-enveloped virion; Hepatitis E virus-open reading frame 3; Innate immunity




http://orcid.org/0000-0003-3162-4070

http://orcid.org/0000-0003-3162-4070

http://orcid.org/0000-0002-4442-8004

http://orcid.org/0000-0002-4442-8004



Yang YL et al. HEV ORF3 protein


p://creativecommons.org/Licenses/by-nc/4.0/



Country/Territory of origin: China

Peer-review report’s scientific quality classificationGrade A (Excellent): A Grade B (Very good): B, B Grade C (Good): C Grade D (Fair): 0 Grade E (Poor): 0

Received: January 23, 2021 Peer-review started: January 23, 2021 First decision: February 25, 2021 Revised: March 10, 2021 Accepted: April 12, 2021 Article in press: April 12, 2021 Published online: May 28, 2021

P-Reviewer: Chiu KW, Filipec Kanizaj T, McQuillan GM, Vaughan G S-Editor: Fan JR L-Editor: Wang TQ P-Editor: Liu JH

Core Tip: Hepatitis E virus (HEV)-open reading frame (ORF) 3 was originally though as an accessory protein with limited function which is not essential for HEV replication. This view has been challenged by recent discoveries, such as HEV-ORF3-associated “quasi-enveloped” HEV particles, regulation of the host innate immune response, and interference with host signaling pathways. More novel function of HEV-ORF3 will be revealed.

Citation: Yang YL, Nan YC. Open reading frame 3 protein of hepatitis E virus: Multi-function protein with endless potential. World J Gastroenterol 2021; 27(20): 2458-2473URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2458.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2458

INTRODUCTIONHepatitis E virus (HEV), a quasi-enveloped, single-stranded positive-sense RNA virus, is classified as a member of the family Hepeviridae[1]. Hepeviridae is a highly diverse family that contains several HEV and HEV-like virus species with zoonotic, anthropo-tropic, and animal-restricted tropisms[2]. Currently, nearly 3 million symptomatic cases of HEV infection are reported annually, resulting in approximately 70000 deaths and 3000 stillbirths in each year[3]. Generally, mortality of HEV ranges from 0.5% to 3% overall, but HEV mortality rates have approached 30% in pregnant women[4,5].

The viral genome of HEV is 7.2 kb in length and is an mRNA-like molecule (capped and poly-adenylated at 5' and 3' ends, respectively)[6]. To date, three well-defined open reading frames (ORFs) have been detected in all HEV genotypes studied (Figure 1)[7,8]. HEV-ORF1 protein is translated directly from HEV genome with HEV-ORF2 and-ORF3 proteins translated from subgenomic RNAs[9]. Moreover, ORF4, whose expression is promoted by an atypical internal ribosome entry site (IRES)-like element, completely overlaps with ORF1 and was identified recently only in HEV-1 isolates[10]. In addition to ORFs, HEV genome contains at least four cis-reactive elements (CREs) that are required for viral replication in vivo[11-13]. Two of these CREs, which are located within intergenic-junctional regions between HEV-ORFs, form “stem-loop” structures that act as promoter-like elements for initiation of subgenomic RNA synthesis[11,12]. Conversely, another two CREs within ORF1 and ORF2, function as a scaffold that generates specific signals that trigger recruitment of viral and host factors for a replication complex[13].

The latest classification system of HEVs includes two genera within the family Hepeviridae, Orthohepevirus (covering all HEV isolates with mammalian and avian origin) and Piscihepevirus (only HEV-like virus with cutthroat trout origin). All four previously characterized HEV genotypes (1-4) that cause human infection are categorized within the species Orthohepevirus A[1], with Orthohepevirus B, C, and D species encompassing HEV isolates found in other non-human hosts[1]. Within Orthohepevirus A, HEV-1 and HEV-2 isolates are anthropotropic viruses without any animal reservoirs, while HEV-3 and HEV-4 isolates are zoonotic[6,14]. The HEV isolates originally identified from Japanese wild boar, containing unique RNA sequences, are categorized into HEV-5 and 6, whereas camel HEV isolates belong to HEV-7 and HEV-8 genotypes[1,15]. Notably, HEV isolated from a human liver transplant patient has been reported to most closely match camel HEV, suggesting that camel HEVs may be zoonotic as well, although this concept requires further confirmation[16]. In addition to mammalian HEVs, two unique groups of HEV-like viruses that have been isolated from avian species and cutthroat trout (Oncorhynchus clarkia) have been assigned to species Orthohepevirus B and to genus Piscihepevirus (as the only member), respectively.

Initially, HEV was assumed to be solely restricted to humans, in whom it induced self-limiting hepatitis symptoms. However, the emergence of HEV and HEV-like isolates in swine and other animal species supports a much wider HEV host tropism, with some HEV genotypes identified as zoonotic pathogens[17]. Currently, hepatitis E cases have been frequently reported in both developing and developed countries and have occurred in step with increasingly more frequent observations of expanding host ranges[18-22]. Generally, inter-species transmission and infection of zoonotic type of virus from animal to humans is considered to be primary routing of HEV transmission







Figure 1 Schematic illustration of hepatitis E virus genome and three well-defined open reading frames. The numbers above or below the RNA boxes indicate nucleotide numbers based on hepatitis E virus-1 prototype Sar55 strain (GenBank accession number AF444002). NCR: Non-coding region; ORF: Open reading frame.

within Western worlds, while fecal-oral transmission appears to be the predominant route of HEV transmission within developing countries[23,24]. Notably, immune serological surveillance data support a high prevalence rate of previous HEV infection in the general population but demonstrate a declining trend in recent years that may be due to undetected endemic HEV circulation[25,26]. Meanwhile, frequent detection of recent cases of chronic HEV, HEV-related acute hepatic failure, and extrahepatic HEV manifestations supports this speculation as well[25,27-30]. Moreover, these observations imply that zoonotic HEV infection is a complicated pathogenic process underlying various forms of HEV-related disease. However, our understanding of HEV remains restricted due to the lack of a robust in vitro HEV cell system. Nevertheless, the HEV genome is known to contain three well-defined ORFs that have been found in all HEV genotypes[7,8]. HEV-ORF3, the smallest ORF found in HEV genomes, encodes a unique protein with multiple indispensable functional roles associated with viral replication and pathogenesis. In this review, we discuss the recent progress toward understanding HEV-ORF3 pathological roles in detail and provide new insights.

PROTEINS ENCODED BY HEV HEV-ORF1 polyprotein as viral replicaseThe HEV-ORF1 protein is the largest protein encoded by the HEV genome and can be directly translated from the mRNA-like genome of HEV[8]. Bioinformatics-based protein homology analysis indicated that at least eight function domains are present within HEV-ORF1 according to similarities to counterparts from other RNA viruses (Figure 2)[31]. These protein domains include methyltransferase domain, the Y domain, papain-like cysteine protease (PCP) domain, a hypervariable region containing previously assigned hypervariable domain and proline-rich domain, the X domain (also named macro-domain), RNA helicase domain, and a RNA polymerase domain (RdRp)[2,32]. The methyltransferase domain and Y domain together are thought to constitute the functional unit of RNA capping enzyme[2,32]. It remains unknown whether HEV-ORF1 protein acts alone to perform all putative viral replicase functions or is cleaved by host protease or viral protease (via PCP domains) to generate independent units resembling viral replicases similar to other positive-sense RNA viruses[2,33-35]. To date, available data suggests that proteolytic cleavage of the HEV-ORF1 product involves PCP domain encoded within HEV-ORF1[2,33-35]. Besides viral replication, another notable characteristic of HEV-ORF1 protein is its flexibility for insertions or deletions within the hypervariable region. The proline-rich domain within the hypervariable region was proposed to act as a linking hinge for the upstream PCP domain and downstream macro-domain, leading to formation of an unstable tertiary structure[31,36-38]. Conversely, the hypervariable region is considered an intrinsically disordered region containing extensive gene segment insertions or deletions and may participate in viral replication[39-42]. As a unique example, one HEV-3 isolate, Kernow-C1 p6 strain, which was originally isolated from an HIV patient chronically infected by HEV as well, contains a 174-nt insertion of a human ribosomal protein S17 sequence originating from the host[43], which confers ORF1 protein with novel nuclear/nucleolar trafficking capability and may promote viral replication in vitro[44,45].



Figure 2 Function domains of hepatitis E virus-open reading frame 1 polyprotein. Putative functional domains of hepatitis E virus (HEV)-open reading frame 1 polyprotein based on HEV-1 prototype Sar55 strain are listed as follows: Methyltransferase domain; Y domain; papain-like cysteine protease; hypervariable region; proline-rich domain; X-domain; Hhelicase; RNA-dependent RNA polymerase. Met: Methyltransferase domain; Y: Y domain; PCP: Papain-like cysteine protease; HV: Hypervariable region; Pro: Proline-rich domain; X: X-domain; Hel: Helicase; RdRp: RNA polymerase; ORF: Open reading frame.

HEV-ORF2 encodes viral capsidsHEV-ORF2 encodes the putative capsid protein of HEV virions with a full-size of 660 aa residues, and has a predicted molecular mass of 72 kDa[46]. Notable, the full length ORF2 protein carries N-linked glycans at three putative glycosylation Asn residues at positions 137, 310, and 562, as well as a 15 aa signal peptide at the N-terminus directing full length ORF2 protein to the endoplasmic reticulum (ER)[47]. One notable characteristic of HEV-ORF2 is the existence of various forms of ORF2-derived proteins with multiple functions. Researchers observed very early that multiple processed ORF2-derived products were detected when recombinant ORF2 was expressed in different systems. The mature HEV capsid protein is generated from the full-length ORF2 precursor via proteolytic processing to remove the first N-terminal 111 aa and the last C-terminal 52 aa. However, at least two other forms of HEV-ORF2 have been detected in HEV infected cells and patients[48,49]. The first was a secreted form of ORF2 protein (ORF2s), which utilized an upstream start codon and contained a signal peptide that earmarked ORF2 for subsequent glycosylation and secretion[50]; the second was a capsid-associated truncated form of ORF2 (ORF2c), which was translated beginning at an internal methionine-encoding AUG start codon (aa16 of ORF2)[50]. Meanwhile, a truncated form of ORF2 (ORF2c) detected in HEV-infected cells may be secreted into the extracellular milieu, as is ORF2s[49].

The mature HEV capsid that lacks N-terminal 111 aa and the last C-terminal 52 aa of full ORF2 can form virus-like particles when expressed in insect cells[51,52]. Genetic analysis of ORF2 sequences of HEV genotypes 1-4 suggests that these proteins share greater than 85% similarity overall, with divergence mainly observed within the first N-terminal 111 aa, which are not incorporated into final virions[53]. A recombinant subunit vaccine using truncated HEV-1 ORF2 protein (HEV239) as major immunogen is licensed in China (Hecolin®)[2].

A multi-functional protein encoded by HEV-ORF3ORF3, the smallest ORF among all HEV ORFs, partially overlaps with the N-terminus of ORF2 for about 300 nt and is translated from a different reading frame[9]. It was initially proposed that ORF3 protein contains 123 aa that are encoded by a subgenomic RNA distinct from the RNA encoding ORF2[7]. However, it was later confirmed that ORF3 protein is translated from a bicistronic subgenomic RNA to generate a 114-aa protein with a predicted molecular weight of 13 kDa (vp13), which is actually 9 aa shorter than the initially predicted length[9,54]. Basic sequence analysis of HEV-ORF3 protein indicated that there are two hydrophobic domains and two proline-rich domains present within the N-terminal half and C-terminal portion of HEV-ORF3 proteins, respectively[55,56], of which the first proline-rich domain contains a mitogen-activated protein kinase (MAPK) phosphorylation site (Ser71)[57]. Furthermore, two presenilin-associated protein (PSAP) motifs within the ORF3 protein were identified in HEV-1 prototype strain Sar55, with the first PSAP motif comprised of aa 86-89 and the second comprised of aa 95-98[58]. Although ORF3 protein is not required for viral RNA replication in vitro[59], it is irreplaceable for in vivo HEV infection and required for viral particle releasing[54,60,61]. In fact, most studies demonstrated that HEV-ORF3 is indispensable for viral particle egress and biogenesis of lipid membrane-wrapped HEV particles, which is now recognized as quasi-enveloped particles. Moreover, the second PSAP motif within the HEV-ORF3 protein has been shown to be required for the formation of membrane-associated HEV particles, a process that relies on an association of ORF3 with lipids[58,62]. Currently, HEV-ORF3 is thought to form an ion channel that shares key structural features with



class I viroporins that are required for virion particle release from cells during infection[63]. This observation aligns with the reported putative role of pORF3[61] and with other evidence indicating that HEV-ORF3 protein interacts with tumor suscept-ibility gene 101 (TSG101), the key component of host endosomal sorting complex required for transport (ESCRT) pathway, which is mainly employed by enveloped virus for budding and acquiring of viral envelope. Formation of ESCRT complex has been shown to lead to biogenesis of quasi-enveloped HEV particles[64-67].

HEV-1 specific ORF4 In recent years, a hidden ORF4 from HEV-1 was characterized[10]. Translation of HEV-ORF4 protein is promoted by a novel element located in HEV-ORF1 appearing to be an atypical IRES sequence and works in combination with a suboptimal Kozak sequence[10]. The exact function of HEV-1 specific pORF4 is still unclear. It was demonstrated that pORF4 stimulates ER stress upon HEV replication[10]. It also interacts with multiple ORF1 domains in vitro which are presumably to form a complex further enhancing RdRp activity [10]. Furthermore, HEV-pORF4 specific antibodies are detectable in HEV-infected patients[10]. Nonetheless, additional invest-igations are needed to understand functions of the ORF4 product that are unique to genotype 1 HEVs.

REGULATION OF HOST INNATE IMMUNITY AND SIGNALING BY HEV-ORF3 PROTEINInitially, HEV-ORF3 protein did not receive much attention, due to its presumed role as an accessory protein involved in regulation of host signaling to promote HEV replication and invasion. This hypothesis was partially evidenced by the fact that ORF3 protein was dispensable for in vitro replication of HEV-RNA[59]. However, subsequent research studies demonstrated that HEV-ORF3-associated putative interference mechanisms acted on multiple host cell signaling pathways, such as those involved in host innate immunity[2,68,69], indicating that HEV-ORF3 activities ultimately promote viral replication and pathogenesis.

As a multifunctional protein, HEV-ORF3 has been demonstrated to play both positive and negative roles in interferon (IFN) induction. In our previous research, we found that HEV-ORF3 protein could enhance retinoic acid-inducible gene I (RIG-I) activation to subsequently enhance IFN induction[68]. More specifically, HEV-1 ORF3 extended protein half-life of RIG-I and interacted with the RIG-I N-terminal portion to enhance ubiquitination-mediated RIG-I activation triggered by addition of the dsRNA analog poly (I:C)[68] (Figure 3A). Interestingly, it is notable that genotypic differences in HEV-ORF3-associated enhancement of RIG-I-mediated IFN induction were observed. For example, ORF3 proteins from the HEV-1 Sar55 strain and HEV-3 kernowC1 p6 strain could enhance RIG-I activation, while HEV-2 and HEV-4 ORF3 proteins could not[68], suggesting that HEV-ORF3 participated in genotype-specific HEV virulence and pathogenic effects. Moreover, these results also aligned with results of a more recent report demonstrating HEV-ORF3-associated increases of IFN-α/β and interferon-stimulated gene 15 levels in hepatoma cell line HepG2/C3A[70]. Conversely, other reports have demonstrated that overexpression of HEV-ORF3 downregulated Toll-like receptor (TLR) 3 and TLR7 and their downstream signaling pathways[71,72] (Figure 3B). Meanwhile, another study has demonstrated that ORF3 protein from an HEV-1 strain interacted with signal transducer and activator of transcription (STAT) 1 to block type I IFN-activated pathway[73] (Figure 3C). Furthermore, another study found that ORF3 proteins blocked nuclear translocation of STAT3 to down-regulate STAT3-dependent gene expression, including expression of acute-phase response proteins[74].

Besides the innate immune response, yeast two-hybrid-based screening detected that binding of ORF3 to Pyst1, a MAPK phosphatase, led to activation of MAPK pathways[75]. Thus, ORF3 appears to regulate host gene expression, as MAPK is related to host gene expression and signaling. Meanwhile, HEV-ORF3 may promote expression of glycolytic pathway enzymes by enhancing phosphorylation and transactivation function of p300/CREB-binding protein as well[76]. Additionally, microarray analysis of Huh7 cells has suggested that liver-specific genes may also be modulated by HEV-ORF3, since it modulated phosphorylation of hepatocyte nuclear factor 4[77]. Moreover, recent research has demonstrated that HEV-ORF3 plays a functional role in virus-cell interactions by influencing expression of integral membrane protein and basement membrane proteins to alter host cell processes



Figure 3 Regulation of host innate immune response by hepatitis E virus-open reading frame 3 proteins. A: Promotion of retinoic acid-inducible gene I mediated activation by hepatitis E virus (HEV)-open reading frame (ORF) 3; B: Inhibition of Toll-like receptor (TLR) 3 and TLR7 by HEV-ORF3; C: Blocking of the phosphorylation of signal transducer and activator of transcription (STAT) 1 to inhibit Janus kinase/STAT signaling. RIG-I: Retinoic acid-inducible gene I; RIP-I: Ribosome-inactivating proteins type I; FADD: Fas-associated protein with death domain; IKK: IkappaB kinase; TBK: Tank-Binding-Kinase; IRF: Interferon regulatory factor; NF-kB: Nuclear factor kB; IFN: Interferon; TLR: Toll-like receptor; TRIF: Toll-interleukin 1 receptor domain-containing adapter inducing interferon-beta; NAPI: Net anthropogenic phosphorus input; IFNAR: Inflammation-the type I interferon receptor; JAK: Janus kinase; TYK: Targeting tyrosine kinase; STAT: Signal transducer and activator of transcription; ISGF: Interferon-stimulated gene factor; ISRE: Interferon-stimulated response element; ISGs: Interferon-stimulated genes; CBP: CREB-binding protein.



associated with apoptosis and lipid metabolism[78]. Taken together, these data suggest that HEV-ORF3 modulates multiple signaling pathways, including those involved in host innate immunity, to ultimately promote HEV pathogenesis.

INVOVEMENT OF HEV-ORF3 PROTEIN IN BIOGENESIS OF QUASI-ENVELOPED HEV PARTICLES The presence of a lipid layer-based envelope has long been used as the basic criterion for virus classification[66]. The presence of a lipid layer can be assessed through treatment of virus preparations with bile salts, a process that abrogates infectivity of enveloped virions, but not non-enveloped virions, by removing their surface lipid layers[66]. Generally, it is believed that during the virion budding process a viral envelope is formed from membranes of infected cells that contain molecules of at least one membrane-embedded virus-encoded glycoprotein (presented as peplomers). Viral envelopes interact with corresponding virus receptors located on target cells to promote membrane fusion of cellular membrane and viral envelop after initial interactions of virions and corresponding receptors. Meanwhile, surface glycoproteins located in viral envelope serve as antibody-neutralization targets in most cases[66], while the lipid layer of the virus envelope prevents internal virus antigens, such as nucleocapsid proteins, from serving as viral neutralization targets[79]. Thus, as compared to a non-enveloped virus, a quasi-enveloped virion would be perceived by the host immune system as antigenically distinct from a naked virion. For example, hepatitis A virus (HAV) was the first non-enveloped virus which is confirmed to hijack host cell membrane similar to enveloped virus as an enveloped form[80]. Biogenesis of enveloped HAV particles has been shown to depend on the ESCRT system[80,81], which is involved in budding of enveloped viruses. Membrane-wrapped or enveloped HAV particles mainly exist in circulation system during acute infection phase of HEV and envelopment confers protection of virus from recognition by neutralizing antibodies, which prevents impairment of virion infectivity[80]. Similar to HAV, HEV was originally classified solely as a non-enveloped virus before membrane-wrapped HEV particles were discovered, with HEV-ORF3 involvement in biogenesis of quasi-enveloped HEV virions confirmed only very recently.

It was observed very early that HEV-ORF3 protein is not required for viral RNA replication in vitro[59]; however, this protein is irreplaceable for HEV replication in vivo and is required for viral particle release from HEV infected cells in vitro[54,60,61]. Meanwhile, antibody-capture assays of HEV virions with or without detergent treatment demonstrated that HEV virion from either serum samples of patients or supernatants of HEV-infected cells were associated with lipid layer and ORF3 protein[82]. Subsequent screening to detect proteins interacting with HEV-ORF3 protein pinpointed the TSG101, a component of the ESCRT complex, as the potential interacting partner of HEV-ORF3[62]. Importantly, the ESCRT complex recognizes and earmarks ubiquitinated proteins for subsequent incorporation into multivesicular bodies (MVBs), an essential step for lysosomal degradation[83]. The ubiquitin E2 variant domain in TSG101 recognizes and interacts with the P(T/S)AP motif present in target proteins to recruit targets to the endosomal membrane[84].

Many enveloped viruses are equipped with a P(T/S)AP motif within their structural proteins that interact with TSG101 to redirect assembled viral components to cell membrane for virion release from infected cells[85]. Originally, two PSAP motifs comprised of aa 86-89 and aa 95-98 were identified within HEV-ORF3 proteins, the second of which was found to be conserved among HEV genotypes[58]. Replacement of the PSAP motif in HEV-ORF3 proteins with heterologous domain motifs (PPPY, YPDL, and PSAA) or mutated PSAP motifs has been shown to affect HEV virion release from infected cells[58,86], including avian HEV release[87]. Additionally, HEV-ORF3 expressed in cells was found to associate with the cellular cytoskeleton fraction, with deletion of the N-terminal hydrophobic domain of vp13 abolishing this association[57]. Moreover, a more recent study demonstrated that green fluorescent protein-tagged ORF3 protein interacted with cellular microtubules and modulated microtubule dynamics[55]. This microtubule-like filament of HEV-ORF3 protein indicated that it was potentially involved in a process that promotes virus egress; this process is reminiscent of the process by which the pUL37 protein of herpes virus interacts with dystonin, an important cytoskeleton cross-linker involved in microtubule-based transport of capsids during virion egress[88]. Although HEV was originally defined as a non-enveloped virus, such HEV-ORF3 functions that were formerly attributed only to enveloped viruses may now be related to HEV in its



recently discovered “quasi-enveloped” form[64-67].Consistent with the aforementioned roles of HEV-ORF3 in virus egress[61], a recent

study found that HEV-ORF3 shares key features with class I viroporins, including its function as an ion channel participating in viral particle egress or release[63], a function that had been previously demonstrated for the well-characterized viroporin of influenza A virus matrix-2 (M2) protein[63]. Meanwhile, a putative transmembrane region identified in pORF3 may be involved in ER localization of this protein[63]. Since viroporins of other viruses, such as M2 of influenza A virus, are components of virions, these observations imply that HEV-ORF3 is a structural HEV virion protein that exists in a membrane-associated state during the formation of envelope structures of quasi-enveloped HEV virions[89]. Interestingly, palmitoylation of cysteine residues within the N-terminal region of HEV-ORF3 has been shown to participate in its association with the membrane and is also required for infectious particle secretion[90].

The unique role that HEV-ORF3 plays during biogenesis of quasi-enveloped virus particles makes it a novel target candidate for antiviral drug development. In fact, one study has shown that the addition of a cyclic peptide inhibitor (CPI) to HEV-infected cells interrupted the interaction between the HEV-ORF3 PSAP motif and TSG101[85] and reduced virion release by over 90% when a 50% inhibitory concentration of CPI of 2 μM was used. Thus, HEV-OR3 has potential as a novel candidate for further development as an anti-HEV drug.

POTENTIAL ROLES OF HEV-ORF3 PROTEIN IN HEV HOST TROPSIMSince the isolation of zoonotic HEV strains from swine HEV and discovery of other HEV-like viral isolates, HEVs have been continually identified from various mammalian hosts. Based on their ability to cause inter-species infection, HEV isolates can be divided three distinct groups: HEV-1 and 2 are only restricted in human; HEV-3, 4, and 7/8 are zoonotic types; while Orthohepevirus C is animal-restricted type. Based on reports in the literature, it implies that either factors or viral determinants contribute to HEV host tropisms and cross-species transmission events.

Among all ORFs, HEV-ORF1 encodes the largest HEV protein and appears to be indispensable for determining HEV host range. An in vitro study demonstrated that swapping of genetic fragment among HEV-1 and HEV-4 infectious clones indicated that chimeric virus formed from an HEV-1 infectious clone bearing surface HEV-4 ORF1 could replicate in porcine kidney cells, while the original HEV-1 virus could not[91]. Meanwhile, chimeric virus containing the junction region between ORF1/2 and the 3' non-coding region (NCR) of HEV-3 or the 3' end of the HEV-1 backbone failed to infect piglets, suggesting that the 5′ NCR and ORF1 are involved in HEV cross-species infection[92]. By contrast, a recent report demonstrated, via genetic fragment swapping of ORF1 regions between HEV-1 and HEV-3 infectious clones, that recombinant chimeric viruses could be generated in vitro. In any case, these chimeric viruses could not infect piglets in vivo[93], suggesting that ORF1 is not the only determinant that can confer cross-species infectivity of HEV in vivo.

As the viral capsid protein, HEV-ORF2 was initially thought to be an unlikely determinant of host tropism, since it is conserved among all major genotypes infecting humans[94]. However, results of in vivo reverse genetics-based studies that swapped segments between different HEV genotypes indicated that HEV-3 or HEV-4 based chimeric viruses inserting ORF2 from HEV-1 was incapable to cause effective infection in swine[95]. Thus, it appears that HEV-ORF2 is also involved in HEV-interspecies infectivity, in agreement with results of another report demonstrating that replacement of the HEV-3 capsid region spanning aa 456 to 605 (the putative virus receptor-binding region) with corresponding region from HEV-1 prevented chimeric virus from entering and infecting swine cells[96]. Therefore, these data imply that in addition to ORF1, viral capsid proteins also determine host preference. Nevertheless, until cellular receptors for HEV are identified, the link between viral capsid residues and cellular receptor determinants underlying HEV host tropism still requires further investigation.

In addition to ORF1/2, other reports have demonstrated that ORF3 proteins may be involved in determining host range of HEV. Up to date, the literature suggests that HEV-ORF3 protein acting as an ion channel essentially resembles viroporins involved in viral particle release during HEV infection[63]. This viroporin-like function depends on the highly conserved PSAP motif spanning aa 95-98 within Orthohepevirus A, which has been proposed to interact with host TSG101. However, truncation analysis has



indicated that the N-terminal 25 aa of HEV-1 ORF3 are required for its association with microtubules as well as virus release[55]. Meanwhile, alignment of ORF3 aa sequences of all eight HEV genotypes indicates that the region containing the M-terminal 25 aa of different HEV-ORF3 proteins are more conserved than the rest of HEV-ORF3. Therefore, the conservation of this region may reflect the conserved role of HEV-ORF3-dependent virion release across all genotypes. Nonetheless, a recent study demonstrated that rat HEV-ORF3 protein possessed the PxYPMP motif in place of the original PSAP motif found in human HEV-ORF3 proteins[97]. Intriguingly, unlike human HEV-ORF3 proteins, rat HEV-ORF3 proteins did not bind to TSG101, but instead utilized MVB-based sorting to achieve virion release; this mechanism differed from the aforementioned TSG101-dependent mechanism for effecting release of human HEV from infected cells[97]. Thus, these results taken together imply that HEV-ORF3 may have an important species-specific function.

Meanwhile, except for the abovementioned conserved motifs, less homology is observed elsewhere in the HEV-ORF3 protein, especially within its C-terminal half (aa 62 to aa 114) (Figure 4), a region that appears to be important for adaptation to various hosts. It is also notable that genomic locations of ORF3-encoding genes vary among species of Orthohepevirus (either of partially or fully overlapping with ORF2, Figure 5)[98,99], which implies a genotype-specific evolution pattern influencing functions of HEV-ORF3 that affect HEV host tropism. This speculation is in line with a genotype-specific enhancement of IFN induction by HEV-ORF3 proteins observed in our previous report[68]. Therefore, the mechanism by which a genotype-specific function of ORF3 product influences HEV host tropism requires further confirmation, although the accumulating literature indicates that a host-specific function exists that may influence host tropism by HEV-ORF3 proteins.

HEV-ORF PROTEIN AS VACCINE TARGET FOR HEV Since the discovery of quasi-enveloped virions, researchers have tried to determine if these particles differ from classically enveloped virions, since the outer lipid bilayers of quasi-enveloped particles, such as those of HAV, are devoid of any viral proteins[66,80], while both HAV virion forms are equally infectious[80]. This apparent paradox raises the question of how membrane-wrapped particles can infect cells in the absence of viral peplomers that are generally thought to be required for enveloped virus infectivity[66]. Nevertheless, HEV appears to differ from HAV, since researchers observed very early before (prior to the identification of quasi-enveloped HEV particles) that HEV-ORF3 protein specific monoclonal antibody (mAb) could capture viral particles from serum samples of HEV patients or cell culture supernatants of HEV infected cells[89]. It is now clear that HEV-ORF3 protein associates with the lipid layer in quasi-enveloped virions produced both in vitro and in vivo, while HEV virions from feces fail to be captured by this mAb due to the lack of the HEV-ORF3-containing envelope[89]. This observation was recently confirmed by electron microscopy showing that immunogold-labeled mAb recognizing HEV-ORF3 proteins bound to quasi-enveloped HEV particles as well[100]. Moreover, although in vitro infectivity appears to be equivalent between quasi-enveloped HAV particles and naked counterparts, quasi-enveloped HEV particles infect fresh cells in a less efficient manner in vitro, as reflected by their need for a longer inoculation time to achieve maximal infectivity[67]. Meanwhile, it appears that cell entry by quasi-enveloped HEV virions depends on endosomal trafficking, which can be abrogated by blocking endosomal acidification[67]. Furthermore, additional investigations have demonstrated that HEV-ORF3 protein acts on ion channel protein and participates in the release of infectious virions from infected cells[63]; this role is similar to that of other well-characterized viroporins such as M2 protein of influenza A virus[63]. It is also notable that two hydrophobic domains located in N-terminal half of HEV-ORF3 demonstrated unique functions, whereby the first one is required associating with microtubules[55], while the second one contains a putative transmembrane region involved in ER localization[63]. Since viroporins of other viruses, such as M2 protein of influenza A virus, are components of virions, these observations imply that HEV-ORF3 is also a structural component of HEV virions, although it is not known if antibody-based neutralization mechanisms differ between the two types of HEV particles. Nevertheless, since HEV-ORF3 is present within quasi-enveloped HEV virions[89], it would be an interesting question to be determined if HEV-ORF3-specific antibodies are capable to neutralize quasi-enveloped HEV viral particles since capsid specific antibodies fail to do so.



Figure 4 Alignment of amino acid sequence of hepatitis E virus-open reading frame 3 from 4 genotypes in Orthohepevirus A virus. Alignment of amino acid sequence of open reading frame 3 from all seven genotypes classified as Orthohepevirus A virus. Hepatitis E virus (HEV)-1 (reference sequence GenBank accession #M73218), HEV-2 (reference sequence GenBank accession #M74506), HEV-3 (reference sequence GenBank accession #AF082843), and HEV-4 (reference sequence GenBank accession #AJ272108) are shown. Those residues that are the same as consensus sequence are shown as “.”. HEV: Hepatitis E virus.

Figure 5 Hepatitis E virus genome organization. Genome location of open reading frame 3 among different hepeviruses. HEV: Hepatitis E virus; NCR: Non-coding region; CRE: Cis-reactive element; ORF: Open reading frame.

It remains unclear whether HEV-ORF3 acts as a potential neutralizing target for HEV. A previous report suggested that a recombinant vaccine candidate using HEV-4 ORF3 protein fused with interleukin-1β might confer partial protection against virus challenge[101]. Moreover, similar to vaccines based on HEV-4 ORF3, our research on avian HEV demonstrated that chickens immunized with recombinant avian HEV-ORF3 protein showed partial protection upon challenge and had milder disease symptoms than did controls[102]. However, it is notable that challenge experiments for these vaccines only employed virus stocks obtained from fecal samples which contained only naked viral particles (without envelopes containing HEV-ORF3 protein)[102]. Therefore, it is possible that antibodies induced by current ORF3-based vaccines cannot prevent first-round infection during initial challenge with naked virus, since naked HEV virions cannot be neutralized by antibodies specific for ORF3-HEV. Nevertheless, partial protection observed in both experiments may be due to ORF3-specific antibody-based neutralization of newly synthesized quasi-enveloped HEV virion entering into circulation after initial infection caused by naked HEV virion used for challenging. In any case, these observations raise the interesting question of whether an ORF2-based vaccine could protect hosts from challenge with quasi-enveloped HEV particles, a concept that warrants further investigation.

Up to date, available data indicated that ORF3 proteins (including avian HEV-ORF3) is highly immunogenic to evoke host humoral response, with most B-cell epitopes located at the C-terminal of HEV-ORF3[103-106]. Thus, the C-terminal half (about 60 aa) of HEV-ORF3 proteins appears to be a promising candidate as a recombinant subunit vaccine. However, an obstacle to employing this region for a vaccine candidate is the potential antigen variation issue as predicted by aa sequence



alignment of HEV-ORF3s from the four reference strains of HEV-1 to -4 (Figure 4)[68]. Therefore, systematic mapping of antigenic epitopes within HEV-ORF3 proteins of different HEV genotypes may be required before using immunogenic ORF3 epitopes as additional components of HEV subunit vaccines.

CONCLUSIONAlthough nearly three decades have elapsed since the identification and character-ization of the complete genome sequence from the first HEV isolate, the full spectrum of this virus remains unclear and HEV infection now is a public health concern in developed countries as well. Currently, cross-species infection and host tropisms of different HEV genotypes remain elusive, due to the lack of easy-to-handle animal model and a robust in vitro system for studying HEV. Although many details about this virus and its pathology have been revealed in recent years, it is notable that HEV-ORF3 protein, the smallest ORF encoded by HEV, appears to have diverse functions and key roles in HEV virion release, biogenesis of quasi-enveloped virus, regulation of the host innate immune response, and neutralization of quasi-enveloped virus. These advances will guide further studies to reveal the basic biology of HEV, functions of HEV proteins, and HEV pathogenic factors toward the development of effective therapeutics and an improved vaccine.

REFERENCESSmith DB, Simmonds P; Members Of The International Committee On The Taxonomy Of Viruses Study Group, Jameel S, Emerson SU, Harrison TJ, Meng XJ, Okamoto H, Van der Poel WHM, Purdy MA. Consensus proposals for classification of the family Hepeviridae. J Gen Virol 2014; 95: 2223-2232 [PMID: 24989172 DOI: 10.1099/vir.0.068429-0]

1

Nan Y, Zhang YJ. Molecular Biology and Infection of Hepatitis E Virus. Front Microbiol 2016; 7: 1419 [PMID: 27656178 DOI: 10.3389/fmicb.2016.01419]

2

Dalton HR, Saunders M, Woolson KL. Hepatitis E virus in developed countries: one of the most successful zoonotic viral diseases in human history? J Virus Erad 2015; 1: 23-29 [PMID: 27482393]

3

Jameel S. Molecular biology and pathogenesis of hepatitis E virus. Expert Rev Mol Med 1999; 1999: 1-16 [PMID: 14987355 DOI: 10.1017/S1462399499001271]

4

Pillot J, Türkoglu S, Dubreuil P, Cosson A, Lemaigre G, Meng J, Lazizi Y. Cross-reactive immunity against different strains of the hepatitis E virus transferable by simian and human sera. C R Acad Sci III 1995; 318: 1059-1064 [PMID: 8556452]

5

Ahmad I, Holla RP, Jameel S. Molecular virology of hepatitis E virus. Virus Res 2011; 161: 47-58 [PMID: 21345356 DOI: 10.1016/j.virusres.2011.02.011]

6

Tam AW, Smith MM, Guerra ME, Huang CC, Bradley DW, Fry KE, Reyes GR. Hepatitis E virus (HEV): molecular cloning and sequencing of the full-length viral genome. Virology 1991; 185: 120-131 [PMID: 1926770]

7

Tsarev SA, Emerson SU, Reyes GR, Tsareva TS, Legters LJ, Malik IA, Iqbal M, Purcell RH. Characterization of a prototype strain of hepatitis E virus. Proc Natl Acad Sci USA 1992; 89: 559-563 [PMID: 1731327]

8

Graff J, Torian U, Nguyen H, Emerson SU. A bicistronic subgenomic mRNA encodes both the ORF2 and ORF3 proteins of hepatitis E virus. J Virol 2006; 80: 5919-5926 [PMID: 16731930 DOI: 10.1128/JVI.00046-06]

9

Nair VP, Anang S, Subramani C, Madhvi A, Bakshi K, Srivastava A, Shalimar, Nayak B, Ranjith Kumar CT, Surjit M. Endoplasmic Reticulum Stress Induced Synthesis of a Novel Viral Factor Mediates Efficient Replication of Genotype-1 Hepatitis E Virus. PLoS Pathog 2016; 12: e1005521 [PMID: 27035822 DOI: 10.1371/journal.ppat.1005521]

10

Cao D, Huang YW, Meng XJ. The nucleotides on the stem-loop RNA structure in the junction region of the hepatitis E virus genome are critical for virus replication. J Virol 2010; 84: 13040-13044 [PMID: 20943962 DOI: 10.1128/JVI.01475-10]

11

Parvez MK. The intergenic-junction variant (genotype 2 isolate) of hepatitis E virus restores the CREX 'stem-loop' structural integrity, essential for viral life cycle. Gene 2015; 559: 149-154 [PMID: 25597766 DOI: 10.1016/j.gene.2015.01.033]

12

Ju X, Xiang G, Gong M, Yang R, Qin J, Li Y, Nan Y, Yang Y, Zhang QC, Ding Q. Identification of functional cis-acting RNA elements in the hepatitis E virus genome required for viral replication. PLoS Pathog 2020; 16: e1008488 [PMID: 32433693 DOI: 10.1371/journal.ppat.1008488]

13

Meng XJ. Recent advances in Hepatitis E virus. J Viral Hepat 2010; 17: 153-161 [PMID: 20040046 DOI: 10.1111/j.1365-2893.2009.01257.x]

14

Woo PC, Lau SK, Teng JL, Cao KY, Wernery U, Schountz T, Chiu TH, Tsang AK, Wong PC, Wong EY, Yuen KY. New Hepatitis E Virus Genotype in Bactrian Camels, Xinjiang, China, 2013.

15


https://dx.doi.org/10.1099/vir.0.068429-0


https://dx.doi.org/10.3389/fmicb.2016.01419



https://dx.doi.org/10.1017/S1462399499001271



https://dx.doi.org/10.1016/j.virusres.2011.02.011




https://dx.doi.org/10.1128/JVI.00046-06


https://dx.doi.org/10.1371/journal.ppat.1005521




https://dx.doi.org/10.1016/j.gene.2015.01.033







Emerg Infect Dis 2016; 22: 2219-2221 [PMID: 27869607 DOI: 10.3201/eid2212.160979]Lee GH, Tan BH, Teo EC, Lim SG, Dan YY, Wee A, Aw PP, Zhu Y, Hibberd ML, Tan CK, Purdy MA, Teo CG. Chronic Infection With Camelid Hepatitis E Virus in a Liver Transplant Recipient Who Regularly Consumes Camel Meat and Milk. Gastroenterology 2016; 150: 355-7. e3 [PMID: 26551551 DOI: 10.1053/j.gastro.2015.10.048]

16

Meng XJ. Zoonotic and foodborne transmission of hepatitis E virus. Semin Liver Dis 2013; 33: 41-49 [PMID: 23564388 DOI: 10.1055/s-0033-1338113]

17

Park WJ, Park BJ, Ahn HS, Lee JB, Park SY, Song CS, Lee SW, Yoo HS, Choi IS. Hepatitis E virus as an emerging zoonotic pathogen. J Vet Sci 2016; 17: 1-11 [PMID: 27051334 DOI: 10.4142/jvs.2016.17.1.1]

18

Montesano C, Giovanetti M, Ciotti M, Cella E, Lo Presti A, Grifoni A, Zehender G, Angeletti S, Ciccozzi M. Hepatitis E Virus Circulation in Italy: Phylogenetic and Evolutionary Analysis. Hepat Mon 2016; 16: e31951 [PMID: 27226798 DOI: 10.5812/hepatmon.31951]

19

Anheyer-Behmenburg HE, Szabo K, Schotte U, Binder A, Klein G, Johne R. Hepatitis E Virus in Wild Boars and Spillover Infection in Red and Roe Deer, Germany, 2013-2015. Emerg Infect Dis 2017; 23: 130-133 [PMID: 27983488 DOI: 10.3201/eid2301.161169]

20

Abravanel F, Lhomme S, El Costa H, Schvartz B, Peron JM, Kamar N, Izopet J. Rabbit Hepatitis E Virus Infections in Humans, France. Emerg Infect Dis 2017; 23: 1191-1193 [PMID: 28628452 DOI: 10.3201/eid2307.170318]

21

Doceul V, Bagdassarian E, Demange A, Pavio N. Zoonotic Hepatitis E Virus: Classification, Animal Reservoirs and Transmission Routes. Viruses 2016; 8 [PMID: 27706110 DOI: 10.3390/v8100270]

22

Pavio N, Meng XJ, Doceul V. Zoonotic origin of hepatitis E. Curr Opin Virol 2015; 10: 34-41 [PMID: 25588602 DOI: 10.1016/j.coviro.2014.12.006]

23

Salines M, Andraud M, Rose N. From the epidemiology of hepatitis E virus (HEV) within the swine reservoir to public health risk mitigation strategies: a comprehensive review. Vet Res 2017; 48: 31 [PMID: 28545558 DOI: 10.1186/s13567-017-0436-3]

24

Sadik S, van Rijckevorsel GG, van Rooijen MS, Sonder GJ, Bruisten SM. Seroprevalence of hepatitis E virus differs in Dutch and first generation migrant populations in Amsterdam, the Netherlands: a cross-sectional study. BMC Infect Dis 2016; 16: 659 [PMID: 27825308 DOI: 10.1186/s12879-016-2007-z]

25

Teshale EH, Denniston MM, Drobeniuc J, Kamili S, Teo CG, Holmberg SD. Decline in hepatitis E virus antibody prevalence in the United States from 1988-1994 to 2009-2010. J Infect Dis 2015; 211: 366-373 [PMID: 25147277 DOI: 10.1093/infdis/jiu466]

26

Dalton HR, Kamar N, van Eijk JJ, Mclean BN, Cintas P, Bendall RP, Jacobs BC. Hepatitis E virus and neurological injury. Nat Rev Neurol 2016; 12: 77-85 [PMID: 26711839 DOI: 10.1038/nrneurol.2015.234]

27

Geng Y, Zhao C, Huang W, Harrison TJ, Zhang H, Geng K, Wang Y. Detection and assessment of infectivity of hepatitis E virus in urine. J Hepatol 2016; 64: 37-43 [PMID: 26362822 DOI: 10.1016/j.jhep.2015.08.034]

28

Feng Z. Causation by HEV of extrahepatic manifestations remains unproven. Liver Int 2016; 36: 477-479 [PMID: 27005694 DOI: 10.1111/liv.13085]

29

Ripellino P, Pasi E, Melli G, Staedler C, Fraga M, Moradpour D, Sahli R, Aubert V, Martinetti G, Bihl F, Bernasconi E, Terziroli Beretta-Piccoli B, Cerny A, Dalton HR, Zehnder C, Mathis B, Zecca C, Disanto G, Kaelin-Lang A, Gobbi C. Neurologic complications of acute hepatitis E virus infection. Neurol Neuroimmunol Neuroinflamm 2020; 7 [PMID: 31806684 DOI: 10.1212/NXI.0000000000000643]

30

Koonin EV, Gorbalenya AE, Purdy MA, Rozanov MN, Reyes GR, Bradley DW. Computer-assisted assignment of functional domains in the nonstructural polyprotein of hepatitis E virus: delineation of an additional group of positive-strand RNA plant and animal viruses. Proc Natl Acad Sci USA 1992; 89: 8259-8263 [PMID: 1518855]

31

Ahola T, Karlin DG. Sequence analysis reveals a conserved extension in the capping enzyme of the alphavirus supergroup, and a homologous domain in nodaviruses. Biol Direct 2015; 10: 16 [PMID: 25886938 DOI: 10.1186/s13062-015-0050-0]

32

Parvez MK. Molecular characterization of hepatitis E virus ORF1 gene supports a papain-like cysteine protease (PCP)-domain activity. Virus Res 2013; 178: 553-556 [PMID: 23978667 DOI: 10.1016/j.virusres.2013.07.020]

33

Paliwal D, Panda SK, Kapur N, Varma SPK, Durgapal H. Hepatitis E virus (HEV) protease: a chymotrypsin-like enzyme that processes both non-structural (pORF1) and capsid (pORF2) protein. J Gen Virol 2014; 95: 1689-1700 [PMID: 24795447 DOI: 10.1099/vir.0.066142-0]

34

Parvez MK. The hepatitis E virus nonstructural polyprotein. Future Microbiol 2017; 12: 915-924 [PMID: 28686042 DOI: 10.2217/fmb-2017-0016]

35

Dosztányi Z, Chen J, Dunker AK, Simon I, Tompa P. Disorder and sequence repeats in hub proteins and their implications for network evolution. J Proteome Res 2006; 5: 2985-2995 [PMID: 17081050 DOI: 10.1021/pr060171o]

36

Dunker AK, Oldfield CJ, Meng J, Romero P, Yang JY, Chen JW, Vacic V, Obradovic Z, Uversky VN. The unfoldomics decade: an update on intrinsically disordered proteins. BMC Genomics 2008; 9 Suppl 2: S1 [PMID: 18831774 DOI: 10.1186/1471-2164-9-S2-S1]

37

Tsai CJ, Ma B, Sham YY, Kumar S, Nussinov R. Structured disorder and conformational selection. Proteins 2001; 44: 418-427 [PMID: 11484219]

38


https://dx.doi.org/10.3201/eid2212.160979


https://dx.doi.org/10.1053/j.gastro.2015.10.048


https://dx.doi.org/10.1055/s-0033-1338113


https://dx.doi.org/10.4142/jvs.2016.17.1.1


https://dx.doi.org/10.5812/hepatmon.31951






https://dx.doi.org/10.3390/v8100270


https://dx.doi.org/10.1016/j.coviro.2014.12.006


https://dx.doi.org/10.1186/s13567-017-0436-3




https://dx.doi.org/10.1093/infdis/jiu466


https://dx.doi.org/10.1038/nrneurol.2015.234






https://dx.doi.org/10.1212/NXI.0000000000000643



https://dx.doi.org/10.1186/s13062-015-0050-0






https://dx.doi.org/10.2217/fmb-2017-0016


https://dx.doi.org/10.1021/pr060171o


https://dx.doi.org/10.1186/1471-2164-9-S2-S1




Purdy MA. Evolution of the hepatitis E virus polyproline region: order from disorder. J Virol 2012; 86: 10186-10193 [PMID: 22811526 DOI: 10.1128/JVI.01374-12]

39

Purdy MA, Lara J, Khudyakov YE. The hepatitis E virus polyproline region is involved in viral adaptation. PLoS One 2012; 7: e35974 [PMID: 22545153 DOI: 10.1371/journal.pone.0035974]

40

Pudupakam RS, Huang YW, Opriessnig T, Halbur PG, Pierson FW, Meng XJ. Deletions of the hypervariable region (HVR) in open reading frame 1 of hepatitis E virus do not abolish virus infectivity: evidence for attenuation of HVR deletion mutants in vivo. J Virol 2009; 83: 384-395 [PMID: 18945785 DOI: 10.1128/JVI.01854-08]

41

Pudupakam RS, Kenney SP, Córdoba L, Huang YW, Dryman BA, Leroith T, Pierson FW, Meng XJ. Mutational analysis of the hypervariable region of hepatitis e virus reveals its involvement in the efficiency of viral RNA replication. J Virol 2011; 85: 10031-10040 [PMID: 21775444 DOI: 10.1128/JVI.00763-11]

42

Shukla P, Nguyen HT, Torian U, Engle RE, Faulk K, Dalton HR, Bendall RP, Keane FE, Purcell RH, Emerson SU. Cross-species infections of cultured cells by hepatitis E virus and discovery of an infectious virus-host recombinant. Proc Natl Acad Sci USA 2011; 108: 2438-2443 [PMID: 21262830 DOI: 10.1073/pnas.1018878108]

43

Kenney SP, Meng XJ. The lysine residues within the human ribosomal protein S17 sequence naturally inserted into the viral nonstructural protein of a unique strain of hepatitis E virus are important for enhanced virus replication. J Virol 2015; 89: 3793-3803 [PMID: 25609799 DOI: 10.1128/JVI.03582-14]

44

Kenney SP, Meng XJ. Identification and fine mapping of nuclear and nucleolar localization signals within the human ribosomal protein S17. PLoS One 2015; 10: e0124396 [PMID: 25853866 DOI: 10.1371/journal.pone.0124396]

45

Robinson RA, Burgess WH, Emerson SU, Leibowitz RS, Sosnovtseva SA, Tsarev S, Purcell RH. Structural characterization of recombinant hepatitis E virus ORF2 proteins in baculovirus-infected insect cells. Protein Expr Purif 1998; 12: 75-84 [PMID: 9473460 DOI: 10.1006/prep.1997.0817]

46

Jameel S, Zafrullah M, Ozdener MH, Panda SK. Expression in animal cells and characterization of the hepatitis E virus structural proteins. J Virol 1996; 70: 207-216 [PMID: 8523527]

47

Ankavay M, Montpellier C, Sayed IM, Saliou JM, Wychowski C, Saas L, Duvet S, Aliouat-Denis CM, Farhat R, de Masson d'Autume V, Meuleman P, Dubuisson J, Cocquerel L. New insights into the ORF2 capsid protein, a key player of the hepatitis E virus lifecycle. Sci Rep 2019; 9: 6243 [PMID: 31000788 DOI: 10.1038/s41598-019-42737-2]

48

Montpellier C, Wychowski C, Sayed IM, Meunier JC, Saliou JM, Ankavay M, Bull A, Pillez A, Abravanel F, Helle F, Brochot E, Drobecq H, Farhat R, Aliouat-Denis CM, Haddad JG, Izopet J, Meuleman P, Goffard A, Dubuisson J, Cocquerel L. Hepatitis E Virus Lifecycle and Identification of 3 Forms of the ORF2 Capsid Protein. Gastroenterology 2018; 154: 211-223. e8 [PMID: 28958858 DOI: 10.1053/j.gastro.2017.09.020]

49

Yin X, Ying D, Lhomme S, Tang Z, Walker CM, Xia N, Zheng Z, Feng Z. Origin, antigenicity, and function of a secreted form of ORF2 in hepatitis E virus infection. Proc Natl Acad Sci USA 2018; 115: 4773-4778 [PMID: 29669922 DOI: 10.1073/pnas.1721345115]

50

Li TC, Yamakawa Y, Suzuki K, Tatsumi M, Razak MA, Uchida T, Takeda N, Miyamura T. Expression and self-assembly of empty virus-like particles of hepatitis E virus. J Virol 1997; 71: 7207-7213 [PMID: 9311793]

51

Li TC, Takeda N, Miyamura T, Matsuura Y, Wang JC, Engvall H, Hammar L, Xing L, Cheng RH. Essential elements of the capsid protein for self-assembly into empty virus-like particles of hepatitis E virus. J Virol 2005; 79: 12999-13006 [PMID: 16189002 DOI: 10.1128/JVI.79.20.12999-13006.2005]

52

Mori Y, Matsuura Y. Structure of hepatitis E viral particle. Virus Res 2011; 161: 59-64 [PMID: 21440590 DOI: 10.1016/j.virusres.2011.03.015]

53

Huang YW, Opriessnig T, Halbur PG, Meng XJ. Initiation at the third in-frame AUG codon of open reading frame 3 of the hepatitis E virus is essential for viral infectivity in vivo. J Virol 2007; 81: 3018-3026 [PMID: 17202216 DOI: 10.1128/JVI.02259-06]

54

Kannan H, Fan S, Patel D, Bossis I, Zhang YJ. The hepatitis E virus open reading frame 3 product interacts with microtubules and interferes with their dynamics. J Virol 2009; 83: 6375-6382 [PMID: 19369329 DOI: 10.1128/JVI.02571-08]

55

Holla RP, Ahmad I, Ahmad Z, Jameel S. Molecular Virology of Hepatitis E Virus. Semin Liver Dis 2013; 33: 3-14 [PMID: 23564385 DOI: 10.1055/s-0033-1338110]

56

Zafrullah M, Ozdener MH, Panda SK, Jameel S. The ORF3 protein of hepatitis E virus is a phosphoprotein that associates with the cytoskeleton. J Virol 1997; 71: 9045-9053 [PMID: 9371561]

57

Nagashima S, Takahashi M, Jirintai, Tanaka T, Yamada K, Nishizawa T, Okamoto H. A PSAP motif in the ORF3 protein of hepatitis E virus is necessary for virion release from infected cells. J Gen Virol 2011; 92: 269-278 [PMID: 21068219 DOI: 10.1099/vir.0.025791-0]

58

Emerson SU, Nguyen H, Torian U, Purcell RH. ORF3 protein of hepatitis E virus is not required for replication, virion assembly, or infection of hepatoma cells in vitro. J Virol 2006; 80: 10457-10464 [PMID: 16928762 DOI: 10.1128/JVI.00892-06]

59

Graff J, Nguyen H, Yu C, Elkins WR, St Claire M, Purcell RH, Emerson SU. The open reading frame 3 gene of hepatitis E virus contains a cis-reactive element and encodes a protein required for infection of macaques. J Virol 2005; 79: 6680-6689 [PMID: 15890906 DOI: 10.1128/JVI.79.11.6680-6689.2005]

60




https://dx.doi.org/10.1371/journal.pone.0035974






https://dx.doi.org/10.1073/pnas.1018878108






https://dx.doi.org/10.1006/prep.1997.0817



https://dx.doi.org/10.1038/s41598-019-42737-2







https://dx.doi.org/10.1128/JVI.79.20.12999-13006.2005








https://dx.doi.org/10.1055/s-0033-1338110







https://dx.doi.org/10.1128/JVI.79.11.6680-6689.2005



Yamada K, Takahashi M, Hoshino Y, Takahashi H, Ichiyama K, Nagashima S, Tanaka T, Okamoto H. ORF3 protein of hepatitis E virus is essential for virion release from infected cells. J Gen Virol 2009; 90: 1880-1891 [PMID: 19339479 DOI: 10.1099/vir.0.010561-0]

61

Nagashima S, Takahashi M, Jirintai S, Tanaka T, Nishizawa T, Yasuda J, Okamoto H. Tumour susceptibility gene 101 and the vacuolar protein sorting pathway are required for the release of hepatitis E virions. J Gen Virol 2011; 92: 2838-2848 [PMID: 21880841 DOI: 10.1099/vir.0.035378-0]

62

Ding Q, Heller B, Capuccino JM, Song B, Nimgaonkar I, Hrebikova G, Contreras JE, Ploss A. Hepatitis E virus ORF3 is a functional ion channel required for release of infectious particles. Proc Natl Acad Sci USA 2017; 114: 1147-1152 [PMID: 28096411 DOI: 10.1073/pnas.1614955114]

63

Nagashima S, Takahashi M, Jirintai S, Tanggis, Kobayashi T, Nishizawa T, Okamoto H. The membrane on the surface of hepatitis E virus particles is derived from the intracellular membrane and contains trans-Golgi network protein 2. Arch Virol 2014; 159: 979-991 [PMID: 24221250 DOI: 10.1007/s00705-013-1912-3]

64

Hurley JH. The ESCRT complexes. Crit Rev Biochem Mol Biol 2010; 45: 463-487 [PMID: 20653365 DOI: 10.3109/10409238.2010.502516]

65

Feng Z, Hirai-Yuki A, McKnight KL, Lemon SM. Naked Viruses That Aren't Always Naked: Quasi-Enveloped Agents of Acute Hepatitis. Annu Rev Virol 2014; 1: 539-560 [PMID: 26958733 DOI: 10.1146/annurev-virology-031413-085359]

66

Yin X, Ambardekar C, Lu Y, Feng Z. Distinct Entry Mechanisms for Nonenveloped and Quasi-Enveloped Hepatitis E Viruses. J Virol 2016; 90: 4232-4242 [PMID: 26865708 DOI: 10.1128/JVI.02804-15]

67

Nan Y, Ma Z, Wang R, Yu Y, Kannan H, Fredericksen B, Zhang YJ. Enhancement of interferon induction by ORF3 product of hepatitis E virus. J Virol 2014; 88: 8696-8705 [PMID: 24850742 DOI: 10.1128/JVI.01228-14]

68

Lhomme S, Migueres M, Abravanel F, Marion O, Kamar N, Izopet J. Hepatitis E Virus: How It Escapes Host Innate Immunity. Vaccines (Basel) 2020; 8 [PMID: 32731452 DOI: 10.3390/vaccines8030422]

69

Wang M, Huang Y, He M, Peng WJ, Tian DY. Effects of hepatitis E virus infection on interferon production via ISG15. World J Gastroenterol 2018; 24: 2173-2180 [PMID: 29853735 DOI: 10.3748/wjg.v24.i20.2173]

70

Lei Q, Li L, Zhang S, Li T, Zhang X, Ding X, Qin B. HEV ORF3 downregulates TLR7 to inhibit the generation of type I interferon via impairment of multiple signaling pathways. Sci Rep 2018; 8: 8585 [PMID: 29872132 DOI: 10.1038/s41598-018-26975-4]

71

He M, Wang M, Huang Y, Peng W, Zheng Z, Xia N, Xu J, Tian D. The ORF3 Protein of Genotype 1 Hepatitis E Virus Suppresses TLR3-induced NF-κB Signaling via TRADD and RIP1. Sci Rep 2016; 6: 27597 [PMID: 27270888 DOI: 10.1038/srep27597]

72

Dong C, Zafrullah M, Mixson-Hayden T, Dai X, Liang J, Meng J, Kamili S. Suppression of interferon-α signaling by hepatitis E virus. Hepatology 2012; 55: 1324-1332 [PMID: 22183878 DOI: 10.1002/hep.25530]

73

Chandra V, Kar-Roy A, Kumari S, Mayor S, Jameel S. The hepatitis E virus ORF3 protein modulates epidermal growth factor receptor trafficking, STAT3 translocation, and the acute-phase response. J Virol 2008; 82: 7100-7110 [PMID: 18448545 DOI: 10.1128/JVI.00403-08]

74

Kar-Roy A, Korkaya H, Oberoi R, Lal SK, Jameel S. The hepatitis E virus open reading frame 3 protein activates ERK through binding and inhibition of the MAPK phosphatase. J Biol Chem 2004; 279: 28345-28357 [PMID: 15096509 DOI: 10.1074/jbc.M400457200]

75

Moin SM, Chandra V, Arya R, Jameel S. The hepatitis E virus ORF3 protein stabilizes HIF-1alpha and enhances HIF-1-mediated transcriptional activity through p300/CBP. Cell Microbiol 2009; 11: 1409-1421 [PMID: 19489936 DOI: 10.1111/j.1462-5822.2009.01340.x]

76

Chandra V, Holla P, Ghosh D, Chakrabarti D, Padigaru M, Jameel S. The hepatitis E virus ORF3 protein regulates the expression of liver-specific genes by modulating localization of hepatocyte nuclear factor 4. PLoS One 2011; 6: e22412 [PMID: 21799848 DOI: 10.1371/journal.pone.0022412]

77

Xu K, Guo S, Zhao T, Zhu H, Jiao H, Shi Q, Pang F, Li Y, Li G, Peng D, Nie X, Cheng Y, Wu K, Du L, Cui K, Zhang W, Wang F. Transcriptome Analysis of HepG2 Cells Expressing ORF3 from Swine Hepatitis E Virus to Determine the Effects of ORF3 on Host Cells. Biomed Res Int 2016; 2016: 1648030 [PMID: 27648443 DOI: 10.1155/2016/1648030]

78

Pütz MM, Midgley CM, Law M, Smith GL. Quantification of antibody responses against multiple antigens of the two infectious forms of Vaccinia virus provides a benchmark for smallpox vaccination. Nat Med 2006; 12: 1310-1315 [PMID: 17086190 DOI: 10.1038/nm1457]

79

Feng Z, Hensley L, McKnight KL, Hu F, Madden V, Ping L, Jeong SH, Walker C, Lanford RE, Lemon SM. A pathogenic picornavirus acquires an envelope by hijacking cellular membranes. Nature 2013; 496: 367-371 [PMID: 23542590 DOI: 10.1038/nature12029]

80

Votteler J, Sundquist WI. Virus budding and the ESCRT pathway. Cell Host Microbe 2013; 14: 232-241 [PMID: 24034610 DOI: 10.1016/j.chom.2013.08.012]

81

Takahashi M, Tanaka T, Takahashi H, Hoshino Y, Nagashima S, Jirintai, Mizuo H, Yazaki Y, Takagi T, Azuma M, Kusano E, Isoda N, Sugano K, Okamoto H. Hepatitis E Virus (HEV) strains in serum samples can replicate efficiently in cultured cells despite the coexistence of HEV antibodies: characterization of HEV virions in blood circulation. J Clin Microbiol 2010; 48: 1112-1125 [PMID: 20107086 DOI: 10.1128/JCM.02002-09]

82








https://dx.doi.org/10.1007/s00705-013-1912-3


https://dx.doi.org/10.3109/10409238.2010.502516


https://dx.doi.org/10.1146/annurev-virology-031413-085359






https://dx.doi.org/10.3390/vaccines8030422




https://dx.doi.org/10.1038/s41598-018-26975-4


https://dx.doi.org/10.1038/srep27597






https://dx.doi.org/10.1074/jbc.M400457200






https://dx.doi.org/10.1155/2016/1648030


https://dx.doi.org/10.1038/nm1457


https://dx.doi.org/10.1038/nature12029


https://dx.doi.org/10.1016/j.chom.2013.08.012


https://dx.doi.org/10.1128/JCM.02002-09



Wen X, He X, Zhang X, Liu L, Guan Y, Zhang Y, Bu Z. Genome sequences derived from pig and dried blood pig feed samples provide important insights into the transmission of African swine fever virus in China in 2018. Emerg Microbes Infect 2019; 8: 303-306 [PMID: 30866784 DOI: 10.1080/22221751.2019.1565915]

83

Neilan JG, Zsak L, Lu Z, Kutish GF, Afonso CL, Rock DL. Novel swine virulence determinant in the left variable region of the African swine fever virus genome. J Virol 2002; 76: 3095-3104 [PMID: 11884534 DOI: 10.1128/jvi.76.7.3095-3104.2002]

84

Wang N, Zhao D, Wang J, Zhang Y, Wang M, Gao Y, Li F, Bu Z, Rao Z, Wang X. Architecture of African swine fever virus and implications for viral assembly. Science 2019; 366: 640-644 [PMID: 31624094 DOI: 10.1126/science.aaz1439]

85

Kenney SP, Wentworth JL, Heffron CL, Meng XJ. Replacement of the hepatitis E virus ORF3 protein PxxP motif with heterologous late domain motifs affects virus release via interaction with TSG101. Virology 2015; 486: 198-208 [PMID: 26457367 DOI: 10.1016/j.virol.2015.09.012]

86

Kenney SP, Pudupakam RS, Huang YW, Pierson FW, LeRoith T, Meng XJ. The PSAP motif within the ORF3 protein of an avian strain of the hepatitis E virus is not critical for viral infectivity in vivo but plays a role in virus release. J Virol 2012; 86: 5637-5646 [PMID: 22438540 DOI: 10.1128/JVI.06711-11]

87

Pasdeloup D, McElwee M, Beilstein F, Labetoulle M, Rixon FJ. Herpesvirus tegument protein pUL37 interacts with dystonin/BPAG1 to promote capsid transport on microtubules during egress. J Virol 2013; 87: 2857-2867 [PMID: 23269794 DOI: 10.1128/JVI.02676-12]

88

Takahashi M, Yamada K, Hoshino Y, Takahashi H, Ichiyama K, Tanaka T, Okamoto H. Monoclonal antibodies raised against the ORF3 protein of hepatitis E virus (HEV) can capture HEV particles in culture supernatant and serum but not those in feces. Arch Virol 2008; 153: 1703-1713 [PMID: 18679765 DOI: 10.1007/s00705-008-0179-6]

89

Gouttenoire J, Pollán A, Abrami L, Oechslin N, Mauron J, Matter M, Oppliger J, Szkolnicka D, Dao Thi VL, van der Goot FG, Moradpour D. Palmitoylation mediates membrane association of hepatitis E virus ORF3 protein and is required for infectious particle secretion. PLoS Pathog 2018; 14: e1007471 [PMID: 30532200 DOI: 10.1371/journal.ppat.1007471]

90

Chatterjee SN, Devhare PB, Pingle SY, Paingankar MS, Arankalle VA, Lole KS. Hepatitis E virus (HEV)-1 harbouring HEV-4 non-structural protein (ORF1) replicates in transfected porcine kidney cells. J Gen Virol 2016; 97: 1829-1840 [PMID: 27072797 DOI: 10.1099/jgv.0.000478]

91

Feagins AR, Córdoba L, Sanford BJ, Dryman BA, Huang YW, LeRoith T, Emerson SU, Meng XJ. Intergenotypic chimeric hepatitis E viruses (HEVs) with the genotype 4 human HEV capsid gene in the backbone of genotype 3 swine HEV are infectious in pigs. Virus Res 2011; 156: 141-146 [PMID: 21195119 DOI: 10.1016/j.virusres.2010.12.011]

92

Tian D, Yugo DM, Kenney SP, Lynn Heffron C, Opriessnig T, Karuppannan AK, Bayne J, Halbur PG, Meng XJ. Dissecting the potential role of hepatitis E virus ORF1 nonstructural gene in cross-species infection by using intergenotypic chimeric viruses. J Med Virol 2020 [PMID: 32589758 DOI: 10.1002/jmv.26226]

93

Meng XJ. Expanding Host Range and Cross-Species Infection of Hepatitis E Virus. PLoS Pathog 2016; 12: e1005695 [PMID: 27490119 DOI: 10.1371/journal.ppat.1005695]

94

Córdoba L, Feagins AR, Opriessnig T, Cossaboom CM, Dryman BA, Huang YW, Meng XJ. Rescue of a genotype 4 human hepatitis E virus from cloned cDNA and characterization of intergenotypic chimeric viruses in cultured human liver cells and in pigs. J Gen Virol 2012; 93: 2183-2194 [PMID: 22837416 DOI: 10.1099/vir.0.043711-0]

95

Nguyen HT, Shukla P, Torian U, Faulk K, Emerson SU. Hepatitis E virus genotype 1 infection of swine kidney cells in vitro is inhibited at multiple levels. J Virol 2014; 88: 868-877 [PMID: 24198420 DOI: 10.1128/JVI.02205-13]

96

Li L, Zhang L, Hu Q, Zhao L, Nan Y, Hou G, Chen Y, Han X, Ren X, Zhao Q, Tao H, Sun Z, Zhang G, Wu C, Wang J, Zhou EM. MYH9 Key Amino Acid Residues Identified by the Anti-Idiotypic Antibody to Porcine Reproductive and Respiratory Syndrome Virus Glycoprotein 5 Involve in the Virus Internalization by Porcine Alveolar Macrophages. Viruses 2019; 12 [PMID: 31905776 DOI: 10.3390/v12010040]

97

Zhao D, Liu R, Zhang X, Li F, Wang J, Zhang J, Liu X, Wang L, Wu X, Guan Y, Chen W, Wang X, He X, Bu Z. Replication and virulence in pigs of the first African swine fever virus isolated in China. Emerg Microbes Infect 2019; 8: 438-447 [PMID: 30898043 DOI: 10.1080/22221751.2019.1590128]

98

Wang J, Li J, Wang N, Ji Q, Li M, Nan Y, Zhou EM, Zhang Y, Wu C. The 40 kDa Linear Polyethylenimine Inhibits Porcine Reproductive and Respiratory Syndrome Virus Infection by Blocking Its Attachment to Permissive Cells. Viruses 2019; 11 [PMID: 31546799 DOI: 10.3390/v11090876]

99

Nagashima S, Takahashi M, Kobayashi T; Tanggis, Nishizawa T, Nishiyama T, Primadharsini PP, Okamoto H. Characterization of the Quasi-Enveloped Hepatitis E Virus Particles Released by the Cellular Exosomal Pathway. J Virol 2017; 91 [PMID: 28878075 DOI: 10.1128/JVI.00822-17]

100

Ma H, Song X, Harrison TJ, Li R, Huang G, Zhang H, Kong W, Wang Y. Immunogenicity and efficacy of a bacterially expressed HEV ORF3 peptide, assessed by experimental infection of primates. Arch Virol 2009; 154: 1641-1648 [PMID: 19763777 DOI: 10.1007/s00705-009-0496-4]

101

Syed SF, Sun Y, Du T, Chen Y, Liu B, Wang X, Li H, Nan Y, Zhou EM, Zhao Q. Evaluation of recombinant Chinese avian hepatitis E virus (CaHEV) ORF2 and ORF3 proteins for protection of chickens against CaHEV infection. Vaccine 2017; 35: 3482-3489 [PMID: 28528763 DOI:

102


https://dx.doi.org/10.1080/22221751.2019.1565915


https://dx.doi.org/10.1128/jvi.76.7.3095-3104.2002


https://dx.doi.org/10.1126/science.aaz1439


https://dx.doi.org/10.1016/j.virol.2015.09.012






https://dx.doi.org/10.1007/s00705-008-0179-6




https://dx.doi.org/10.1099/jgv.0.000478




https://dx.doi.org/10.1002/jmv.26226








https://dx.doi.org/10.3390/v12010040


https://dx.doi.org/10.1080/22221751.2019.1590128


https://dx.doi.org/10.3390/v11090876




https://dx.doi.org/10.1007/s00705-009-0496-4




10.1016/j.vaccine.2017.05.030]Zhao Q, Sun YN, Hu SB, Wang XJ, Xiao YH, Hsu WH, Xiao SQ, Wang CB, Mu Y, Hiscox JA, Zhou EM. Characterization of antigenic domains and epitopes in the ORF3 protein of a Chinese isolate of avian hepatitis E virus. Vet Microbiol 2013; 167: 242-249 [PMID: 24021883 DOI: 10.1016/j.vetmic.2013.08.006]

103

Semiletov IuA, Dement'evaEV, Iashina TL, Favorov MO, Shibnev VA. [Synthesis and antigenic activity of peptides from the ORF3 protein sequence of hepatitis E virus]. Bioorg Khim 1995; 21: 156-157 [PMID: 7748209]

104

Dement'eva EV, Semiletov IuA, Iashina TL, Shibnev VA. [Analysis of the organization of immunodominant epitopes of hepatitis virus E orf2 and orf3 proteins using synthetic peptides]. Vopr Virusol 1997; 42: 102-105 [PMID: 9297336]

105

Yang Y, Lin S, Nan Y, Ma Z, Yang L, Zhang Y. A Linear Surface Epitope in a Proline-Rich Region of ORF3 Product of Genotype 1 Hepatitis E Virus. Viruses 2016; 8: 227 [PMID: 27548202 DOI: 10.3390/v8080227]

106

https://dx.doi.org/10.1016/j.vaccine.2017.05.030


https://dx.doi.org/10.1016/j.vetmic.2013.08.006




https://dx.doi.org/10.3390/v8080227


World Journal of



REVIEW

Breakthroughs and challenges in the management of pediatric viral hepatitis

Emanuele Nicastro, Lorenzo Norsa, Angelo Di Giorgio, Giuseppe Indolfi, Lorenzo D'Antiga

ORCID number: Emanuele Nicastro 0000-0002-4518-9375; Lorenzo Norsa 0000-0003-3322-2921; Angelo Di Giorgio 0000-0003-0363-5565; Giuseppe Indolfi 0000-0003-3830-9823; Lorenzo D'Antiga 0000-0001-7150-3148.

Author contributions: Nicastro E, Norsa L, Di Giorgio A and Indolfi G contributed literature review, manuscript draft, manuscript revision; D'Antiga L contributed manuscript revision.

Conflict-of-interest statement: The authors declare that they have no competing interests.

Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/Licenses/by-nc/4.0/


Emanuele Nicastro, Lorenzo Norsa, Angelo Di Giorgio, Lorenzo D'Antiga, Department of Pediatric Gastroenterology Hepatology and Transplantation, ASST Papa Giovanni XXIII, Bergamo 24127, Italy

Giuseppe Indolfi, Department of Neurofarba, Meyer Children's University Hospital of Florence, Florence 50137, Italy

Corresponding author: Emanuele Nicastro, MD, PhD, Consultant, Department of Pediatric Gastroenterology Hepatology and Transplantation, ASST Papa Giovanni XXIII, Piazza OMS 1, Bergamo 24127, Italy. [email protected]

AbstractChronic infections by hepatitis B virus (HBV) and hepatitis C virus (HCV) major causes of advanced liver disease and mortality worldwide. Although regarded as benign infections in children, their persistence through adulthood is undoubtedly of concern. Recent advances in HCV treatment have restored the visibility of these conditions and raised expectations for HBV treatment, which is currently far from being curative. Herein we describe direct-acting antivirals available for pediatric HCV (sofosbuvir/ledipasvir, sofosbuvir/velpatasvir, glecaprevir/pibrentasvir) and their real-world use. A critical review of the HBV pediatric classification is provided. Anti-HBV investigational compounds are reviewed in light of the pathophysiology in the pediatric population, including capsid assembly modulators, antigen secretion inhibitors, silencing RNAs, and immune modifiers. Recommendations for screening and management of immunosuppressed children or those with other risk factors or comorbidities are also summarized.

Key Words: Hepatitis C; Hepatitis B; Direct acting antivirals; Liver cirrhosis; Children


Core Tip: Chronic hepatitis B and C account for substantial morbidity and mortality worldwide. Infection control in children has great clinical and epidemiological implic-ations. This review discusses recent achievements and forthcoming opportunities in the management of pediatric viral hepatitis. Removing barriers to the access to novel direct-acting antivirals, and understanding the relevance of tolerance-breaking new



http://orcid.org/0000-0002-4518-9375

http://orcid.org/0000-0002-4518-9375

http://orcid.org/0000-0003-3322-2921

http://orcid.org/0000-0003-3322-2921

http://orcid.org/0000-0003-0363-5565

http://orcid.org/0000-0003-0363-5565

http://orcid.org/0000-0003-3830-9823

http://orcid.org/0000-0003-3830-9823

http://orcid.org/0000-0001-7150-3148

http://orcid.org/0000-0001-7150-3148

http://orcid.org/0000-0001-7150-3148





Nicastro E et al. Viral hepatitis in children



Country/Territory of origin: Italy

Peer-review report’s scientific quality classificationGrade A (Excellent): A Grade B (Very good): 0 Grade C (Good): 0 Grade D (Fair): 0 Grade E (Poor): 0


P-Reviewer: Yang Y S-Editor: Gao CC L-Editor: Filipodia P-Editor: Ma YJ

approaches to infection control, are major challenges for pediatric hepatologists.

Citation: Nicastro E, Norsa L, Di Giorgio A, Indolfi G, D'Antiga L. Breakthroughs and challenges in the management of pediatric viral hepatitis. World J Gastroenterol 2021; 27(20): 2474-2494URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2474.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2474

INTRODUCTIONThe World Health Organization has estimated that - in 2015 - hepatitis B virus (HBV) and hepatitis C virus (HCV) infections caused 1.34 million deaths worldwide, exceeding the death toll of human immunodeficiency virus (HIV) and malaria and reaching that of tuberculosis[1]. These epidemics now move into the spotlight for the forthcoming innovations in their treatment armamentarium. We are living in the aftermath of the breakthrough of direct-acting antivirals (DAAs) for HCV. Those drugs have radically changed the epidemiology of adult chronic liver disease in developed countries, but they are still poorly accessible in resource-limited settings, and their use in children is just beginning. Improvements in HBV treatment are just around the corner. Notably, most investigational compounds directed against HBV tackle the host immune response towards the virus.

The role of children in this scenario is very important. Treating children chronically infected by HBV and HCV is complementary to prevention of mother-to-child transmission, with the purpose of the epidemiological control. On the clinical side, treatments capable of restoring immune tolerance or exhaustion, and promoting viral clearance, are paramount to prevent advanced liver disease in adulthood. In this review, we identify and discuss the next challenges in treatment of pediatric viral hepatitis. Overcoming the barriers to the availability of treatment for all children is a major objective in HCV infection. Conversely, identifying homogeneous patients phenotypes of chronic infection, selecting timely and appropriate treatment, bearing in mind forthcoming therapeutic opportunities, are keys to meeting the current challenges in the management of pediatric HBV infection.

NATURAL HISTORY OF HCV INFECTION IN CHILDRENHCV infection differs in children and adults in the mode of transmission, rate of clearance, progression of fibrosis, and duration of chronic infection when acquired at birth[2]. Vertical transmission of HCV from mother to child is reported to be the leading cause of pediatric infection worldwide. A recent review found that the risk of vertical HCV infection to the children of HCV antibody-positive and HCV-RNA-positive women was 5.8% [95% confidence interval (CI), 4.2-7.8] for children of HIV-negative women and 10.8% (95%CI, 7.6-15.2) for children of HIV-positive women[3].

Symptomatic acute HCV is rare in childhood, but can present with lethargy, fever, and myalgia[4]. Approximately 20%-25% of acute HCV infections can be cleared spontaneously. Spontaneous clearance of the virus following vertical transmission is reported in around 20% of infected patients, usually by 4 years of age[5]. Children who do not clear HCV spontaneously develop chronic infection that, is usually asympto-matic in pediatric patients[6]. During the chronic course, transaminase levels may be normal or intermittently elevated. Serum HCVRNA levels may considerably fluctuate as well, but without immediate prognostic relevance[7,8]. Histological findings are usually unremarkable, but the presence of cirrhosis is reported in 1%-4% of patients[7,9,10]. Overall, the risk of severe hepatic complications in pediatric patients is low[7].

Information on childhood to adult progression of liver is lacking, and the proportion of HCV-infected children who develop serious long-term liver disease is not clear[10]. Hepatic fibrosis is reported in less than 2% of pediatric cases, but the percentage is much higher in patients with long-term follow-up, suggesting that the development of fibrosis correlates with age and the duration of infection[11-14]. In a recent study from the United Kingdom including 1049 patients with a chronic hepatitis





C acquired during childhood, one-third developed liver disease, with a median of 33 years after infection. Patients who acquired the infection vertically developed cirrhosis at an earlier age (36 years) compared with patients who acquired HCV in childhood through drug abuse (48 years), blood transfusion (46 years), or with an unknown route of infection (52 years, P < 0.0001). In that cohort, the incidence of hepatocellular carcinoma (HCC) was 5%; 4% required a liver transplant, and death occurred in 3%[11]. Long-term liver complications included cirrhosis, HCC, or requirement for transplantation[11,12,14]. In adults, the natural history of chronic HCV infection is affected by associated medical and social factors including hepatic steatosis, alcohol consumption, malignancy, and viral HIV and HBV coinfection[11,15]. Extrahepatic manifestations of chronic HCV (e.g., positive nonorgan-specific autoantibodies, autoimmune thyroiditis, glomerulonephritis) are rare in children[15,16]. In conclusion, early acquired hepatitis C infection is a clinically and histologically silent condition. Nevertheless, it may become insidious. Although the percentage of children developing liver cirrhosis in adulthood is low, progression beyond the second decade of life is likely. Thus, the main aim of therapeutic interventions in pediatric patients is not the treatment of an ongoing liver disease, but the prevention of progression by early eradication of the infection[17].

THE BREAKTHROUGH OF DIRECT-ACTING ANTIVIRALSIn 2011 boceprevir and telaprevir, two first-generation NS3/4A protease inhibitors, were approved for use in combination with pegylated interferon (peg-IFN) and ribavirin for the treatment of adults with chronic HCV infection[18,19]. The response rates to triple therapy with boceprevir or telaprevir was improved when compared with dual therapy with peg-IFN and ribavirin, but was accompanied by significant side effects. Since then, the European Medicines Agency (EMA) and the United States Food and Drug Administration (FDA) have approved 13 different DAAs and six fixed-dose combinations for the treatment of adults with chronic HCV infection (sofosbuvir/ledipasvir, ombitasvir/paritaprevir/ritonavir, elbasvir/grazoprevir, glecaprevir/pibrentasvir, sofosbuvir/velpatasvir, and sofosbuvir/ velpatasvir/ voxilaprevir)[20]. DAAs are classified into several categories by their molecular targets. Those agents were designed to inhibit specific viral proteins that have critical roles in HCV replication. They include NS3/4A protease inhibitors (e.g., simeprevir, paritaprevir, grazoprevir, voxilaprevir, and glecaprevir), nucleotide (e.g., sofosbuvir) and non-nucleotide (e.g., dasabuvir) inhibitors of NS5B polymerase, and NS5A inhibitors (e.g., daclatasvir, ledipasvir, ombitasvir, velpatasvir, elbasvir, and pibrentasvir). The development of combinations of DAAs is based on the concept that at least two drugs are needed to achieve the treatment goal of a virological response of > 95%) without selecting resistant mutants. When the backbone of the treatment is a nucleoside NS5B inhibitor like sofosbuvir, only one other drug, an NS3/4A protease inhibitor or a NS5A-inhibitor, is usually required. Conversely, a non-nucleoside NS5B inhibitor like dasabuvir should be used together with both NS3/4A proteases and NS5A inhibitors.

The introduction of DAAs changed the HCV treatment paradigm. These regimens are oral, patient-friendly, have treatment schedules as short as 8 wk, are highly effective, and have few side effects. Given these clear advantages over IFN-based therapies, DAAs have become the preferred treatment for adults with HCV. Remarkably, the latest generation of DAAs (glecaprevir/pibrentasvir, sofosbuvir/ velpatasvir, and sofosbuvir/velpatasvir/voxilaprevir) have pan-genotypic activity, thus simplifying treatment decisions.

The aims of treating HCV with DAAs are multiple. The first is to prevent the complications of HCV-related liver disease, namely cirrhosis, liver failure, and HCC. Secondly, the treatment can improve the quality of life of patients with chronic infection and remove the social stigma of being HCV-positive. Finally, treatment effective for HCV elimination also prevents further transmission of the infection, which targets the goal of global HCV elimination. The excellent efficacy and safety profile of DAAs, have also made antiviral therapy possible for patients with advanced liver disease and for those on the waiting list for liver transplantation (LT), leading to remarkable clinical improvement, avoiding the infection of the graft, and even allowing the delisting of up to one third of the patients[21]. Unfortunately, the initial lack of clinical data specific to DAA treatment in adolescents and children with chronic HCV infection has delayed the introduction of those agents for pediatric use. Several recent studies have demonstrated the safety and efficacy of various DAA-based



regimens in pediatric patients[22], supporting their use in that setting (Table 1).

TREATING CHILDREN WITH HCVIFN-based treatment regimens are no longer recommended as therapeutic options in adolescents and children with HCV, given the high toxicity and modest sustained virological response (SVR) rates[4,23,24]. In 2017, the United States. FDA and the EMA approved the use of sofosbuvir/ledipasvir for adolescents 12 years of age or older with HCV genotype 1, 4, 5, or 6 infection, and sofosbuvir in combination with ribavirin for those with HCV genotype 2 or 3. Since then, four different regimens with age-specific limitations have been approved based on the results of four phase II-III, open-label, multicenter, multicohort studies (Table 2).

Sofosbuvir/ledipasvir was tested in 100 adolescents 12-17 years of age, 92 children 6-11 years of age, and 34 toddlers 3-5 years of age with chronic HCV infections[25-27]. The efficacy of the combination across the three different age cohorts was high (Table 1). One adolescent discontinued treatment and one did not attend post-treatment follow-up visits after having achieved the end-of-treatment response. One cirrhotic child 6 years of age with HCV genotype 1a infection relapsed by the 4-wk follow-up visit, and one 3-year-old patient discontinued treatment after 5 days because of “abnormal drug taste” and vomiting. The treatment was well tolerated. No serious or grade 3-4 drug-related adverse events were reported, and no patient discontinued treatment because of an adverse event. The efficacy of this combination has been confirmed in real-world studies[28-32].

The combination of sofosbuvir and ribavirin was tested in 52 adolescents, 41 children 6-11 years of age and 13 toddlers 3-5 years of age with chronic HCV infection, showing high efficacy (Table 1)[33,34]. One adolescent with HCV genotype 3 infection who did not achieve SVR 12 wk after the end of treatment (SVR12), was lost to follow-up after achieving the end-of-treatment response and SVR4 (HCV-RNA negative 4 wk after the end of treatment). In the younger cohorts, a 4-year-old child who did not achieve the primary endpoint of SVR12 discontinued treatment after 3 d because of an “abnormal drug taste”. No patients had a virologic nonresponse, breakthrough, or relapse. Treatment was well tolerated. One serious adverse event, accidental ribavirin overdose requiring hospitalization for monitoring was reported in a 3-year-old patient. The child completed treatment and achieved SVR12.

A sofosbuvir/velpatasvir fixed-dose formulation was tested in 102 adolescents and 73 children 6-11 years of age with chronic HCV infection[35]. Two treatment failures were reported, a 17-year-old girl with genotype 1a who became pregnant and discon-tinued therapy at week 4, and a 10-year-old girl with genotype 1a. Eight patients were lost to follow-up. The treatment was well tolerated. No serious or grade 3-4 drug-related adverse events were reported, and no patient discontinued treatment because of an adverse event. The combination of glecaprevir/pibrentasvir was tested in 47 adolescents[36], and an SVR12 of 100% was reported. No patients had virologic nonresponse, breakthrough infection, or relapse. The treatment was well tolerated. No serious adverse events were reported, and no patient discontinued treatment because of an adverse event.

Sofosbuvir/ledipasvir and sofosbuvir plus ribavirin are approved for children as young as 3 years of age. Sofosbuvir/velpatasvir and glecaprevir/pibrentasvir are approved for those older than 6 years older than 12 years of age, respectively. The results of the trials of those regimens for younger age cohorts (3-5 years of age for sofosbuvir/velpatasvir and 3-11 years of age for glecaprevir/pibrentasvir) are now available[35,37], and expansion of the indication is expected in late 2021. Remarkably, adherence to the treatment regimen seems to be the main factor impacting the effect-iveness of DAA regimens. Noncompliance was responsible of most of the treatment failures across the different trials[22].

REAL-WORLD USE OF DAA IN CHILDRENAfter the approval of DAAs by international regulatory agencies and their inclusion in clinical guidelines in the early 2010’s, other barriers had to be overcome before they actually became available. In the adult setting, delay in private third-party payer authorization was an obstacle to access to DAA[38,39]. In the pediatric setting, the situation is paradoxical. DAA treatment has been approved by the EMA and the FDA, but the drugs are not available in many countries because of a lack of indications



Table 1 Pediatric studies of different combinations of direct-acting antivirals

Drug regimen Age range in yr

Sample size Genotype(s) Number of patients with

SVR12, % Ref.

12-17 100 1 98/100 (98) Balistreri et al[25]

6-11 92 1, 3, 4 91/92 (99) Murray et al[26]

Sofosbuvir/ledipasvir

3-5 34 1, 4 33/34 (97) Schwarz et al[27]

12-17 52 51/52 (98) Wirth et al[33]

6-11 41 41/41 (100) Rosenthal et al[34]

Sofosbuvir plus ribavirin

3-5 13

2, 3

12/13 (92) Rosenthal et al[34]

12-17 102 97/102 (95) Sokal et al[35]

6-11 73 68/73 (93) Sokal et al[35]

Sofosbuvir/velpatasvir

3-5 41

1, 2, 3, 4

34/41 (83) Sokal et al[35]

12-17 47 47/47 (100) Jonas et al[36]

6-11 32 31/32 (97) Jonas et al[37]

Glecaprevir/pibrentasvir

3-5 16

1, 2, 3, 4

15/16 (96) Jonas et al[37]

12-17 38 1, 4 38/38 (100) Leung et al[134]Ombitasvir/paritaprevir/ritonavir ± dasabuvir ± ribavirin

6-11 26 1 25/26 (95) Rosenthal et al[135]

12-17 22 22/22 (100) Wirth et al[136]

6-11 17 17/17 (100) Wirth et al[136]

Elbasvir/grazoprevir

3-5 18

1, 4

18/18 (100) Wirth et al[136]

Sofosbuvir/velpatasvir/voxilaprevir 12-17 21 1, 2, 3, 4 21/21 (100) Bansal et al[137]

15-17 13 13/13 (100) El-Sayed et al[138]

13-17 10 10/10 (100) El-Shabrawi et al[139]

12-17 30

4

29/29 (100) Yakoot et al[140]

8-17 40 1, 4 39/39 (100) Abdel Ghaffar et al[141]

Sofosbuvir plus daclatasvir

7-13 14 3 14/14 (100) Padhi et al[57]

SVR12: Sustained virological response 12 wk after the end of treatment.

approved by the national agencies[40]. Nevertheless, eradicating HCV infection in early life has been demonstrated to impact favorably on the quality of life and to be cost-effective[41,42]. In fact, reports of real-world experience with the pediatric use of DAAs are increasing, and have shown that the excellent results of the experimental trials can be replicated in adolescents. A study in a cohort of 78 Italian children 12-17 years of age with chronic genotype 1, 3, and 4 HCV infections treated with sofosbuvir/ledipasvir had an SVR12 of 98.7%[28]. Similar results were obtained in three cohorts of Egyptian children. Thirty adolescents with genotype 2 and 144 with genotype 3 infections achieved 99%-100% SVR12s[31,32], and 20 children 6-12 years of age had an SVR12 of 95% after fixed half-dose sofosbuvir/ledipasvir, with only one patient lost to follow-up[43]. Real-world experience has also confirmed that children are ideal candidates for a shortened DAA course, as they typically exhibit little or no fibrosis. In fact, noncirrhotic, treatment-naïve children with genotype 1 and 4 infections and low viremia below 6.8 log10 IU/mL HCV-RNA have been reported to achieve 100% SVR12 after only 8 wk of sofosbuvir/ledipasvir[29,44].

HCV TREATMENT OF HCV IN CHILDREN AT RISKThe advent of DAA has radically changed the approach to HCV eradication treatment



Table 2 Direct-acting antiviral regimens approved for treatment of chronic hepatitis C virus infection in children and adolescents

Regimen Genotype and duration of treatment Formulations Dosage

GT 1, 4, 5, 6: 12 wk > 35 kg: 400/90 mg/d

17-35 kg: 200/45 mg/d

Tablet (FDC) 400/90 mg

Tablet (FDC) 200/45 mg

Sofosbuvir/ledipasvir

GT 1, treatment-experienced, cirrhosis: 24 wk

Pellets 200/45 mg and 150/33.75

< 17 kg, older than 3 yr of age: 150/33.75 mg/d

GT 2: 12 wk Sofosbuvir: > 35 kg: 400 mg/dSofosbuvir: tablet 400 mg

17-35 kg: 200 mg/d

Tablet 100 mg < 17 kg, older than 3 yr of age: 200 mg if ≥ 17 kg

Sofosbuvir + ribavirin

GT 3: 24 wk

Capsules 50 mg containing granules

150 mg/d if < 17 kg. Ribavirin: 15 mg/kg per day in two divided doses

> 30 kg: 400/100 mg/dTablet (FDC) 400/100 mgSofosbuvir/velpatasvir All GTs: 12 wkdecompensated cirrhosis: 12 wk + ribavirin

Tablet (FDC) 200/50 mg 17-30 kg, older than 6 yr of age: 200/50 mg/d. Ribavirin: 15 mg/kg per day in two divided doses

All GTs: 8 wk

All GTs, cirrhosis: 12 wk

Glecaprevir/pibrentasvir

GT 3 treatment-experienced: 16 wk

Tablet (FDC) 100/40 mg/d

12-17 yr or > 45 kg: 300/120 mg/d

FDC: Fixed-dose combination; GT: Genotype.

not only for otherwise healthy children but also for those with comorbidities. The high safety profile of DAA fostered wide experimental use for previously neglected clinical conditions for which IFN-based treatment strategy was proscribed[45]. HCV recurrence after LT is a well-known problem in the adult setting, and can progress rapidly to graft cirrhosis and early loss[46]. HCV eradication with DAA before and after LT brought a tremendous improvement in transplantation outcomes[47-49]. The type of post-LT immunosuppression regimen seems not to impact the DAA treatment response[50], but fluctuation of calcineurin inhibitor trough levels during treatment may account for graft immune-mediated dysfunction[51,52]. Although a similar DAA efficacy might be expected in pediatric LT recipients, no relevant pediatric experience has been published.

So far only seven children receiving DAAs after hemopoietic stem cell transplan-tation (HSCT) for hematological disease have been described in the literature. One had received HSCT for acute lymphoblastic leukemia (4 years of age, genotype 1a)[53], and another for sickle cell disease (15-year-old, genotype 4)[54], and were treated with a combination of sofosbuvir/simeprevir for 24 wk and 12 wk, respectively. They achieved stable viral clearance with calcineurin inhibitor-based regimens (cyclosporin + mycophenolate mofetil, and tacrolimus + sirolimus, respectively). The other five children were 5-12 years of age, and received haploidentical allogeneic HSCT for genotype 1b refractory lymphoblastic leukemia, and were under cyclosporin maintenance[55]. All of them were treated with 24 wk of sofosbuvir/velpatasvir after a median HSCT follow-up of 15 mo, and achieved SVR by 1 mo. El-Shabrawi et al[56] reported on a 12-wk course of sofosbuvir/daclatasvir in 20 genotype 4-infected children in full remission of a hematological malignancy for more than 18 mo. All patients achieved SVR24 without notable adverse events. Two teams of researchers have reported on DAA-driven HCV eradication in thalassemic patients. In total, 25 children were enrolled, 14 with genotype 3 received 12-wk sofosbuvir/daclatasvir and 11 with genotype 4 received 12-wk sofosbuvir/ledipasvir). All achieved SVR12 without complaining of any serious adverse reactions[57,58].

Due to their efficacy and safety profile, DAA use is expanding to cutting-edge therapeutic scenarios. Hematopoietic stem cell-gene therapy (HSC-GT)—a life-saving option for inborn error diseases—is contraindicated in HCV-infected children because of the infection risk of bone marrow cells used as starting material for the manufacture. A 12-wk course with sofosbuvir/ledipasvir was found to allow autologous HSC-GT for the correction of severe combined immunodeficiency caused



by adenosine deaminase deficiency as described in a pioneering study[59]. Pediatric studies have not yet replicated the encouraging results with DAA in adults with HIV/HCV coinfection[60-62], even when vertically transmitted[63], nor for the treatment of HCV in hemodialyzed[64] and kidney transplant patients[65,66].

NATURAL HISTORY OF HBV IN CHILDRENSimilar to HCV, childhood HBV infections occur mostly by mother-to-child transmission. In highly endemic countries, up to one third of the HBV incident transmissions are horizontal, caused by child-to-child, household or intrafamilial contacts[67]. Unlike adults, the prevalence of HBV infections in children reflects the extent of infant and HBV vaccination at birth, which is done with the goal of eradication in the absence of an effective antiviral treatment[68]. According to the last World Health Organization hepatitis report, global hepatitis B surface antigen (HBsAg) prevalence in preschool children dropped from approximately 4.7% between the 1980 and the early 2000s to 1.3% in 2015, following the widespread adoption of universal infant immunization[69]. The age at transmission determines the development of chronic HBV infection, which occurs in approximately 90% of infected newborns and infants compared with only 10% of infections acquired later in childhood or in adults[70,71]. Once chronic infection is established, the subsequent disease progression is the result of the adaptive immune response to HBV, immune-inflammatory liver injury, and the pathogen itself.

Until recently, patients have been classified by immune activity and control criteria, assuming that the clinical phenotype of HBV infection represented a clear-cut immunological state. Due to the lack of immunological data, the latest classification by the European Association for the Study of the Liver (EASL) replaces states with “phases”, which focused more on the infection course in adults, but do not fit the pediatric scenario[72]. In fact, the natural history of chronic HBV infection across all ages has been described by few large, long-term, prospective studies[73-80]. After infection, the vast majority of children enter a state characterized by detectable HBsAg and hepatitis B e antigen (HBeAg), high viremia, and normal or near-normal transa-minase levels. This “tolerant” phase can last for decades, and Asian children or those carrying genotypes C or D have a higher likelihood to remain so[81-83]. Eventually, at a median age of 30 years, 3%-6% of the patients achieve HBeAg/hepatitis B e antibody (HBeAb) seroconversion annually, and most of those patients succeed in suppressing viremia[73,75,81,82]. Spontaneous HBsAg/hepatitis B surface antibody (HBsAb) seroconversion, which potentially allows long-term infection control, rarely occurs, with a rate of approximately 0.5%/year[73,75].

The major concern in children with chronic pediatric HBV infection is the risk of cirrhosis and HCC, which occur in 1%-5% and 0.03%-2% of the patients, respectively[73,74]. As this risk increases in parallel with the activity and the duration of chronic hepatitis B (CHB) (e.g., transaminase elevation in the presence of elevated viral load)[84], the ultimate long-term goal is the achievement of sustained viral suppression, either as a result of the immunologic control or with the use of antiviral drugs.

PHENOTYPE CHARACTERIZATION OF CHILDREN WITH CHRONIC HBV INFECTIONThe current classification of chronic HBV infection recognizes different phases and considers their interchangeability[72]. The phases are identified by the presence or absence of HBeAg (i.e. HBeAg positive and HBeAg negative) and by increased transa-minases, which distinguishes chronic hepatitis from chronic infection. These categories do not fit the pediatric scenario well, where HBeAg-positive chronic infection tends to persist and HBeAg-negative hepatitis is anecdotal[85].

Classifying patients with HBV infection by the clinical phenotype, which reflects the immune status against the virus, has some advantages. Phenotype classification has been progressively refined to harmonize population data, estimate the need for antiviral treatment, and possibly predict the outcome of the infection[81,82]. It describes three childhood phenotypes: (1) immune-active (HBeAg +/-, elevated transaminases, HBV-DNA > log10 4 IU/mL); (2) immune-tolerant (HBeAg-positive, normal transaminases, HBV-DNA > log10 4 IU/mL); and (3) inactive carriers (HBeAg



+/-, normal transaminases, HBV-DNA ≤ log10 4 IU/mL); with the remaining being “indeterminant”[81-83]. So defined, the immune phenotype may help predict the fate towards infection control. Indeed, HBeAg-positive immune-active children are much more likely to achieve spontaneous HBeAg/HBeAb seroconversion (28% and 7.47%/year) compared with immune-tolerant children (11% and 2.29%/year)[81]. In a scenario characterized by the lack of clear treatment indications, and disappointing therapeutic outcomes, the definition of phenotypes has proven to bear prognostic value, and to help define homogeneous groups for patient selection in future trials.

CONTROVERSIES IN THE TREATMENT OF CHILDREN WITH HBVThe current pediatric guidelines recommend limiting treatment to children with prolonged (> 6 mo) active hepatitis B and evidence of fibrosis, and to observe those with the immune-tolerant phenotype[86]. The rationale for this approach, is that HBeAg-positive CHB, especially when protracted and beginning early, at < 3 years of age, has the highest risk of progression to cirrhosis, which is between 1% and 5% by the third decade[73,74]. Many centers tend to delay treatment based on the sound conception that CHB is harmless and that transaminase activity heralds spontaneous immune clearance[73]. Moreover, there is evidence that treatment only accelerates HBeAg/HBeAb seroconversion without influencing the proportion of patients who seroconvert over time[75]. On the other hand, treatment could be indicated to break tolerance. The EASL recommends nucleoside or nucleotide analogues (NAs) for long-term viral suppression in immune-tolerant patients 30 years of age or older[72], to reduce the risk of fibrosis and cirrhosis in those with delayed immune clearance [87,88]. In fact, studies highlight that high viremia in persistently HBeAg-positive patients is associated with the highest risk of cirrhosis, HCC, and liver-related mortality and that functional infection control following HBeAg/HBeAb serocon-version lowers those risks[87,89,90]. Thus, the question has arisen whether treatment should also be offered to immune-tolerant children.

The pediatric age has been regarded as a good time window to break viral tolerance. The first encouraging experience of treating immune-tolerant children was published in 2006. HBV-DNA clearance and HBsAg/HBsAb seroconversion were achieved in 5 of 22 children (23%) after 10 mo of sequential combination therapy with lamivudine and IFN-a[91]. That result was confirmed in a case-control study in which 6 of 22 immune-tolerant children achieved HBsAb-dependent functional control[92]. Another study reported a success rate of 33%[93]. Those studies reported rates of HBsAb seroconversion of 20%-25%. Conversely, a clinical trial using a combination of entecavir and peg-IFN-a 2a resulted in only 2 of 60 children achieving the primary endpoint of HBeAg loss and sustained HBV-DNA levels < 1000 IU/mL), with adverse events reported in more than 50% of the children[94].

In favor of treating immune-tolerant children is evidence that their spontaneous seroconversion rate is much lower than that seen in adults[87]. Also, immune-tolerant children treated with a combination of lamivudine and IFN achieved a high rate of loss of HBsAg. Moreover, unlike adults, immune tolerance in young patients is charac-terized by a high level of HBV DNA integration and clonal hepatocyte expansion with malignant potential despite a relatively preserved anti-HBV T-cell response[95]. Several new compounds in the pipeline that aim at increasing immune responses and overcoming immune exhaustion, with probably play a role in this specific group of children.

DRUGS APPROVED FOR CHILDREN WITH CHBSeven different drugs have been approved for children and adolescents with chronic HBV infection by the United States FDA and the EMA (Table 3). IFN- and peg-IFN- are immune modulators that can be administered for a predefined duration with the aim of inducing immune-mediated control of HBV infection and achieving long-lasting suppression of off-treatment viral replication[68]. NA are potent HBV inhibitors that are used as long-term oral treatments to suppress viral replication, or as treatments of finite duration with or without IFN, to obtain a sustained off-treatment virological response. NAs are also classified as genetic barriers to resistance by the threshold of mutations required for clinically meaningful loss of drug susceptibility. Lamivudine, adefovir, and telbivudine have low and tenofovir, and entecavir have high, genetic barriers to resistance[68]. Treatment is indicated for children and adolescents with



Table 3 Antiviral drugs approved for adolescents and children with chronic hepatitis B virus infection

Drug Use in children Dose Formulation

Interferon-α-2b ≥ 1 yr 6 million IU/m2 3 times a week Subcutaneous injections

Pegylated interferon-α-2a ≥ 3 yr 180 µg/1.73 m2 once a week Subcutaneous injections

Oral solution (5 mg/mL)Lamivudine ≥ 3 yr 3 mg/kg daily (max 100 mg)

tablets (100 mg)

Oral solution (0.05 mg/mL)Entecavir ≥ 2 yr 10-30 kg: 0.015 mg/kg daily (max 0·5 mg)

tablets (0.5 mg and 1 mg)

Adefovir ≥ 12 yr 10 mg daily Tablets (10 mg)

≥ 12 yr (FDA) Oral powder (40 mg per 1 g)Tenofovir disoproxil fumarate

≥ 2 yr (EMA)

300 mg daily

tablets (150, 200, 250 and 300 mg)

Tenofovir alafenamide ≥ 12 yr (EMA) 25 mg daily Tablets (25 mg)

EMA: European Medicines Agency; FDA: Food and Drug Administration.

active viral replication (detectable HBV-DNA levels), prolonged (6-12 mo) active hepatitis (elevated serum transaminase levels) and/or inflammation or fibrosis on liver biopsy. Unless within a clinical trial, treatment is contraindicated when transam-inases are normal[72,86,96].

IFN--2b, peg-IFN--2a, lamivudine, adefovir, tenofovir disoproxil fumarate, and entecavir were approved for the treatment of children and adolescents with chronic HBV infection following six randomized placebo-controlled trials[97-102]. Tenofovir alafenamide was approved on the basis of studies of its use in HIV infection (Table 4). IFN therapy may be associated with higher rates of HBsAg loss compared with NAs. In all studies, a good overall treatment response defined by reduction of serum HBV-DNA to undetectable concentrations, loss of serum HBeAg, and normalization of transaminases, was associated with and increased baseline histology activity index score, increased baseline transaminase concentrations, and decreased baseline HBV-DNA concentrations.

Peg-IFN, entecavir, and tenofovir disoproxil fumarate are considered the drugs of choice for the treatment of CHB in children by the major international societies [72,86,96]. The advantages of IFN and peg-IFN use in children, compared with NAs, are the absence of viral resistance and a predictable, finite duration of treatment. However, the use of IFN and peg-IFN is demanding for children because it requires subcutaneous injection and is associated with a nearly certain occurrence of flu-like symptoms.

DRUGS IN THE PIPELINE: RELEVANCE TO PEDIATRIC AGECurrently available drugs against HBV have inherent limitations. NAs are safe and well tolerated and usually succeed in suppressing replication. They are not curative, as they act at a late stage of the viral cycle and do not prevent HBV-DNA persistence in episomal or integrated forms[96]. On the other hand, IFNs have limited efficacy and frequent side effects, but they are the most likely to achieve a definite “functional” cure with HBeAg or HBsAg loss, seroconversion, normal transaminases, and undetectable viremia)[103]. Various new classes of compounds are under investigation with the aim of achieving high rates of HBsAg seroconversion[104]. The candidate drugs are relevant to the pediatric field, as almost all act as immune modifiers to achieve tolerance breakthrough. The molecules currently in phase II and later trials are listed and summarized in Table 5.

Capsid assembly inhibitors are antiviral agents complementary to NAs. JNJ-56136379 is an oral compound with two distinct mechanisms, the inhibition of the encapsidation of the pregenomic RNA (pgRNA) and the formation of covalently closed circular (ccc)DNA. It has been studied in 57 subjects with CHB treated for 28 d. HBV-DNA and HBV-RNA decreased at all tested doses and HBV-DNA was undetectable at the end of the study in one third of the patients. Nonetheless, none achieved HBsAg/HBsAb seroconversion[105]. ABI-H0731 (Vebicorvir) is an oral



Table 4 Summary of results of clinical trials of hepatitis B antiviral therapy in children

Interferon-α-2b Lamivudine Adefovir Tenofovir

DF Entecavir Pegylated interferon-α-2a

Number treated 144 191 173 52 120 101

Duration of treatment in wk 24 52 48 72 48 48

Age, median (range) 5 (1-17) 9 (2-17) 11 15.5 (12-17) 12 (2-17) 11 (3-7)

Virological response as HBeAg negative HBV DNA undetectable (% treated vs placebo)

26 (vs 11) 23 (vs 13) 10.6 (vs 0) 21.2 (vs 0) 24.2 (vs 3.3) 19.8 (vs 2)

HBsAg negative (% treated vs placebo) 10 (vs 1) 2 (vs 0) 0.8 (vs 0) 1.9 (vs 0) 5.8 (vs 0) 8.9 (0)

Ref. Sokal et al[97] Jonas et al[98]

Jonas et al[99]

Murray et al[100]

Jonas et al[101]

Wirth et al[102]

HBeAg: Hepatitis B e antigen; HBsAg: Hepatitis B surface antigen.

Table 5 Investigational drugs for chronic hepatitis B infection

Class Compound Mechanism Known side effects Route Status Ref.

Entry inhibitors

Bulevirtide Blocks entry via interaction with NTCP

Injection site reactions, cholestasis

Subcutaneous (iv)

Approved for CHD (HBV/HDV) Phase II for HBeAg-CHB

European Association for the Study of the Liver[117] and Bogomolov et al[118]

Capsid assembly modulators

JNJ-56136379 Prevents encapsidation of pgRNA Prevents formation of cccDNA

Hypertransaminasemia Oral Phase II for CHB ± NA

Zoulim et al[105]

Capsid assembly modulators

ABI-H0731 (Vebicorvir)

Prevents packaging of pgRNA into capsids

Skin rash, dizziness Oral Phase II for CHB ± NA

Yuen et al[106]

HBsAg secretion inhibitors

REP-2139 Inhibits the secretion of HBsAg subviral particles

Fever, chills, hypertransaminasemia, leukopenia

iv Phase II for CHB + NA Phase II for CHD + peg-IFN-α

Bazinet et al[109,110]

HBsAg secretion inhibitors

REP-2165 Inhibits the secretion of HBsAg subviral particles

Fever, chills, hypertransaminasemia, leukopenia

iv Phase II for CHD + peg-IFN-α

Bazinet et al[110]

RNA interference

JNJ-3989 Silences the mRNA viral transcripts reducing HBsAg

- iv Phase II for CHB Yuen et al[111] and Gane et al[112]

RNA interference

VIR-2218 Silences the mRNA viral transcripts reducing HBsAg

- iv Phase II for CHB VIR Biotechnology[113]

RIG-1 agonist Inarigivir (SB9200)

Activates PRR RIG-1 and IFN-I response

Headache, dizziness, UTI, ILI, GI symptoms, hypertransaminasemia

Oral Phase II for CHB + NA

Yuen et al[114]

Immune modifier

Selgantolimod (GS-9688)

TLR8 agonist Nausea, vomiting, chills, headache, UTI

Oral Phase II for CHB + NA

Gane et al[116]

CHB: Chronic hepatitis B; CHD: Chronic hepatitis D; GI: Gastrointestinal.; HBeAg: Hepatitis B e antigen; HBsAg: Hepatitis B surface antigen; ILI: Influenza-like illness; NA: Nucleotide analogue; peg-IFN-: Pegylated interferon α; UTI: Urinary tract infection.

compound inhibiting encapsidation, binding the core protein, and thus blocking the packaging of pgRNA into nucleocapsids. A phase I study conducted in healthy volunteers and 38 patients with CHB reported good tolerability and a prompt but temporary drop in viremia[106]. Two phase II studies are ongoing, whose interim results show that in already NA-suppressed patients, the addition of Vebicorvir significantly suppressed HBV-RNA levels. In treatment-naïve patients, its association with standard care resulted in a greater decrease in HBV DNA levels[107].

Nucleic acid polymeric secretion inhibitors reduce the release of HBsAg small viral particles, considered crucial in immune system exhaustion, thus favoring HBsAg loss



and the seroconversion to HBsAb. The polymer REP-2139, administered intravenously once weekly, has been selected for its tolerability within this class[108]. The sequential use of REP-2139 and peg-IFN-a in chronic HBV/HDV coinfection over a period of 63 wk, resulted in sustained HBsAg loss and seroconversion to HBsAb in six of 12 patients. HBV-DNA and HDV-RNA were negative in seven of the patients at the end of the treatment and in nine after 1 year of follow-up[109]. In HBeAg-negative CHB, REP-2139 or its analogue REP-2165 were used in combination with tenofovir and peg-IFN-a and achieved sustained HBsAg/HBsAb seroconversion in 41% of the patients and functional control (undetectable HBV-DNA and normal transaminases) in 77%, an unprecedented result[110].

RNA interference is another promising strategy that aims at silencing the translation of viral transcripts and subsequently decreasing HBsAg. Preliminary reports of efficacy are available from ongoing phase I/II studies on CHB entailing 3 moly administration of the small interfering (si)RNA JNJ-3989 (ARO-HBV). Regardless of HBeAg status and previous treatment, HBsAg decreased by 97%-100% after one dose and the majority of participants achieved HBsAg loss and dramatically reduced HBV-DNA shortly after protocol completion[111,112]. This drug aims at breaking the immune stall toward the virus, as suggested by an ongoing trial in immune-tolerant adults. Similar interim results have been reported in a trial of the siRNA VIR-2218[113].

Immune modulators are a heterogeneous class of candidate antivirals that target different effectors of innate immunity. Inarigivir is an RIG-1 pattern recognition receptor agonist, whose binding activates the IFN-I response. Final results of the phase II ACHIEVE trial demonstrated dose-dependent HBV-DNA reduction after inarigivir monotherapy, and the endpoint of an HBsAg reduction > 0.5 log10 was achieved in 22% of the patients[114]. Selgantolimod (formerly GS-9688), is a potent, orally administered Toll-like-receptor (TLR8) agonist capable of inducing tumor necrosis factor-, IFN-γ, interleukin (IL)-12, and IL-18 expression[115]. Interim results of a phase II study show that it induced a significant HBsAg reduction in 16%-30% of CHB patients and occasional HBsAg loss 24 wk after treatment onset[116]. Bulevirtide is the only viral-entry inhibitor approved by the EMA in 2020 for HBV/HDV coinfection, while it is on a phase II study for HBeAg-negative CHB (NCT02881008). It binds sodium taurocholate cotransporting polypeptide (NTCP) to prevent HBV from entering hepatocytes. Combined with peg-IFN-a, it was shown to significantly decrease HBV-DNA and HDV-DNA compared with peg-IFN-a alone[117,118]. Figure 1 depicts the different mechanisms of action of the HBV investigational products. Although many steps remain to achieve availability in the pediatric population, these drugs will certainly change the burden of HBV in children as much as in adults. The perspective of feasible and curative treatments aiming to break tolerance will increase the efforts to eradicate HBV early in life.

SCREENING AND TREATMENT OF HBV IN CHILDREN AT RISKData on the risk of hepatitis B reactivation in children who need to start immunosup-pressive treatment for concomitant diseases are extremely scarce. For that reason, the recommendations expressed in a recent position paper of the Hepatology committee of the ESPGHAN[119] are mostly derived from adult evidence[72,96,120]. Experts recommend HBV screening, with HBsAg, HBsAb, and hepatitis B core antibody (HBcAb) testing, of all patients at risk of HBV reactivation, including those who are going to start immunosuppressive treatment. The tests should be performed even if HBV vaccination is complete, because, as shown in inflammatory bowel diseases, immunosuppressed children have low serologic protection from childhood vaccines[121]. Patients with negative HBV screening should be vaccinated before starting immunosuppressive treatment. This statement was supported by a recent study of 580 children that demonstrated a high seroconversion rate after the vaccine booster even after immunosuppression initiation[122].

Reactivation risk is classified as mild, moderate, or severe depending on the administered immunosuppressive agents[119]. The risk classes are based exclusively on adult evidence, as no corresponding pediatric study results have been published. Children scheduled for moderate or high-risk drugs should start antiviral prophylaxis, while a preemptive approach is preferred for children starting on low-risk drugs[123,124]. Even in the absence of pediatric evidence, a robust network meta-analysis reported that entecavir or tenofovir should be preferred for HBV reactivation treatment in immunosuppressed patients[125].



Figure 1 Novel investigational approaches to chronic hepatitis. cccDNA: Covalently closed circular DNA; DNA-: Negative-sense DNA filament synthesis; DNA+: Positive-sense DNA filament synthesis; EGFR: Epidermal growth factor receptor; NTCP: Sodium taurocholate cotransporting polypeptide; HBcAg: Hepatitis B core antigen; HBeAg: Hepatitis B e antigen; HBsAg: Hepatitis B surface antigen; NAPs: Nucleic acid polymers; pgRNA: Pregenomic RNA; preC RNA: Pre-core RNA; PRR: Pattern recognition receptor; rcDNA: Relaxed circular DNA; SVP: Small viral particle; TLR: Toll-like receptor.

Currently, reactivation of HBV infection in pediatric solid organ transplant recipients is anecdotal, as HBcAb-positive donors are almost no longer used, and end-stage liver disease in HBsAg-positive recipients is exceptional. However, current recommendations replicate those for adults. In HBsAg-positive recipients, tenofovir or entecavir treatment should be started as soon as possible before transplant to achieve undetectable HBV-DNA[72,96]. NA treatment should be continued indefinitely, and HBV-specific immunoglobulins can be stopped after 5-7 d, unless there is a history of drug resistance or poor compliance[72,96,126]. The use of grafts from HBcAb positive/HBsAg negative donors might be acceptable in case of organ shortage and in highly endemic countries. In those situations, recipients with HBsAb titers > 200 IU/mL might be protected from infection. However, ultimately the overall risk of developing infection depends on multiple factors and implies that recipients of such organs receive NAs (e.g., entecavir, tenofovir, or tenofovir alafenamide) for at least 1 year after transplant. Discontinuation might then be carefully evaluated on a case-by-case basis in HBsAb-positive children[127].

Another challenge for HBV treatment is represented by HBeAg-negative CHB, which is the most prevalent chronic hepatitis in many countries[128]. In adults, the clinical profile is characterized by wide fluctuations in viral replication and biochemical activity, with temporary spontaneous remissions[129]. The risk of cirrhosis in HBeAg-negative CHB is higher (8%-10%/year) than in HBeAg-positive (2%-5%/year)[130,131]. Infants with fulminant hepatitis caused by mother-to-child transmission of HBV e-minus variants are described[132]. This behavior is thought to result from absence of the tolerogenic effect of HBeAg. Adults are treated with long courses of NAs[133], but evidence of the benefits in children with HBeAg-negative hepatitis is lacking. However, pediatric HBeAg-negative and HBeAg-positive CHB have the same therapeutic approach, which is based on IFN- and indefinite use of NAs. The target should be decrease in HBsAg, HBV-DNA clearance, and normal-ization of transaminases.



CONCLUSIONHBV and HCV infections in childhood usually progress to chronic hepatitis through different mechanisms of immune tolerance and exhaustion. For HCV infection, different combinations of DAAs are increasingly available, including pan-genotypic combinations, with very few side effects and extremely high SVRs. For HBV infection, recent cohort studies have clarified that several factors including immune host phenotype, viral genotype, and ethnicity, contribute to spontaneous control. Viral hepatitis should not be a barrier to the use of immunosuppressive medications in the treatment of autoimmune conditions, nor to antineoplastic chemotherapy, provided that timely screening and appropriate pharmacological interventions are performed. Finally, new drugs in development for the treatment of HBV, mostly acting by fostering the breaking of tolerance, could dramatically improve the treatment outcome of CHB in children.

REFERENCESWorld Health Organization. Viral hepatitis: a hidden killer gains visibility. [cited 2 March 2021]. In: World Health Organization [Internet]. Available from: http://www.who.int/publications/10-year-review/hepatitis/en/

1

Robinson JL, Doucette K. The natural history of hepatitis C virus infection acquired during childhood. Liver Int 2012; 32: 258-270 [PMID: 22098487 DOI: 10.1111/j.1478-3231.2011.02633.x]

2

Benova L, Mohamoud YA, Calvert C, Abu-Raddad LJ. Vertical transmission of hepatitis C virus: systematic review and meta-analysis. Clin Infect Dis 2014; 59: 765-773 [PMID: 24928290 DOI: 10.1093/cid/ciu447]

3

Indolfi G, Easterbrook P, Dusheiko G, El-Sayed MH, Jonas MM, Thorne C, Bulterys M, Siberry G, Walsh N, Chang MH, Meyers T, Giaquinto C, Wirth S, Chan PL, Penazzato M. Hepatitis C virus infection in children and adolescents. Lancet Gastroenterol Hepatol 2019; 4: 477-487 [PMID: 30982721 DOI: 10.1016/S2468-1253(19)30046-9]

4

Resti M, Jara P, Hierro L, Azzari C, Giacchino R, Zuin G, Zancan L, Pedditzi S, Bortolotti F. Clinical features and progression of perinatally acquired hepatitis C virus infection. J Med Virol 2003; 70: 373-377 [PMID: 12766999 DOI: 10.1002/jmv.10405]

5

Seeff LB. Natural history of chronic hepatitis C. Hepatology 2002; 36: S35-S46 [PMID: 12407575 DOI: 10.1053/jhep.2002.36806]

6

Bortolotti F, Verucchi G, Cammà C, Cabibbo G, Zancan L, Indolfi G, Giacchino R, Marcellini M, Marazzi MG, Barbera C, Maggiore G, Vajro P, Bartolacci S, Balli F, Maccabruni A, Guido M; Italian Observatory for HCV Infection and Hepatitis C in Children. Long-term course of chronic hepatitis C in children: from viral clearance to end-stage liver disease. Gastroenterology 2008; 134: 1900-1907 [PMID: 18439604 DOI: 10.1053/j.gastro.2008.02.082]

7

Jara P, Resti M, Hierro L, Giacchino R, Barbera C, Zancan L, Crivellaro C, Sokal E, Azzari C, Guido M, Bortolotti F. Chronic hepatitis C virus infection in childhood: clinical patterns and evolution in 224 white children. Clin Infect Dis 2003; 36: 275-280 [PMID: 12539067 DOI: 10.1086/345908]

8

Goodman ZD, Makhlouf HR, Liu L, Balistreri W, Gonzalez-Peralta RP, Haber B, Jonas MM, Mohan P, Molleston JP, Murray KF, Narkewicz MR, Rosenthal P, Smith LJ, Robuck PR, Schwarz KB. Pathology of chronic hepatitis C in children: liver biopsy findings in the Peds-C Trial. Hepatology 2008; 47: 836-843 [PMID: 18167062 DOI: 10.1002/hep.22094]

9

Indolfi G, Guido M, Azzari C, Resti M. Histopathology of hepatitis C in children, a systematic review: implications for treatment. Expert Rev Anti Infect Ther 2015; 13: 1225-1235 [PMID: 26202832 DOI: 10.1586/14787210.2015.1070668]

10

Modin L, Arshad A, Wilkes B, Benselin J, Lloyd C, Irving WL, Kelly DA. Epidemiology and natural history of hepatitis C virus infection among children and young people. J Hepatol 2019; 70: 371-378 [PMID: 30496763 DOI: 10.1016/j.jhep.2018.11.013]

11

Jara P, Hierro L. Treatment of hepatitis C in children. Expert Rev Gastroenterol Hepatol 2010; 4: 51-61 [PMID: 20136589 DOI: 10.1586/egh.09.76]

12

Guido M, Bortolotti F, Leandro G, Jara P, Hierro L, Larrauri J, Barbera C, Giacchino R, Zancan L, Balli F, Crivellaro C, Cristina E, Pucci A, Rugge M. Fibrosis in chronic hepatitis C acquired in infancy: is it only a matter of time? Am J Gastroenterol 2003; 98: 660-663 [PMID: 12650803 DOI: 10.1111/j.1572-0241.2003.07293.x]

13

Abdel-Hady M, Bunn SK, Sira J, Brown RM, Brundler MA, Davies P, Kelly DA. Chronic hepatitis C in children--review of natural history at a National Centre. J Viral Hepat 2011; 18: e535-e540 [PMID: 21914074 DOI: 10.1111/j.1365-2893.2011.01456.x]

14

Garazzino S, Calitri C, Versace A, Alfarano A, Scolfaro C, Bertaina C, Vatrano S, Mignone F, Licciardi F, Gabiano C, Tovo PA. Natural history of vertically acquired HCV infection and associated autoimmune phenomena. Eur J Pediatr 2014; 173: 1025-1031 [PMID: 24585099 DOI: 10.1007/s00431-014-2286-6]

15

http://www.who.int/publications/10-year-review/hepatitis/en/

http://www.who.int/publications/10-year-review/hepatitis/en/




https://dx.doi.org/10.1093/cid/ciu447


https://dx.doi.org/10.1016/S2468-1253(19)30046-9


https://dx.doi.org/10.1002/jmv.10405


https://dx.doi.org/10.1053/jhep.2002.36806




https://dx.doi.org/10.1086/345908




https://dx.doi.org/10.1586/14787210.2015.1070668




https://dx.doi.org/10.1586/egh.09.76






https://dx.doi.org/10.1007/s00431-014-2286-6



Sugiura T, Yamada T, Kimpara Y, Fujita N, Goto K, Koyama N. Effects of pegylated interferon alpha-2a on hepatitis-C-virus-associated glomerulonephritis. Pediatr Nephrol 2009; 24: 199-202 [PMID: 18696119 DOI: 10.1007/s00467-008-0948-1]

16

Indolfi G, Thorne C, El Sayed MH, Giaquinto C, Gonzalez-Peralta RP. The Challenge of Treating Children With Hepatitis C Virus Infection. J Pediatr Gastroenterol Nutr 2017; 64: 851-854 [PMID: 28362694 DOI: 10.1097/MPG.0000000000001589]

17

Sarrazin C, Rouzier R, Wagner F, Forestier N, Larrey D, Gupta SK, Hussain M, Shah A, Cutler D, Zhang J, Zeuzem S. SCH 503034, a novel hepatitis C virus protease inhibitor, plus pegylated interferon alpha-2b for genotype 1 nonresponders. Gastroenterology 2007; 132: 1270-1278 [PMID: 17408662 DOI: 10.1053/j.gastro.2007.01.041]

18

Reesink HW, Zeuzem S, Weegink CJ, Forestier N, van Vliet A, van de Wetering de Rooij J, McNair L, Purdy S, Kauffman R, Alam J, Jansen PL. Rapid decline of viral RNA in hepatitis C patients treated with VX-950: a phase Ib, placebo-controlled, randomized study. Gastroenterology 2006; 131: 997-1002 [PMID: 17030169 DOI: 10.1053/j.gastro.2006.07.013]

19

Jakobsen JC, Nielsen EE, Feinberg J, Katakam KK, Fobian K, Hauser G, Poropat G, Djurisic S, Weiss KH, Bjelakovic M, Bjelakovic G, Klingenberg SL, Liu JP, Nikolova D, Koretz RL, Gluud C. Direct-acting antivirals for chronic hepatitis C. Cochrane Database Syst Rev 2017; 9: CD012143 [PMID: 28922704 DOI: 10.1002/14651858.CD012143.pub3]

20

Belli LS, Berenguer M, Cortesi PA, Strazzabosco M, Rockenschaub SR, Martini S, Morelli C, Donato F, Volpes R, Pageaux GP, Coilly A, Fagiuoli S, Amaddeo G, Perricone G, Vinaixa C, Berlakovich G, Facchetti R, Polak W, Muiesan P, Duvoux C; European Liver and Intestine Association (ELITA). Delisting of liver transplant candidates with chronic hepatitis C after viral eradication: A European study. J Hepatol 2016; 65: 524-531 [PMID: 27212241 DOI: 10.1016/j.jhep.2016.05.010]

21

Indolfi G, Giometto S, Serranti D, Bettiol A, Bigagli E, De Masi S, Lucenteforte E. Systematic review with meta-analysis: the efficacy and safety of direct-acting antivirals in children and adolescents with chronic hepatitis C virus infection. Aliment Pharmacol Ther 2020; 52: 1125-1133 [PMID: 32809230 DOI: 10.1111/apt.16037]

22

Indolfi G, Hierro L, Dezsofi A, Jahnel J, Debray D, Hadzic N, Czubkowski P, Gupte G, Mozer-Glassberg Y, van der Woerd W, Smets F, Verkade HJ, Fischler B. Treatment of Chronic Hepatitis C Virus Infection in Children: A Position Paper by the Hepatology Committee of European Society of Paediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr 2018; 66: 505-515 [PMID: 29287014 DOI: 10.1097/MPG.0000000000001872]

23

Indolfi G, Fischler B, Gonzalez-Peralta RP, Ciocca M, Porta G, Neelam M, El-Guindi M, Kelly D, Ni YH, Sibal A, Leung DH, Chang MH; Hepatitis Expert Team of the Federation of International Societies of Pediatric Gastroenterology, Hepatology, and Nutrition (FISPGHAN). Comparison of Recommendations for Treatment of Chronic Hepatitis C Virus Infection in Children and Adolescents: A Position Paper of the Federation of International Societies of Pediatric Gastroenterology, Hepatology, and Nutrition. J Pediatr Gastroenterol Nutr 2020; 70: 711-717 [PMID: 32205770 DOI: 10.1097/MPG.0000000000002710]

24

Balistreri WF, Murray KF, Rosenthal P, Bansal S, Lin CH, Kersey K, Massetto B, Zhu Y, Kanwar B, German P, Svarovskaia E, Brainard DM, Wen J, Gonzalez-Peralta RP, Jonas MM, Schwarz K. The safety and effectiveness of ledipasvir-sofosbuvir in adolescents 12-17 years old with hepatitis C virus genotype 1 infection. Hepatology 2017; 66: 371-378 [PMID: 27997679 DOI: 10.1002/hep.28995]

25

Murray KF, Balistreri WF, Bansal S, Whitworth S, Evans HM, Gonzalez-Peralta RP, Wen J, Massetto B, Kersey K, Shao J, Garrison KL, Parhy B, Brainard DM, Arnon R, Gillis LA, Jonas MM, Lin CH, Narkewicz MR, Schwarz K, Rosenthal P. Safety and Efficacy of Ledipasvir-Sofosbuvir With or Without Ribavirin for Chronic Hepatitis C in Children Ages 6-11. Hepatology 2018; 68: 2158-2166 [PMID: 30070726 DOI: 10.1002/hep.30123]

26

Schwarz KB, Rosenthal P, Murray KF, Honegger JR, Hardikar W, Hague R, Mittal N, Massetto B, Brainard DM, Hsueh CH, Shao J, Parhy B, Narkewicz MR, Rao GS, Whitworth S, Bansal S, Balistreri WF. Ledipasvir-Sofosbuvir for 12 Weeks in Children 3 to <6 Years Old With Chronic Hepatitis C. Hepatology 2020; 71: 422-430 [PMID: 31220349 DOI: 10.1002/hep.30830]

27

Serranti D, Nebbia G, Cananzi M, Nicastro E, Di Dato F, Nuti F, Garazzino S, Silvestro E, Giacomet V, Forlanini F, Pinon M, Calvo PL, Riva S, Dodi I, Cangelosi AM, Antonucci R, Ricci S, Bartolini E, Mastrangelo G, Trapani S, Lenge M, Gaio P, Vajro P, Iorio R, D'Antiga L, Indolfi G. Efficacy of Sofosbuvir/Ledipasvir in Adolescents With Chronic Hepatitis C Genotypes 1, 3, and 4: A Real-world Study. J Pediatr Gastroenterol Nutr 2021; 72: 95-100 [PMID: 32810039 DOI: 10.1097/MPG.0000000000002900]

28

Serranti D, Dodi I, Nicastro E, Cangelosi AM, Riva S, Ricci S, Bartolini E, Trapani S, Mastrangelo G, Vajro P, D'Antiga L, Resti M, Indolfi G. Shortened 8-Week Course of Sofosbuvir/Ledipasvir Therapy in Adolescents With Chronic Hepatitis C Infection. J Pediatr Gastroenterol Nutr 2019; 69: 595-598 [PMID: 31335836 DOI: 10.1097/MPG.0000000000002449]

29

El-Khayat H, Kamal EM, Yakoot M, Gawad MA, Kamal N, El Shabrawi M, Sameh Y, Haseeb A, Fouad Y, Attia D. Effectiveness of 8-week sofosbuvir/Ledipasvir in the adolescent chronic hepatitis C-infected patients. Eur J Gastroenterol Hepatol 2019; 31: 1004-1009 [PMID: 30676473 DOI: 10.1097/MEG.0000000000001360]

30

El-Khayat HR, Kamal EM, El-Sayed MH, El-Shabrawi M, Ayoub H, RizK A, Maher M, El 31


https://dx.doi.org/10.1007/s00467-008-0948-1


https://dx.doi.org/10.1097/MPG.0000000000001589










https://dx.doi.org/10.1111/apt.16037



















Sheemy RY, Fouad YM, Attia D. The effectiveness and safety of ledipasvir plus sofosbuvir in adolescents with chronic hepatitis C virus genotype 4 infection: a real-world experience. Aliment Pharmacol Ther 2018; 47: 838-844 [PMID: 29349793 DOI: 10.1111/apt.14502]El-Karaksy H, Mogahed EA, Abdullatif H, Ghobrial C, El-Raziky MS, El-Koofy N, El-Shabrawi M, Ghita H, Baroudy S, Okasha S. Sustained Viral Response in Genotype 4 Chronic Hepatitis C Virus-infected Children and Adolescents Treated With Sofosbuvir/Ledipasvir. J Pediatr Gastroenterol Nutr 2018; 67: 626-630 [PMID: 30216203 DOI: 10.1097/MPG.0000000000002101]

32

Wirth S, Rosenthal P, Gonzalez-Peralta RP, Jonas MM, Balistreri WF, Lin CH, Hardikar W, Kersey K, Massetto B, Kanwar B, Brainard DM, Shao J, Svarovskaia E, Kirby B, Arnon R, Murray KF, Schwarz KB. Sofosbuvir and ribavirin in adolescents 12-17 years old with hepatitis C virus genotype 2 or 3 infection. Hepatology 2017; 66: 1102-1110 [PMID: 28543053 DOI: 10.1002/hep.29278]

33

Rosenthal P, Schwarz KB, Gonzalez-Peralta RP, Lin CH, Kelly DA, Nightingale S, Balistreri WF, Bansal S, Jonas MM, Massetto B, Brainard DM, Hsueh CH, Shao J, Parhy B, Davison S, Feiterna-Sperling C, Gillis LA, Indolfi G, Sokal EM, Murray KF, Wirth S. Sofosbuvir and Ribavirin Therapy for Children Aged 3 to <12 Years With Hepatitis C Virus Genotype 2 or 3 Infection. Hepatology 2020; 71: 31-43 [PMID: 31222783 DOI: 10.1002/hep.30821]

34

Sokal EM, Schwarz KB, Rosenthal, Philip, Verucchi G, Lin C-H, Balistreri W, Wen J, Whitworth S, Leung DH, Bansal S, Wikrom K, Mangia A, Rao GS, Shao J, Hsueh C-H, Parhy B, Gaggar A, Kersey K, Narkewicz MR, Gonzalez-Peralta RP, Murray KF, Romero R, Jonas MM. Safety and efficacy of sofosbuvir/velpatasvir for the treatment of chronic hepatitis c infection in adolescents and children aged 3 to 17 years old through 24 weeks post-treatment. Hepatology 2020; 570A

35

Jonas MM, Squires RH, Rhee SM, Lin CW, Bessho K, Feiterna-Sperling C, Hierro L, Kelly D, Ling SC, Strokova T, Del Valle-Segarra A, Lovell S, Liu W, Ng TI, Porcalla A, Gonzalez YS, Burroughs M, Sokal E. Pharmacokinetics, Safety, and Efficacy of Glecaprevir/Pibrentasvir in Adolescents With Chronic Hepatitis C Virus: Part 1 of the DORA Study. Hepatology 2020; 71: 456-462 [PMID: 31254392 DOI: 10.1002/hep.30840]

36

Jonas MM, Lon HK, Rhee SM, Gilmour SM, Gonzalez-Peralta RP, Leung DH, Ling SC, Lobritto S, Narkewicz MR, Sabharwal V, Del Valle-Segarra A, Wen J, Lovell S, Porcalla A, Tripathi R, Sokal EM. Pharmacokinetics of glecaprevir/pibrentasvir in children with chronic hcv infection: interim analysis of part 2 of the Dora study. Hepatology 2019; 934-935

37

Do A, Mittal Y, Liapakis A, Cohen E, Chau H, Bertuccio C, Sapir D, Wright J, Eggers C, Drozd K, Ciarleglio M, Deng Y, Lim JK. Drug Authorization for Sofosbuvir/Ledipasvir (Harvoni) for Chronic HCV Infection in a Real-World Cohort: A New Barrier in the HCV Care Cascade. PLoS One 2015; 10: e0135645 [PMID: 26312999 DOI: 10.1371/journal.pone.0135645]

38

Torabi J, Rocca JP, Ajaimy M, Melvin J, Campbell A, Akalin E, Liriano LE, Azzi Y, Pynadath C, Greenstein SM, Le M, Goldstein DY, Fox AS, Carrero J, Weiss JM, Powell T, Racine AD, Reinus JF, Kinkhabwala MM, Graham JA. Commercial insurance delays direct-acting antiviral treatment for hepatitis C kidney transplantation into uninfected recipients. Transpl Infect Dis 2021; 23: e13449 [PMID: 32810315 DOI: 10.1111/tid.13449]

39

Marascio N, Mazzitelli M, Pavia G, Giancotti A, Barreca GS, Costa C, Pisani V, Greco G, Serapide F, Trecarichi EM, Casalinuovo F, Liberto MC, Matera G, Torti C. Clinical, Virological Characteristics, and Outcomes of Treatment with Sofosbuvir/Ledipasvir in Two Pediatric Patients Infected by HCV Genotype 4. Cells 2019; 8 [PMID: 31060315 DOI: 10.3390/cells8050416]

40

Younossi ZM, Stepanova M, Balistreri W, Schwarz K, Murray KF, Rosenthal P, Bansal S, Hunt S. Health-related Quality of Life in Adolescent Patients With Hepatitis C Genotype 1 Treated With Sofosbuvir and Ledipasvir. J Pediatr Gastroenterol Nutr 2018; 66: 112-116 [PMID: 28957984 DOI: 10.1097/MPG.0000000000001754]

41

Ruggeri M, Romano F, Basile M, Coretti S, Rolli FR, Drago C, Cicchetti A. Cost-Effectiveness Analysis of Early Treatment of Chronic HCV with Sofosbuvir/Velpatasvir in Italy. Appl Health Econ Health Policy 2018; 16: 711-722 [PMID: 30039348 DOI: 10.1007/s40258-018-0410-x]

42

El-Shabrawi MHF, Kamal NM, El-Khayat HR, Kamal EM, AbdElgawad MMAH, Yakoot M. A pilot single arm observational study of sofosbuvir/Ledipasvir (200 + 45 mg) in 6- to 12- year old children. Aliment Pharmacol Ther 2018; 47: 1699-1704 [PMID: 29696674 DOI: 10.1111/apt.14677]

43

Behairy BE, El-Araby HA, El-Guindi MA, Basiouny HM, Fouad OA, Ayoub BA, Marei AM, Sira MM. Safety and Efficacy of 8 Weeks Ledipasvir/Sofosbuvir for Chronic Hepatitis C Genotype 4 in Children Aged 4-10 Years. J Pediatr 2020; 219: 106-110 [PMID: 32008767 DOI: 10.1016/j.jpeds.2019.12.034]

44

Suda G, Ogawa K, Morikawa K, Sakamoto N. Treatment of hepatitis C in special populations. J Gastroenterol 2018; 53: 591-605 [PMID: 29299684 DOI: 10.1007/s00535-017-1427-x]

45

Berenguer M, Schuppan D. Progression of liver fibrosis in post-transplant hepatitis C: mechanisms, assessment and treatment. J Hepatol 2013; 58: 1028-1041 [PMID: 23262248 DOI: 10.1016/j.jhep.2012.12.014]

46

Brown RS Jr, O'Leary JG, Reddy KR, Kuo A, Morelli GJ, Burton JR Jr, Stravitz RT, Durand C, Di Bisceglie AM, Kwo P, Frenette CT, Stewart TG, Nelson DR, Fried MW, Terrault NA; Hepatitis C Therapeutic Registry Research Network Study Group. Interferon-free therapy for genotype 1 hepatitis C in liver transplant recipients: Real-world experience from the hepatitis C therapeutic registry and research network. Liver Transpl 2016; 22: 24-33 [PMID: 26519873 DOI: 10.1002/lt.24366]

47

Agarwal K, Castells L, Müllhaupt B, Rosenberg WMC, McNabb B, Arterburn S, Camus G, 48














https://dx.doi.org/10.1111/tid.13449


https://dx.doi.org/10.3390/cells8050416




https://dx.doi.org/10.1007/s40258-018-0410-x




https://dx.doi.org/10.1016/j.jpeds.2019.12.034









McNally J, Stamm LM, Brainard DM, Mani Subramanian G, Mariño Z, Dufour JF, Forns X. Sofosbuvir/velpatasvir for 12 wk in genotype 1-4 HCV-infected liver transplant recipients. J Hepatol 2018; 69: 603-607 [PMID: 29886154 DOI: 10.1016/j.jhep.2018.05.039]Rupp C, Hippchen T, Neuberger M, Sauer P, Pfeiffenberger J, Stremmel W, Gotthardt DN, Mehrabi A, Weiss KH. Successful combination of direct antiviral agents in liver-transplanted patients with recurrent hepatitis C virus. World J Gastroenterol 2018; 24: 1353-1360 [PMID: 29599610 DOI: 10.3748/wjg.v24.i12.1353]

49

El-Hamamsy M, Montasser IF, Mansy AE, Nabet DE, El-Meteini M. Effect of cyclosporine A vs tacrolimus on the response to antiviral therapy after hepatitis C genotype-4 recurrence post-liver transplantation: A prospective cohort trial. J Clin Pharm Ther 2019; 44: 447-453 [PMID: 30714175 DOI: 10.1111/jcpt.12807]

50

Bixby AL, Fitzgerald L, Leek R, Mellinger J, Sharma P, Tischer S. Impact of direct-acting antivirals for hepatitis C virus therapy on tacrolimus dosing in liver transplant recipients. Transpl Infect Dis 2019; 21: e13078 [PMID: 30884055 DOI: 10.1111/tid.13078]

51

Chan C, Schiano T, Agudelo E, Paul Haydek J, Hoteit M, Laurito MP, Norvell JP, Terrault N, Verna EC, Yang A, Levitsky J. Immune-mediated graft dysfunction in liver transplant recipients with hepatitis C virus treated with direct-acting antiviral therapy. Am J Transplant 2018; 18: 2506-2512 [PMID: 30075485 DOI: 10.1111/ajt.15053]

52

Thomas P, Santiago T, Dallas MH. Treatment of hepatitis C in a pediatric patient using simeprevir and sofosbuvir immediately after an umbilical cord blood transplantation. Bone Marrow Transplant 2016; 51: 735-737 [PMID: 26752145 DOI: 10.1038/bmt.2015.309]

53

Fischler B, Priftakis P, Sundin M. Sofosbuvir and Simeprevir Treatment of a Stem Cell Transplanted Teenager With Chronic Hepatitis C Infection. Pediatr Infect Dis J 2016; 35: 708-710 [PMID: 26928522 DOI: 10.1097/INF.0000000000001122]

54

Jaiswal SR, Bhakuni P, Soni M, Gupta M, Thatai A, Chakrabarti S. Safety and efficacy of Sofosbuvir and Velpatasvir in children with active hepatitis C virus infection undergoing haploidentical transplantation. Transpl Infect Dis 2020; e13490 [PMID: 33037697 DOI: 10.1111/tid.13490]

55

El-Shabrawi MH, Sherief LM, Yakoot M, Kamal NM, Almalky MA, AbdElgawad MM, Mahfouz AA, Helmy S, Kamal EM, Attia D, El-Khayat HR. Effects of dual sofosbuvir/daclatasvir therapy on, chronic hepatitis C infected, survivors of childhood malignancy. World J Clin Cases 2019; 7: 2247-2255 [PMID: 31531319 DOI: 10.12998/wjcc.v7.i16.2247]

56

Padhi S, Maharshi S, Gupta GK, Garg K, Nijhawan S. Efficacy and Safety of Direct Acting Antiviral Therapy for Chronic Hepatitis C in Thalassemic Children. J Pediatr Hematol Oncol 2018; 40: 511-514 [PMID: 29771862 DOI: 10.1097/MPH.0000000000001217]

57

Makhlouf NA, Abdelmalek MO, Ibrahim ME, Abu-Faddan NH, Kheila AE, Mahmoud AA. Ledipasvir/Sofosbuvir in Adolescents With Chronic Hepatitis C Genotype 4 With and Without Hematological Disorders: Virological Efficacy and Impact on Liver Stiffness. J Pediatric Infect Dis Soc 2021; 10: 7-13 [PMID: 32060510 DOI: 10.1093/jpids/piaa006]

58

Tucci F, Calbi V, Barzaghi F, Migliavacca M, Ferrua F, Bernardo ME, Canarutto D, Consiglieri G, Recupero S, Calzatini F, Gabaldo M, Lucano C, Casiraghi M, Darin S, Dionisio F, Marktel S, Cestone E, Finazzi R, Mieli-Vergani G, Boeri E, Appleby J, Abd Elaziz D, Ciceri F, Aiuti A, Cicalese MP. Successful Treatment With Ledipasvir/Sofosbuvir in an Infant With Severe Combined Immunodeficiency Caused by Adenosine Deaminase Deficiency With HCV Allowed Gene Therapy with Strimvelis. Hepatology 2018; 68: 2434-2437 [PMID: 30014500 DOI: 10.1002/hep.30160]

59

Naggie S, Cooper C, Saag M, Workowski K, Ruane P, Towner WJ, Marks K, Luetkemeyer A, Baden RP, Sax PE, Gane E, Santana-Bagur J, Stamm LM, Yang JC, German P, Dvory-Sobol H, Ni L, Pang PS, McHutchison JG, Stedman CA, Morales-Ramirez JO, Bräu N, Jayaweera D, Colson AE, Tebas P, Wong DK, Dieterich D, Sulkowski M; ION-4 Investigators. Ledipasvir and Sofosbuvir for HCV in Patients Coinfected with HIV-1. N Engl J Med 2015; 373: 705-713 [PMID: 26196665 DOI: 10.1056/NEJMoa1501315]

60

Molina JM, Orkin C, Iser DM, Zamora FX, Nelson M, Stephan C, Massetto B, Gaggar A, Ni L, Svarovskaia E, Brainard D, Subramanian GM, McHutchison JG, Puoti M, Rockstroh JK; PHOTON-2 study team. Sofosbuvir plus ribavirin for treatment of hepatitis C virus in patients co-infected with HIV (PHOTON-2): a multicentre, open-label, non-randomised, phase 3 study. Lancet 2015; 385: 1098-1106 [PMID: 25659285 DOI: 10.1016/S0140-6736(14)62483-1]

61

Rockstroh JK, Nelson M, Katlama C, Lalezari J, Mallolas J, Bloch M, Matthews GV, Saag MS, Zamor PJ, Orkin C, Gress J, Klopfer S, Shaughnessy M, Wahl J, Nguyen BY, Barr E, Platt HL, Robertson MN, Sulkowski M. Efficacy and safety of grazoprevir (MK-5172) and elbasvir (MK-8742) in patients with hepatitis C virus and HIV co-infection (C-EDGE CO-INFECTION): a non-randomised, open-label trial. Lancet HIV 2015; 2: e319-e327 [PMID: 26423374 DOI: 10.1016/S2352-3018(15)00114-9]

62

Carrasco I, Sainz T, Frick MA, Jiménez de Ory S, Fortuny C, Burgos J, Montero M, Gavilán C, Falcón MD, Couceiro JA, Bernardino JI, Bisbal O, Guerrero C, Aldámiz-Echevarría MT, Berenguer J, Navarro ML; the Pediatric National AIDS Research Network of Spain (CoRISpe) integrated in the Translational Research Network in Pediatric Infectious Diseases (RITIP). Response to direct-acting antivirals for hepatitis C treatment in vertically HIV/HCV co-infected patients. J Viral Hepat 2020; 27: 955-958 [PMID: 32347645 DOI: 10.1111/jvh.13308]

63

Roth D, Nelson DR, Bruchfeld A, Liapakis A, Silva M, Monsour H Jr, Martin P, Pol S, Londoño 64






https://dx.doi.org/10.1111/jcpt.12807




https://dx.doi.org/10.1111/ajt.15053


https://dx.doi.org/10.1038/bmt.2015.309


https://dx.doi.org/10.1097/INF.0000000000001122




https://dx.doi.org/10.12998/wjcc.v7.i16.2247


https://dx.doi.org/10.1097/MPH.0000000000001217


https://dx.doi.org/10.1093/jpids/piaa006






https://dx.doi.org/10.1016/S0140-6736(14)62483-1


https://dx.doi.org/10.1016/S2352-3018(15)00114-9


https://dx.doi.org/10.1111/jvh.13308



MC, Hassanein T, Zamor PJ, Zuckerman E, Wan S, Jackson B, Nguyen BY, Robertson M, Barr E, Wahl J, Greaves W. Grazoprevir plus elbasvir in treatment-naive and treatment-experienced patients with hepatitis C virus genotype 1 infection and stage 4-5 chronic kidney disease (the C-SURFER study): a combination phase 3 study. Lancet 2015; 386: 1537-1545 [PMID: 26456905 DOI: 10.1016/S0140-6736(15)00349-9]Fernández-Ruiz M, Polanco N, García-Santiago A, Muñoz R, Hernández AM, González E, Mercado VR, Fernández I, Aguado JM, Praga M, Andrés A. Impact of anti-HCV direct antiviral agents on graft function and immunosuppressive drug levels in kidney transplant recipients: a call to attention in the mid-term follow-up in a single-center cohort study. Transpl Int 2018; 31: 887-899 [PMID: 29356211 DOI: 10.1111/tri.13118]

65

González-Corvillo C, Beneyto I, Sánchez-Fructuoso A, Perelló M, Alonso A, Mazuecos A, Jiménez C, Zárraga S, Paul J, Lauzurica R, Hernández D, Guirado L, Franco A, Ruiz JC, Llorente S, Crespo M, Rodríguez-Benot A, de Gracia Guindo MDC, Díaz-Corte C, Gentil MÁ. Direct-acting Antivirals for the Treatment of Kidney Transplant Patients With Chronic Hepatitis C Virus Infection in Spain: A Long-term Prospective Observational Study. Transplant Direct 2019; 5: e510 [PMID: 32095505 DOI: 10.1097/TXD.0000000000000954]

66

Gupta S, Gupta R, Joshi YK, Singh S. Role of horizontal transmission in hepatitis B virus spread among household contacts in north India. Intervirology 2008; 51: 7-13 [PMID: 18309243 DOI: 10.1159/000118790]

67

Indolfi G, Easterbrook P, Dusheiko G, Siberry G, Chang MH, Thorne C, Bulterys M, Chan PL, El-Sayed MH, Giaquinto C, Jonas MM, Meyers T, Walsh N, Wirth S, Penazzato M. Hepatitis B virus infection in children and adolescents. Lancet Gastroenterol Hepatol 2019; 4: 466-476 [PMID: 30982722 DOI: 10.1016/S2468-1253(19)30042-1]

68

World Health Organization. Global hepatitis report, 2017. [cited 2 March 2021]. In: World Health Organization [Internet]. Available from: https://www.who.int/publications-detail-redirect/global-hepatitis-report-2017

69

McMahon BJ, Alward WL, Hall DB, Heyward WL, Bender TR, Francis DP, Maynard JE. Acute hepatitis B virus infection: relation of age to the clinical expression of disease and subsequent development of the carrier state. J Infect Dis 1985; 151: 599-603 [PMID: 3973412 DOI: 10.1093/infdis/151.4.599]

70

Stevens CE, Beasley RP, Tsui J, Lee WC. Vertical transmission of hepatitis B antigen in Taiwan. N Engl J Med 1975; 292: 771-774 [PMID: 1113797 DOI: 10.1056/NEJM197504102921503]

71

European Association for the Study of the Liver. EASL 2017 Clinical Practice Guidelines on the management of hepatitis B virus infection. J Hepatol 2017; 67: 370-398 [PMID: 28427875 DOI: 10.1016/j.jhep.2017.03.021]

72

Bortolotti F, Guido M, Bartolacci S, Cadrobbi P, Crivellaro C, Noventa F, Morsica G, Moriondo M, Gatta A. Chronic hepatitis B in children after e antigen seroclearance: final report of a 29-year longitudinal study. Hepatology 2006; 43: 556-562 [PMID: 16496323 DOI: 10.1002/hep.21077]

73

Wen WH, Chang MH, Hsu HY, Ni YH, Chen HL. The development of hepatocellular carcinoma among prospectively followed children with chronic hepatitis B virus infection. J Pediatr 2004; 144: 397-399 [PMID: 15001956 DOI: 10.1016/j.jpeds.2003.11.022]

74

Marx G, Martin SR, Chicoine JF, Alvarez F. Long-term follow-up of chronic hepatitis B virus infection in children of different ethnic origins. J Infect Dis 2002; 186: 295-301 [PMID: 12134225 DOI: 10.1086/341508]

75

Wu JF, Su YR, Chen CH, Chen HL, Ni YH, Hsu HY, Wang JL, Chang MH. Predictive effect of serial serum alanine aminotransferase levels on spontaneous HBeAg seroconversion in chronic genotype B and C HBV-infected children. J Pediatr Gastroenterol Nutr 2012; 54: 97-100 [PMID: 21716132 DOI: 10.1097/MPG.0b013e31822a033e]

76

Ni YH, Chang MH, Wang KJ, Hsu HY, Chen HL, Kao JH, Yeh SH, Jeng YM, Tsai KS, Chen DS. Clinical relevance of hepatitis B virus genotype in children with chronic infection and hepatocellular carcinoma. Gastroenterology 2004; 127: 1733-1738 [PMID: 15578511 DOI: 10.1053/j.gastro.2004.09.048]

77

Roushan MR, Bijani A, Ramzaninejad S, Roushan MH, Amiri MJ, Baiani M. HBeAg seroconversion in children infected during early childhood with hepatitis B virus. J Clin Virol 2012; 55: 30-33 [PMID: 22727680 DOI: 10.1016/j.jcv.2012.05.007]

78

Tseng YR, Wu JF, Ni YH, Chen HL, Chen CC, Wen WH, Hsu HY, Chang MH. Long-term effect of maternal HBeAg on delayed HBeAg seroconversion in offspring with chronic hepatitis B infection. Liver Int 2011; 31: 1373-1380 [PMID: 21745315 DOI: 10.1111/j.1478-3231.2011.02574.x]

79

Wu JF, Chiu YC, Chang KC, Chen HL, Ni YH, Hsu HY, Chang MH. Predictors of hepatitis B e antigen-negative hepatitis in chronic hepatitis B virus-infected patients from childhood to adulthood. Hepatology 2016; 63: 74-82 [PMID: 26389515 DOI: 10.1002/hep.28222]

80

Nicastro E, Mangili B, Giacomet V, Benincaso AR, Di Giorgio A, Sansotta N, Callegaro A, D'Antiga L. Longitudinal Immune Phenotype Assessment and Serological Outcome in Foreign-born Children With Chronic Hepatitis B. J Pediatr Gastroenterol Nutr 2020; 71: 381-387 [PMID: 32826716 DOI: 10.1097/MPG.0000000000002804]

81

Schwarz KB, Lombardero M, Di Bisceglie AM, Murray KF, Rosenthal P, Ling SC, Cloonan YK, Rodriguez-Baez N, Schwarzenberg SJ, Hoofnagle JH, Teckman J; Hepatitis B Research Network (HBRN). Phenotypes of Chronic Hepatitis B in Children From a Large North American Cohort. J Pediatr Gastroenterol Nutr 2019; 69: 588-594 [PMID: 31436702 DOI:

82


https://dx.doi.org/10.1016/S0140-6736(15)00349-9


https://dx.doi.org/10.1111/tri.13118


https://dx.doi.org/10.1097/TXD.0000000000000954


https://dx.doi.org/10.1159/000118790


https://dx.doi.org/10.1016/S2468-1253(19)30042-1

https://www.who.int/publications-detail-redirect/global-hepatitis-report-2017

https://www.who.int/publications-detail-redirect/global-hepatitis-report-2017


https://dx.doi.org/10.1093/infdis/151.4.599










https://dx.doi.org/10.1086/341508


https://dx.doi.org/10.1097/MPG.0b013e31822a033e




https://dx.doi.org/10.1016/j.jcv.2012.05.007










10.1097/MPG.0000000000002467]Di Bisceglie AM, Lombardero M, Teckman J, Roberts L, Janssen HL, Belle SH, Hoofnagle JH; Hepatitis B Research Network (HBRN). Determination of hepatitis B phenotype using biochemical and serological markers. J Viral Hepat 2017; 24: 320-329 [PMID: 27917600 DOI: 10.1111/jvh.12643]

83

Lee MH, Yang HI, Liu J, Batrla-Utermann R, Jen CL, Iloeje UH, Lu SN, You SL, Wang LY, Chen CJ; R. E.V.E.A.L.-HBV Study Group. Prediction models of long-term cirrhosis and hepatocellular carcinoma risk in chronic hepatitis B patients: risk scores integrating host and virus profiles. Hepatology 2013; 58: 546-554 [PMID: 23504622 DOI: 10.1002/hep.26385]

84

Arnone OC, Serranti D, Bartolini E, Mastrangelo G, Stinco M, Trapani S, Ricci S, Resti M, Indolfi G. Chronic hepatitis B in children, report of a single-centre longitudinal study on 152 children. J Viral Hepat 2020; 27: 1344-1351 [PMID: 32853482 DOI: 10.1111/jvh.13382]

85

Sokal EM, Paganelli M, Wirth S, Socha P, Vajro P, Lacaille F, Kelly D, Mieli-Vergani G; European Society of Pediatric Gastroenterology, Hepatology and Nutrition. Management of chronic hepatitis B in childhood: ESPGHAN clinical practice guidelines: consensus of an expert panel on behalf of the European Society of Pediatric Gastroenterology, Hepatology and Nutrition. J Hepatol 2013; 59: 814-829 [PMID: 23707367 DOI: 10.1016/j.jhep.2013.05.016]

86

Chu CM, Hung SJ, Lin J, Tai DI, Liaw YF. Natural history of hepatitis B e antigen to antibody seroconversion in patients with normal serum aminotransferase levels. Am J Med 2004; 116: 829-834 [PMID: 15178498 DOI: 10.1016/j.amjmed.2003.12.040]

87

Wong GL, Wong VW, Choi PC, Chan AW, Chim AM, Yiu KK, Chan HY, Chan FK, Sung JJ, Chan HL. Clinical factors associated with liver stiffness in hepatitis B e antigen-positive chronic hepatitis B patients. Clin Gastroenterol Hepatol 2009; 7: 227-233 [PMID: 19121647 DOI: 10.1016/j.cgh.2008.10.023]

88

Chen YC, Chu CM, Liaw YF. Age-specific prognosis following spontaneous hepatitis B e antigen seroconversion in chronic hepatitis B. Hepatology 2010; 51: 435-444 [PMID: 19918971 DOI: 10.1002/hep.23348]

89

Chen CF, Lee WC, Yang HI, Chang HC, Jen CL, Iloeje UH, Su J, Hsiao CK, Wang LY, You SL, Lu SN, Chen CJ; Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer in HBV (REVEAL–HBV) Study Group. Changes in serum levels of HBV DNA and alanine aminotransferase determine risk for hepatocellular carcinoma. Gastroenterology 2011; 141: 1240-1248, 1248.e1-1248. e2 [PMID: 21703214 DOI: 10.1053/j.gastro.2011.06.036]

90

D'Antiga L, Aw M, Atkins M, Moorat A, Vergani D, Mieli-Vergani G. Combined lamivudine/interferon-alpha treatment in "immunotolerant" children perinatally infected with hepatitis B: a pilot study. J Pediatr 2006; 148: 228-233 [PMID: 16492434 DOI: 10.1016/j.jpeds.2005.09.020]

91

Poddar U, Yachha SK, Agarwal J, Krishnani N. Cure for immune-tolerant hepatitis B in children: is it an achievable target with sequential combo therapy with lamivudine and interferon? J Viral Hepat 2013; 20: 311-316 [PMID: 23565612 DOI: 10.1111/jvh.12007]

92

Zhu S, Zhang H, Dong Y, Wang L, Xu Z, Liu W, Gan Y, Tang H, Chen D, Wang F, Zhao P. Antiviral therapy in hepatitis B virus-infected children with immune-tolerant characteristics: A pilot open-label randomized study. J Hepatol 2018; 68: 1123-1128 [PMID: 29452204 DOI: 10.1016/j.jhep.2018.01.037]

93

Rosenthal P, Ling SC, Belle SH, Murray KF, Rodriguez-Baez N, Schwarzenberg SJ, Teckman J, Lin HS, Schwarz KB; Hepatitis B Research Network (HBRN). Combination of Entecavir/Peginterferon Alfa-2a in Children With Hepatitis B e Antigen-Positive Immune Tolerant Chronic Hepatitis B Virus Infection. Hepatology 2019; 69: 2326-2337 [PMID: 30318613 DOI: 10.1002/hep.30312]

94

Wong GL. Management of chronic hepatitis B patients in immunetolerant phase: what latest guidelines recommend. Clin Mol Hepatol 2018; 24: 108-113 [PMID: 29353469 DOI: 10.3350/cmh.2017.0068]

95

Terrault NA, Lok ASF, McMahon BJ, Chang KM, Hwang JP, Jonas MM, Brown RS Jr, Bzowej NH, Wong JB. Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD 2018 hepatitis B guidance. Hepatology 2018; 67: 1560-1599 [PMID: 29405329 DOI: 10.1002/hep.29800]

96

Sokal EM, Conjeevaram HS, Roberts EA, Alvarez F, Bern EM, Goyens P, Rosenthal P, Lachaux A, Shelton M, Sarles J, Hoofnagle J. Interferon alfa therapy for chronic hepatitis B in children: a multinational randomized controlled trial. Gastroenterology 1998; 114: 988-995 [PMID: 9558288 DOI: 10.1016/s0016-5085(98)70318-x]

97

Jonas MM, Mizerski J, Badia IB, Areias JA, Schwarz KB, Little NR, Greensmith MJ, Gardner SD, Bell MS, Sokal EM; International Pediatric Lamivudine Investigator Group. Clinical trial of lamivudine in children with chronic hepatitis B. N Engl J Med 2002; 346: 1706-1713 [PMID: 12037150 DOI: 10.1056/NEJMoa012452]

98

Jonas MM, Kelly D, Pollack H, Mizerski J, Sorbel J, Frederick D, Mondou E, Rousseau F, Sokal E. Safety, efficacy, and pharmacokinetics of adefovir dipivoxil in children and adolescents (age 2 to <18 years) with chronic hepatitis B. Hepatology 2008; 47: 1863-1871 [PMID: 18433023 DOI: 10.1002/hep.22250]

99

Murray KF, Szenborn L, Wysocki J, Rossi S, Corsa AC, Dinh P, McHutchison J, Pang PS, Luminos LM, Pawlowska M, Mizerski J. Randomized, placebo-controlled trial of tenofovir disoproxil fumarate in adolescents with chronic hepatitis B. Hepatology 2012; 56: 2018-2026

100











https://dx.doi.org/10.1016/j.amjmed.2003.12.040


https://dx.doi.org/10.1016/j.cgh.2008.10.023














https://dx.doi.org/10.3350/cmh.2017.0068











[PMID: 22544804 DOI: 10.1002/hep.25818]Jonas MM, Chang MH, Sokal E, Schwarz KB, Kelly D, Kim KM, Ling SC, Rosenthal P, Oraseanu D, Reynolds L, Thiry A, Ackerman P. Randomized, controlled trial of entecavir vs placebo in children with hepatitis B envelope antigen-positive chronic hepatitis B. Hepatology 2016; 63: 377-387 [PMID: 26223345 DOI: 10.1002/hep.28015]

101

Wirth S, Zhang H, Hardikar W, Schwarz KB, Sokal E, Yang W, Fan H, Morozov V, Mao Q, Deng H, Huang Y, Yang L, Frey N, Nasmyth-Miller C, Pavlovic V, Wat C. Efficacy and Safety of Peginterferon Alfa-2a (40KD) in Children With Chronic Hepatitis B: The PEG-B-ACTIVE Study. Hepatology 2018; 68: 1681-1694 [PMID: 29689122 DOI: 10.1002/hep.30050]

102

Smolders EJ, Burger DM, Feld JJ, Kiser JJ. Review article: clinical pharmacology of current and investigational hepatitis B virus therapies. Aliment Pharmacol Ther 2020; 51: 231-243 [PMID: 31840863 DOI: 10.1111/apt.15581]

103

Hepatitis B Foundation. Drug Watch. [cited 2 March 2021]. In: Hepatitis B Foundation [Internet]. Available from: https://www.hepb.org/treatment-and-management/drug-watch/

104

Zoulim F, Lenz O, Vandenbossche JJ, Talloen W, Verbinnen T, Moscalu I, Streinu-Cercel A, Bourgeois S, Buti M, Crespo J, Manuel Pascasio J, Sarrazin C, Vanwolleghem T, Shukla U, Fry J, Yogaratnam JZ. JNJ-56136379, an HBV Capsid Assembly Modulator, Is Well-Tolerated and Has Antiviral Activity in a Phase 1 Study of Patients With Chronic Infection. Gastroenterology 2020; 159: 521-533. e9 [PMID: 32343960 DOI: 10.1053/j.gastro.2020.04.036]

105

Yuen MF, Agarwal K, Gane EJ, Schwabe C, Ahn SH, Kim DJ, Lim YS, Cheng W, Sievert W, Visvanathan K, Ruby E, Liaw S, Yan R, Huang Q, Colonno R, Lopatin U. Safety, pharmacokinetics, and antiviral effects of ABI-H0731, a hepatitis B virus core inhibitor: a randomised, placebo-controlled phase 1 trial. Lancet Gastroenterol Hepatol 2020; 5: 152-166 [PMID: 31711752 DOI: 10.1016/S2468-1253(19)30346-2]

106

Ma X, Lalezari J, Nguyen T, Bae H, Schiff ER, Fung S, Yuen M-F, Hassanein T, Hann H-W, Elkhashab M, Dieterich D, Sulkowski PK, Nahass R, Agarwal K, Ramji A, Park J, Ravendhran N, Chan S, Weilert F, Han S-H, Ayoub W, Gane EJ, Jacobson I, Bennett M, Huang Q, Yan R, Huey V, Ruby E, Liaw S, Colonno R, Lopatin U. Interim safety and efficacy results of the ABI-H0731 phase 2a program exploring the combination of ABI-H0731 with Nuc therapy in treatment-naive and treatment-suppressed chronic hepatitis B patients. [cited 2 March 2021]. Available from: https://program.manage.com/ProgramSearch/DownloadAbstractOfPresentation/ilc2019?id=417422

107

Al-Mahtab M, Bazinet M, Vaillant A. Safety and Efficacy of Nucleic Acid Polymers in Monotherapy and Combined with Immunotherapy in Treatment-Naive Bangladeshi Patients with HBeAg+ Chronic Hepatitis B Infection. PLoS One 2016; 11: e0156667 [PMID: 27257978 DOI: 10.1371/journal.pone.0156667]

108

Bazinet M, Pântea V, Cebotarescu V, Cojuhari L, Jimbei P, Albrecht J, Schmid P, Le Gal F, Gordien E, Krawczyk A, Mijočević H, Karimzadeh H, Roggendorf M, Vaillant A. Safety and efficacy of REP 2139 and pegylated interferon alfa-2a for treatment-naive patients with chronic hepatitis B virus and hepatitis D virus coinfection (REP 301 and REP 301-LTF): a non-randomised, open-label, phase 2 trial. Lancet Gastroenterol Hepatol 2017; 2: 877-889 [PMID: 28964701 DOI: 10.1016/S2468-1253(17)30288-1]

109

Bazinet M, Pântea V, Placinta G, Moscalu I, Cebotarescu V, Cojuhari L, Jimbei P, Iarovoi L, Smesnoi V, Musteata T, Jucov A, Dittmer U, Krawczyk A, Vaillant A. Safety and Efficacy of 48 Weeks REP 2139 or REP 2165, Tenofovir Disoproxil, and Pegylated Interferon Alfa-2a in Patients With Chronic HBV Infection Naïve to Nucleos(t)ide Therapy. Gastroenterology 2020; 158: 2180-2194 [PMID: 32147484 DOI: 10.1053/j.gastro.2020.02.058]

110

Yuen MF, Locarnini S, Huey Lim T, Strasser S, Sievert W, Cheng W, Thompson A, Given B, Schluep T, Hamilton J, Cloherty G, Wong D, Schwabe C, Jackson K, Ferrari C, Lai C-L, Gish R, Gane E. Short term RNA interference therapy in chronic hepatitis B using JNJ-3989 brings majority of patients to HBsAg < 100 IU/mL threshold. [cited 2 March 2021]. Available from: https://program.m-anage.com/ilc2019/en-GB/ProgramSearch/DownloadAbstract/151378

111

Gane E, Locarnini S, Huey Lim T, Strasser S, Sievert W, Cheng W, Thompson A, Given B, Schluep T, Hamilton J, Li Z, Cloherty G, Wong D, Schwabe C, Jackson K, Ferrari C, Lai C-L, Gish R, Yuen MF. First results with RNA interference (RNAi) in chronic hepatitis B (CHB) using ARO-HBV. The Liver Meeting; 2018 November; San Francisco, USA

112

VIR Biotechnology. VIR-2218 Demonstrates Dose-Dependent and Durable Reductions of Hepatitis B Surface Antigen in Phase 1/2 Trial. [cited 2 March 2021]. In: VIR Biotechnology [Internet]. Available from: https://investors.vir.bio/news-releases/news-release-details/vir-2218-demonstrates-dose-dependent-and-durable-reductions/

113

Yuen MF, Chen CY, Liu CJ, Jeng RWJ, Elkhashab M, Coffin C, Kim W, Greenbloom S, Ramji A, Lim YS, Kim YJ, Fung S, Kim DJ, Jang JW, Lee KS, Afdhal N, Radhakrishnan I, Macfarlane C, Jackson K, Locarnini S, Chan H. Ascending dose cohort study of inarigivir - A novel RIG I agonist in chronic HBV patients: Final results of the ACHIEVE trial. J Hepatol 2019; 70: e47-E48 [DOI: 10.1016/S0618-8278(19)30084-2]

114

Reyes M, Lutz JD, Lau AH, Gaggar A, Grant EP, Joshi A, Mackman RL, Ling J, Tan SK, Ayithan N, Daffis S, Woo J, Wu P, Lam T, Fletcher SP, Kottilil S, Poonia B, Gane EJ, Mathias A, German P. Safety, pharmacokinetics and pharmacodynamics of selgantolimod, an oral Toll-like receptor 8 agonist: a Phase Ia study in healthy subjects. Antivir Ther 2020; 25: 171-180 [PMID: 32667286 DOI: 10.3851/IMP3363]

115









https://www.hepb.org/treatment-and-management/drug-watch/




https://dx.doi.org/10.1016/S2468-1253(19)30346-2

https://program.manage.com/ProgramSearch/DownloadAbstractOfPresentation/ilc2019?id=417422




https://dx.doi.org/10.1016/S2468-1253(17)30288-1



https://program.m-anage.com/ilc2019/en-GB/ProgramSearch/DownloadAbstract/151378

https://investors.vir.bio/news-releases/news-release-details/vir-2218-demonstrates-dose-dependent-and-durable-reductions/

https://investors.vir.bio/news-releases/news-release-details/vir-2218-demonstrates-dose-dependent-and-durable-reductions/

https://dx.doi.org/10.1016/S0618-8278(19)30084-2


https://dx.doi.org/10.3851/IMP3363



Gane E, Zhao Y, Tan SK, Lau AH, Gaggar A, Subramanian M, Wallin J, Brooks AE, Dubar PR, Kottilil S, Tang L. Efficacy and Safety of Oral TLR8 Agonist Selgantolimod in Virally Suppressed Adult Patients With Chronic Hepatitis B: a Phase 2, Randomized, Double-Blind, Placebo-Controlled, Multicenter Study. Hepatology 2019

116

European Association for The Study of The Liver. Bulevirtide shows promise in the treatment of chronic hepatitis B/D (HBV/HDV) coinfection. [cited 2 March 2021]. In: European Association for The Study of The Liver [Internet]. Available from: https://2019.ilc-congress.eu/press-release/bulevirtide-shows-promise-in-the-treatment-of-chronic-hepatitis-b-d-hbv-hdv-coinfection/

117

Bogomolov P, Alexandrov A, Voronkova N, Macievich M, Kokina K, Petrachenkova M, Lehr T, Lempp FA, Wedemeyer H, Haag M, Schwab M, Haefeli WE, Blank A, Urban S. Treatment of chronic hepatitis D with the entry inhibitor myrcludex B: First results of a phase Ib/IIa study. J Hepatol 2016; 65: 490-498 [PMID: 27132170 DOI: 10.1016/j.jhep.2016.04.016]

118

Indolfi G, Abdel-Hady M, Bansal S, Debray D, Smets F, Czubkowski P, van der Woerd W, Samyn M, Jahnel J, Gupte G, Zellos A, Mozer-Glassberg Y, Verkade HJ, Sokal E, Fischler B. Management of Hepatitis B Virus Infection and Prevention of Hepatitis B Virus Reactivation in Children With Acquired Immunodeficiencies or Undergoing Immune Suppressive, Cytotoxic, or Biological Modifier Therapies. J Pediatr Gastroenterol Nutr 2020; 70: 527-538 [PMID: 31977956 DOI: 10.1097/MPG.0000000000002628]

119

Sarin SK, Kumar M, Lau GK, Abbas Z, Chan HL, Chen CJ, Chen DS, Chen HL, Chen PJ, Chien RN, Dokmeci AK, Gane E, Hou JL, Jafri W, Jia J, Kim JH, Lai CL, Lee HC, Lim SG, Liu CJ, Locarnini S, Al Mahtab M, Mohamed R, Omata M, Park J, Piratvisuth T, Sharma BC, Sollano J, Wang FS, Wei L, Yuen MF, Zheng SS, Kao JH. Asian-Pacific clinical practice guidelines on the management of hepatitis B: a 2015 update. Hepatol Int 2016; 10: 1-98 [PMID: 26563120 DOI: 10.1007/s12072-015-9675-4]

120

deBruyn JCC, Soon IS, Fonseca K, Feng S, Purtzki M, Goedhart C, Kuhn S, Vanderkooi OG, Wrobel I. Serologic Status of Routine Childhood Vaccines, Cytomegalovirus, and Epstein-Barr Virus in Children With Inflammatory Bowel Disease. Inflamm Bowel Dis 2019; 25: 1218-1226 [PMID: 30551205 DOI: 10.1093/ibd/izy366]

121

Aljaberi N, Ghulam E, Smitherman EA, Favier L, Dykes DMH, Danziger-Isakov LA, Brady RC, Huggins J. Maintaining Hepatitis B Protection in Immunocompromised Pediatric Rheumatology and Inflammatory Bowel Disease Patients. J Rheumatol 2020 [PMID: 32739895 DOI: 10.3899/jrheum.200283]

122

Reddy KR, Beavers KL, Hammond SP, Lim JK, Falck-Ytter YT; American Gastroenterological Association Institute. American Gastroenterological Association Institute guideline on the prevention and treatment of hepatitis B virus reactivation during immunosuppressive drug therapy. Gastroenterology 2015; 148: 215-9; quiz e16 [PMID: 25447850 DOI: 10.1053/j.gastro.2014.10.039]

123

Loomba R, Liang TJ. Hepatitis B Reactivation Associated With Immune Suppressive and Biological Modifier Therapies: Current Concepts, Management Strategies, and Future Directions. Gastroenterology 2017; 152: 1297-1309 [PMID: 28219691 DOI: 10.1053/j.gastro.2017.02.009]

124

Zhang MY, Zhu GQ, Shi KQ, Zheng JN, Cheng Z, Zou ZL, Huang HH, Chen FY, Zheng MH. Systematic review with network meta-analysis: Comparative efficacy of oral nucleos(t)ide analogues for the prevention of chemotherapy-induced hepatitis B virus reactivation. Oncotarget 2016; 7: 30642-30658 [PMID: 27121321 DOI: 10.18632/oncotarget.8907]

125

Fung J, Wong T, Chok K, Chan A, Cheung TT, Dai JW, Sin SL, Ma KW, Ng K, Ng KT, Seto WK, Lai CL, Yuen MF, Lo CM. Long-term outcomes of entecavir monotherapy for chronic hepatitis B after liver transplantation: Results up to 8 years. Hepatology 2017; 66: 1036-1044 [PMID: 28370215 DOI: 10.1002/hep.29191]

126

Huprikar S, Danziger-Isakov L, Ahn J, Naugler S, Blumberg E, Avery RK, Koval C, Lease ED, Pillai A, Doucette KE, Levitsky J, Morris MI, Lu K, McDermott JK, Mone T, Orlowski JP, Dadhania DM, Abbott K, Horslen S, Laskin BL, Mougdil A, Venkat VL, Korenblat K, Kumar V, Grossi P, Bloom RD, Brown K, Kotton CN, Kumar D. Solid organ transplantation from hepatitis B virus-positive donors: consensus guidelines for recipient management. Am J Transplant 2015; 15: 1162-1172 [PMID: 25707744 DOI: 10.1111/ajt.13187]

127

Hadziyannis SJ, Papatheodoridis GV. Hepatitis B e antigen-negative chronic hepatitis B: natural history and treatment. Semin Liver Dis 2006; 26: 130-141 [PMID: 16673291 DOI: 10.1055/s-2006-939751]

128

Brunetto MR, Oliveri F, Colombatto P, Coco B, Ciccorossi P, Bonino F. Treatment of HBeAg-negative chronic hepatitis B with interferon or pegylated interferon. J Hepatol 2003; 39 Suppl 1: S164-S167 [PMID: 14708697 DOI: 10.1016/s0168-8278(03)00329-5]

129

Brunetto MR, Oliveri F, Rocca G, Criscuolo D, Chiaberge E, Capalbo M, David E, Verme G, Bonino F. Natural course and response to interferon of chronic hepatitis B accompanied by antibody to hepatitis B e antigen. Hepatology 1989; 10: 198-202 [PMID: 2663695 DOI: 10.1002/hep.1840100213]

130

Hsu YS, Chien RN, Yeh CT, Sheen IS, Chiou HY, Chu CM, Liaw YF. Long-term outcome after spontaneous HBeAg seroconversion in patients with chronic hepatitis B. Hepatology 2002; 35: 1522-1527 [PMID: 12029639 DOI: 10.1053/jhep.2002.33638]

131

Ogunbosi B, Smuts H, Eley B, Korsman S, De Lacy R, Hardie DR. Fulminant hepatitis B virus (HBV) infection in an infant following mother-to-child transmission of an e-minus HBV mutant: Time to relook at HBV prophylaxis in South African infants. S Afr Med J 2018; 108: 389-392

132

https://2019.ilc-congress.eu/press-release/bulevirtide-shows-promise-in-the-treatment-of-chronic-hepatitis-b-d-hbv-hdv-coinfection/

https://2019.ilc-congress.eu/press-release/bulevirtide-shows-promise-in-the-treatment-of-chronic-hepatitis-b-d-hbv-hdv-coinfection/






https://dx.doi.org/10.1007/s12072-015-9675-4


https://dx.doi.org/10.1093/ibd/izy366


https://dx.doi.org/10.3899/jrheum.200283












https://dx.doi.org/10.1055/s-2006-939751


https://dx.doi.org/10.1016/s0168-8278(03)00329-5




https://dx.doi.org/10.1053/jhep.2002.33638



[PMID: 29843851 DOI: 10.7196/SAMJ.2018.v108i5.13017]Lampertico P, Viganò M, Di Costanzo GG, Sagnelli E, Fasano M, Di Marco V, Boninsegna S, Farci P, Fargion S, Giuberti T, Iannacone C, Regep L, Massetto B, Facchetti F, Colombo M; PegBeLiver Study Group. Randomised study comparing 48 and 96 wk peginterferon α-2a therapy in genotype D HBeAg-negative chronic hepatitis B. Gut 2013; 62: 290-298 [PMID: 22859496 DOI: 10.1136/gutjnl-2011-301430]

133

Leung DH, Wirth S, Yao BB, Viani RM, Gonzalez-Peralta RP, Jonas MM, Lobritto SJ, Narkewicz MR, Sokal E, Fortuny C, Hsu EK, Del Valle-Segarra A, Zha J, Larsen L, Liu L, Shuster DL, Cohen DE, Rosenthal P. Ombitasvir/Paritaprevir/Ritonavir With or Without Dasabuvir and With or Without Ribavirin for Adolescents With HCV Genotype 1 or 4. Hepatol Commun 2018; 2: 1311-1319 [PMID: 30411078 DOI: 10.1002/hep4.1250]

134

Rosenthal P, Narkewicz MR, Yao BB, Jolley CD, Lobritto SJ, Wen J, Molleston JP, Hsu EK, Jonas MM, Zha J, Liu L, Leung DH. Ombitasvir, Paritaprevir, Ritonavir, and Dasabuvir Mini-Tabs Plus Ribavirin for Children Aged 3-11 Years with Hepatitis C Genotype 1a. Adv Ther 2020; 37: 3299-3310 [PMID: 32451952 DOI: 10.1007/s12325-020-01389-9]

135

Wirth S, Gonzalez-Peralta RP, Squires RH, Mutschler F, Lang T, Pawlowska M, Sluzewski W, Majda-Stanislawska E, Fischler B, Balistreri WF, Jonas MM, Blondet N, Rosenthal P, Alkhouri N, Romero R, Grandhi A, Castronuovo P, Caro L, Rosenblum D, Du L, Haber BH. Elbasvir/grazoprevir in children aged 3 to <18 years with chronic hepatitis c virus (hcv) genotype (gt) 1 or 4 infection: final results from an iterative pharmacokinetic (pk) modeling study. Hepatology 2020; 555-556

136

Bansal S, Pawlowska M, Nebbia G, Di Dato F, Sluzewski W, Giacomet V, Szenborn L, Shao J, Yue MS, Hsueh C-H, Parhy B, Gaggar A, Kersey K, Mangia A, Kelly DA, Indolfi G. Safety and efficacy of sofosbuvir/velpatasvir/voxilaprevir in adolescents with chronic hepatitis c virus infection. Hepatology 2020; 571

137

El-Sayed MH, Hassany M, Asem N. A pilot study for safety and efficacy of 12 wk sofosbuvir plus daclatasvir with or without ribavirin in Egyptian adolescents with chronic hepatitis C virus Infection. J Hepatology 2017; 178

138

El-Shabrawi MH, Abdo AM, El-Khayat HR, Yakoot M. Shortened 8 Weeks Course of Dual Sofosbuvir/Daclatasvir Therapy in Adolescent Patients, With Chronic Hepatitis C Infection. J Pediatr Gastroenterol Nutr 2018; 66: 425-427 [PMID: 29135824 DOI: 10.1097/MPG.0000000000001838]

139

Yakoot M, El-Shabrawi MH, AbdElgawad MM, Mahfouz AA, Helmy S, Abdo AM, El-Khayat HR. Dual Sofosbuvir/Daclatasvir Therapy in Adolescent Patients With Chronic Hepatitis C Infection. J Pediatr Gastroenterol Nutr 2018; 67: 86-89 [PMID: 29570560 DOI: 10.1097/MPG.0000000000001968]

140

Abdel Ghaffar TY, El Naghi S, Abdel Gawad M, Helmy S, Abdel Ghaffar A, Yousef M, Moafy M. Safety and efficacy of combined sofosbuvir/daclatasvir treatment of children and adolescents with chronic hepatitis C Genotype 4. J Viral Hepat 2019; 26: 263-270 [PMID: 30380158 DOI: 10.1111/jvh.13032]

141


https://dx.doi.org/10.7196/SAMJ.2018.v108i5.13017


https://dx.doi.org/10.1136/gutjnl-2011-301430


https://dx.doi.org/10.1002/hep4.1250


https://dx.doi.org/10.1007/s12325-020-01389-9








World Journal of



MINIREVIEWS

Pancreatitis after endoscopic retrograde cholangiopancreatography: A narrative review

Igor Braga Ribeiro, Epifanio Silvino do Monte Junior, Antonio Afonso Miranda Neto, Igor Mendonça Proença, Diogo Turiani Hourneaux de Moura, Mauricio Kazuyoshi Minata, Edson Ide, Marcos Eduardo Lera dos Santos, Gustavo de Oliveira Luz, Sergio Eiji Matuguma, Spencer Cheng, Renato Baracat, Eduardo Guimarães Hourneaux de Moura

ORCID number: Igor Braga Ribeiro 0000-0003-1844-8973; Epifanio Silvino do Monte Junior 0000-0001-7304-8222; Antonio Afonso Miranda Neto 0000-0002-9439-9088; Igor Mendonça Proença 0000-0003-0274-038X; Diogo Turiani Hourneaux de Moura 0000-0002-7446-0355; Mauricio Kazuyoshi Minata 0000-0002-9243-1371; Edson Ide 0000-0003-4533-6117; Marcos Eduardo Lera dos Santos 0000-0001-9759-3807; Gustavo de Oliveira Luz 0000-0001-7396-8440; Sergio Eiji Matuguma 0000-0002-9956-7183; Spencer Cheng 0000-0001-9584-203X; Renato Baracat 0000-0002-2701-9006; Eduardo Guimarães Hourneaux de Moura 0000-0003-1215-5731.

Author contributions: Ribeiro IB performed the acquisition of data, analysis, interpretation of data, drafting the article, revising the article, final approval; do Monte Junior ES, Miranda Neto AA, Proença IM, de Moura DTH, and Minata MK conducted data analysis and interpretation, revised the article, and final approval; Ide E, and de Moura EGH conducted data analysis and interpretation, drafted the article, and final approval; dos Santos MEL, Matuguma SE, Cheng S, and Baracat R revised, edited and

Igor Braga Ribeiro, Epifanio Silvino do Monte Junior, Antonio Afonso Miranda Neto, Igor Mendonça Proença, Diogo Turiani Hourneaux de Moura, Mauricio Kazuyoshi Minata, Edson Ide, Marcos Eduardo Lera dos Santos, Gustavo de Oliveira Luz, Sergio Eiji Matuguma, Spencer Cheng, Renato Baracat, Eduardo Guimarães Hourneaux de Moura, Department of Gastrointestinal Endoscopy Unit, University of São Paulo School of Medicine, São Paulo 05403-010, Brazil

Corresponding author: Igor Braga Ribeiro, MD, Research Scientist, Surgeon, Department of Gastrointestinal Endoscopy Unit, University of São Paulo School of Medicine, Av. Dr Enéas de Carvalho Aguiar, 225, 6o andar, bloco 3, Cerqueira Cesar, São Paulo 05403-010, Brazil. [email protected]

AbstractAcute post-endoscopic retrograde cholangiopancreatography pancreatitis (PEP) is a feared and potentially fatal complication that can be as high as up to 30% in high-risk patients. Pre-examination measures, during the examination and after the examination are the key to technical and clinical success with a decrease in adverse events. Several studies have debated on the subject, however, numerous topics remain controversial, such as the effectiveness of prophylactic medications and the amylase dosage time. This review was designed to provide an update on the current scientific evidence regarding PEP available in the literature.

Key Words: Endoscopic retrograde cholangiopancreatography; Pan-creatitis; Post-endoscopic retrograde cholangiopancreatography pancreatitis; Adverse events; Pancreatitis; Prevention


Core Tip: Acute post-endoscopic retrograde cholangiopancreatography pancreatitis (PEP) is a feared and potentially fatal complication. Early diagnosis remains the key to the clinical success of these patients. Unfortunately, several topics remain contro-versial, especially early diagnosis with hyperamylasemia still being mistaken for PEP.



http://orcid.org/0000-0003-1844-8973

http://orcid.org/0000-0003-1844-8973

http://orcid.org/0000-0001-7304-8222

http://orcid.org/0000-0001-7304-8222

http://orcid.org/0000-0001-7304-8222

http://orcid.org/0000-0002-9439-9088

http://orcid.org/0000-0002-9439-9088

http://orcid.org/0000-0003-0274-038X

http://orcid.org/0000-0003-0274-038X

http://orcid.org/0000-0002-7446-0355

http://orcid.org/0000-0002-7446-0355

http://orcid.org/0000-0002-9243-1371

http://orcid.org/0000-0002-9243-1371

http://orcid.org/0000-0002-9243-1371

http://orcid.org/0000-0003-4533-6117

http://orcid.org/0000-0003-4533-6117

http://orcid.org/0000-0003-4533-6117

http://orcid.org/0000-0001-9759-3807

http://orcid.org/0000-0001-9759-3807

http://orcid.org/0000-0001-7396-8440

http://orcid.org/0000-0001-7396-8440

http://orcid.org/0000-0001-7396-8440

http://orcid.org/0000-0002-9956-7183

http://orcid.org/0000-0002-9956-7183

http://orcid.org/0000-0001-9584-203X

http://orcid.org/0000-0001-9584-203X

http://orcid.org/0000-0002-2701-9006

http://orcid.org/0000-0002-2701-9006

http://orcid.org/0000-0002-2701-9006

http://orcid.org/0000-0003-1215-5731

http://orcid.org/0000-0003-1215-5731

http://orcid.org/0000-0003-1215-5731


Ribeiro IB et al. Pancreatitis after ERCP


drafted the article, and final approval.

Conflict-of-interest statement: Dr. Moura reports personal fees from Boston Scientific, personal fees from Olympus, outside the submitted work.




Country/Territory of origin: Brazil

Peer-review report’s scientific quality classificationGrade A (Excellent): A, A Grade B (Very good): 0 Grade C (Good): C Grade D (Fair): 0 Grade E (Poor): 0

Received: December 10, 2020 Peer-review started: December 10, 2020 First decision: January 27, 2021 Revised: January 30, 2021 Accepted: March 18, 2021 Article in press: March 18, 2021 Published online: May 28, 2021

P-Reviewer: Draganov P, Xiao B S-Editor: Fan JR L-Editor: Webster JR P-Editor: Ma YJ

The purpose of this review is to demonstrate the evidence in the current literature on PEP.

Citation: Ribeiro IB, do Monte Junior ES, Miranda Neto AA, Proença IM, de Moura DTH, Minata MK, Ide E, dos Santos MEL, Luz GO, Matuguma SE, Cheng S, Baracat R, de Moura EGH. Pancreatitis after endoscopic retrograde cholangiopancreatography: A narrative review. World J Gastroenterol 2021; 27(20): 2495-2506URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2495.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2495

INTRODUCTIONStarting in 1968, endoscopic retrograde cholangiopancreatography (ERCP) was a watershed in the diagnosis and treatment of biliopancreatic diseases. Since then, an accurate indication for this examination is very important given the potential adverse effects associated with the procedure[1].

Early recognition and proper management of potential adverse events are essential to reduce associated morbidity and mortality.

As in other endoscopic procedures, there are safety determinants for ERCP, in addition to the precise indication, the clinical condition of the patient, age, sex, the type of sedation used, what type of therapeutic procedure performed, the appropriate use of accessories and the training of the endoscopist and assistants are taken into consideration[2].

Acute pancreatitis is the most common serious complication after ERCP[3,4], often confused with an increase in serum amylase concentration that occurs in up to 75% of patients[5,6].

Acute clinical pancreatitis itself, defined as a clinical syndrome of abdominal pain and hyperamylasemia which requires hospitalization, is much less common than it appears. There are still some controversies in the literature on the subject. The purpose of this review is to provide an update on post-ERCP pancreatitis and its prevention.

PATHOGENESISThe determinants of the inflammatory process in the pancreas are multifactorial. Several proposed factors can act independently or in combination to induce post-ERCP pancreatitis (PEP). The two most important are mechanical injury due to instru-mentation in the pancreatic duct and hydrostatic injury due to contrast injection[7].

During ERCP and sphincterotomy, the pancreas is exposed to various forms of trauma: mechanical, chemical, hydrostatic, thermal, and even allergic[8].

It is also known that prolonged manipulation around the papillary orifice, inadvertent cannulation of the pancreatic duct and multiple injections into the pancreatic duct are common when selective cannulation of the bile duct is difficult[9,10]. This can result in mechanical damage to the duct or ampoule. Thermal injury to the electrocautery current can also produce edema of the pancreatic orifice, leading to obstruction of the duct, impairing the emptying of pancreatic secretions[11].

Hydrostatic injury due to excessive injection of contrast into the pancreatic duct is probably an important cause of PEP[12].

Either by allergy or chemical injury, contrast agents can lead to injuries. In a study by George et al[13], there was no statistically significant difference between the types of contrast in the analysis of randomized studies.

EPIDEMIOLOGY AND RISK FACTORSIncidenceThe incidence of pancreatitis post-ERCP can vary from 1% to 10%, reaching an alarming 30% in high-risk patients[14,15]. Stratification of the degree of post-examination pancreatitis shows incidence rates of 3.6% to 4% for mild acute pancre-








atitis, 1.8% to 2.8% for moderate acute pancreatitis, and 0.3% to 0.5% for severe acute pancreatitis[16,17] with a mortality rate of 0.2%[18]. Higher rates are observed in patients undergoing evaluation for possible sphincter of Oddi dysfunction[19].

Risk factorsAccording to the guidelines of the European Society for Gastrointestinal Endoscopy (ESGE)[20] and the American Society for Gastrointestinal Endoscopy (ASGE)[2]: History of pancreatitis, suspected sphincter of Oddi dysfunction, female gender, and young age are definitely “patient-related risk factors” for PEP. On the other hand, difficult cannulation, pancreatic injection, and pre-cut sphincterotomy are "risk factors related to the procedure[3,4].

Patient-related factorsThere are several factors related to the patient, the most common factors are female gender, normal levels of bilirubin, young adults, history of recurrent pancreatitis, and patients with suspected sphincter of Oddi dysfunction. Patients with a history of chronic pancreatitis have a protective effect against PEP[2].

Unfortunately, risk factors are additive[7,21,22]. For example, the combination of female gender, patients with suspected sphincter of Oddi dysfunction, young age, difficult cannulation, bilirubin within the acceptable standard, and absence of bile duct stones are associated with a risk of the pancreatitis of more than 40%.

Operator-related factorsThese are the most subjective factors. It is believed that the experience of the endoscopist, the presence of fellows and multiple operators is an independent risk factor for PEP[23,24].

Procedure-related factorsThe factors related to the procedure are the best studied and discussed in the literature. Pre-cut sphincterotomy, often used in difficult ERCP, time and number of cannulation times, trauma, and edema of the major duodenal papillae due to the number of attempts are independent factors for PEP[25].

In a systematic review with a meta-analysis that included 25 randomized controlled trials (RCTs) evaluating the incidence of PEP in patients undergoing sphincterotomy, ballooning dilation of the major duodenal papilla without sphincterotomy and patients undergoing both procedures, it was concluded that the incidence of PEP was similar between the groups[26].

The risk factors can be divided into three groups and are shown in Table 1[7,27-29].

CLINICAL MANIFESTATIONSThe clinical manifestations of PEP are the same as those seen in patients with acute pancreatitis due to other causes.

These include epigastric or upper right quadrant pain, abdominal tenderness and high levels of amylase and lipase.

Post-ERCP acute pancreatitis can be classified as mild, moderate or severe based on the American Gastroenterology Association[30] and the American College of Gastroenterology[31]: (1) mild-amylase levels 24 h after the examination, remaining above up to three times the reference value with necessary hospitalization; (2) moderate-need for hospitalization of 4 to 10 d; and (3) severe-need for hospitalization over 10 d or need for invasive therapeutic intervention.

DIAGNOSISMost patients with PEP have an acute onset of severe and persistent epigastric abdominal pain and in approximately 50% of patients, the pain radiates to the back. Approximately 90% of patients experience nausea and vomiting that can persist for several hours[32].

Patients with severe acute pancreatitis may have dyspnea due to diaphragmatic inflammation secondary to pancreatitis, pleural effusions, or acute respiratory distress syndrome, and 5% to 10% of patients with severe acute pancreatitis may have painless disease and unexplained hypotension[33].



Table 1 Risk factors for post-endoscopic retrograde cholangiopancreatography pancreatitis related to the operator and the procedure

Operator related Factors related to the procedure

Inadequate training Extended procedure time (≥ 30 min)

Lack of experience Difficult cannulation (≥ 15 min)

Patient-related risk factors Injection of contrast into the pancreatic duct

Young age Sphincter of Oddi manometry

Women Pancreatic sphincterotomy

Normal serum bilirubin Small papillary sphincterotomy

Recurrent pancreatitis Biliary balloon sphincteroplasty

Previous ERCP-induced pancreatitis Endoscopic papillectomy with loop

Sphincter of Oddi dysfunction Pancreatic intraductal ultrasound

Precut sphincterotomy

ERCP: Endoscopic retrograde cholangiopancreatography.

For diagnostic confirmation, radiological evidence with computed tomography may be necessary[34] but biochemical tests are more commonly used, as they are inexpensive and sensitive[35].

Early diagnosis of PEP is crucial as late diagnosis can be fatal[36,37].

Pancreatic enzymesThe diagnosis of PEP can be complicated, since elevations in pancreatic enzymes are common after the examination, but are generally not associated with clinical pancre-atitis.

There is no consensus in the literature on the ideal time after examination to request serum amylase levels and their real meaning. Two prospective studies including 263 and 886 patients found that the 4-h post-ERCP amylase level proved useful in predicting PEP[38,39]. We suggest that the patient should fast for the next 12 h and amylase analysis should be requested for all patients.

In patients with suspected pancreatitis, the degree and speed of elevations in pancreatic enzymes may be a way of differentiating patients with PEP from those in pain due to other causes. Some studies state that patients with PEP often have serum amylase levels more than five times the upper limit of normal[40,41].

Patients undergoing a contrast study of the main pancreatic duct should be admitted if the 4-h amylase level is greater than 2.5 times the upper reference limit. Patients who have not undergone a contrast study should be admitted if the 4-h amylase level is greater than five times the upper limit of normal[38]. The 4-h post-ERCP amylase level was useful in predicting PEP in two prospective studies including 263 and 886 patients, respectively[38,39].

DIFFERENTIAL DIAGNOSISNot all patients with pain after ERCP have pancreatitis. Other causes of abdominal pain after ERCP include discomfort due to air insufflation[42-44] and perforation.

In patients with discomfort due to air insufflation, the pain is generally not as severe as that seen with PEP, and pancreatic enzyme levels may be normal or elevated, as pancreatic enzymes are elevated in most patients after ERCP[5].

If serum lipase is less than three times the upper limit of normal, pancreatitis is unlikely (specificity of 85 to 98%). However, it should be borne in mind that amylase and lipase start to increase several hours after the onset of pancreatitis; thus, blood tests taken soon after ERCP can show false negative results.

If the clinical suspicion of pancreatitis is high, tests should be repeated at least 4-6 h after ERCP. Perforated patients may experience diffuse abdominal pain, bloating, tachycardia, fever, and leukocytosis.

Symptoms can be immediate after the examination or hours later[45]. Many of the perforation symptoms overlap with those of acute pancreatitis and, if perforation is



suspected, an abdominal tomography should be performed immediately for intraperi-toneal and retroperitoneal evaluation[46].

TREATMENTMost of the patients who develop PEP requiring hospitalization are classified as mild. In severe cases, admission to an intensive care unit may be necessary[30-31]. Initial treatment should focus on the following:

Pain controlThis usually manifests as abdominal pain and must be one of the main pillars in the treatment, since its non-control can lead to hemodynamic instability. There is still a lot of controversy in the use of opioids such as morphine as it has been shown to increase pressure in the sphincter of Oddi, but without clinical data that this has resulted in worsening of pancreatitis. Indicated: Meperidine, fentanyl, and morphine[47].

Particular attention should be given to patients who are dehydrated or who have not received an adequate amount of fluids since hypovolemia and hemoconcentration can cause ischemic pain and increased lactic acidosis.

Fluid replacementFluid replacement is one of the main items in the treatment of patients with PEP. The use of crystalloid solutions, mainly Ringer Lactate, from 5 to 10 mL/kg/h is recommended in patients without restrictions. In critically ill patients, with hemodynamic instability, 20 mL/kg is recommended in 30 min followed by 3 mL/kg/h in the next 8 to 12 h[48,49].

MonitoringAs these patients’ condition may worsen in the next 24 h, it is recommended that they be monitored for at least 48 h. This surveillance includes vital signs, urine volume, electrolytes, and blood glucose[48].

AntibioticsProphylactic antibiotics are not recommended in patients with PEP regardless of the type or severity of the disease. Antibiotics should only be used in about 20% of patients who develop extrapancreatic infections[48,50].

NutritionFasting is recommended for all patients with PEP. The time for restarting oral feeding is dependent on the severity of pancreatitis[51].

PREVENTIONCertain measures can reduce the incidence of PEP[7]: (1) adequate training and experience of endoscopists and assistants; (2) use of wire-guided techniques for biliary cannulation; (3) minimizing the number of cannulation attempts; (4) placement of a prophylactic pancreatic stent in patients at high risk of developing PEP; (5) placement of prophylactic pancreatic stents in patients who require the assistance of a pancreatic guidewire for biliary cannulation (double guidewire technique); (6) selective cannulation of the bile duct if an assessment of the pancreatic duct is not necessary; (7) minimizing the volume of contrast medium injected into the pancreatic duct, if necessary; (8) careful use of the electrocautery current during sphincterotomy; (9) high-risk patients should undergo ERCP in specialized centers; and (10) use of carbon dioxide for luminal insufflation to decrease post-procedure abdominal pain that can be mistaken for pancreatitis.

Effectiveness of preventive measuresEndoscopic techniques: The endoscopic technique is an important factor in the development of PEP. Cannulation guided by a hydrophilic-coated wire, careful use of electrocautery during sphincterotomy, and placement of a prophylactic pancreatic stent should be undertaken in patients at high risk of developing PEP.



Cannulation techniques: Various instruments such as guidewires are available and can decrease the risk of PEP as suggested by the ASGE and ESGE[52-55].

A systematic review that included only randomized trials, evaluating a total of 3450 patients, demonstrated that cannulation guided by a guidewire was superior to contrast-assisted cannulation technique[56]. Cannulation rates were higher for the wire-guided technique, and the risk of PEP was halved.

In a multicenter RCT, including 274 patients with naïve papilla undergoing ERCP using wire-guided cannulation in whom the guidewire was inadvertently inserted into the main pancreatic duct, the patients were randomized to undergo the double guidewire technique or a new cannulation attempt with a single wire. Conversion to the double guidewire technique did not facilitate selective bile duct cannulation and did not decrease the incidence of PEP compared to the new single guidewire cannulation attempt. However, double guidewire cannulation was more effective in patients with malignant biliary stenosis[57].

Electrocautery: In a recent systematic review evaluating 11 randomized studies involving 1791 patients, it was found that the performance of sphincterotomy with electrocautery in pure cut mode leads to a higher incidence of mild bleeding compared to endocut and blend. However, this modality may have a lower incidence of pancre-atitis. Monopolar mode causes higher rates of pancreatitis compared to bipolar mode[11].

Pancreatic stent: Pancreatic stent placement can be performed as prophylaxis for PEP mainly in high-risk patients. We suggest the use in patients undergoing pancreatic sphincterotomy, a contrasting study of the main pancreatic duct when it is necessary to use the double guidewire technique, in patients with suspected sphincter of Oddi dysfunction, and in patients undergoing pre-cut sphincterotomy[7].

The possible benefit is believed to be related to a reduction in pancreatic intraductal pressure of papillary edema.

Studies have shown that in special situations, the passage of a pancreatic stent in the DPP may be necessary to prevent the evolution of pancreatitis after ERCP. This procedure must be performed 8 to 20 h after the start of PEP[58-60].

Pancreatic stents should be short (less than 5 cm and small in diameter (5 French), plastic, and not have flanges distally[7]. Non-flanged stents can lead to spontaneous migration to the gastrointestinal tract, which occurs in 95% of cases within 10 d[55]. If radiographs show evidence of persistent stent within 1 wk, a high endoscopy should be performed to remove the stent[55].

Intravenous hydration: ASGE guidelines suggest the use of periprocedural intra-venous hydration with lactated Ringer to decrease the risk of PEP[2].

In a RCT of 150 patients, the PEP rate was lower in patients who received aggressive intravenous hydration compared to standard therapy[61].

In patients with contraindications to rectal non-steroidal anti-inflammatory drugs (NSAIDs), who are not at risk of fluid overload and a pancreatic stent has not been placed, the suggested alternative is aggressive hydration with lactated Ringer's solution (3 mL/kg/h during ERCP, 20 mL/kg bolus after ERCP and 3 mL/kg/h for 8 h after the examination)[25].

ChemopreventionSince 1977, more than 35 different drugs have been evaluated for the prevention of PEP with variable results[62,63]. The available options are discussed below:

NSAIDsRectal NSAIDs: The ASGE and the ESGE recommend the administration of NSAIDs to reduce the incidence and severity of PEP (for example, 100 mg of indomethacin or diclofenac rectally immediately before or after ERCP)[2,20].

A systematic review with meta-analysis evaluating 21 RCTs with a total of 6854 patients, found that the rectal administration of NSAIDs in all patients adequately reduced the incidence of PEP and that mild pancreatitis was the only preventable result. In this context, both diclofenac and indomethacin are considered effective[64].

Rectal NSAIDs were also compared indirectly with stenting of the pancreatic duct. A meta-analysis showed that rectal NSAIDs were superior to pancreatic duct stenting for the prevention of PEP (OR 0.48, 95%CI: 0.26-0.87)[65].

Non-rectal NSAIDs: There are no data in the current literature to support the prophy-lactic use of any NSAIDs administered by any non-rectal route or in combination with



other agents.In a multicenter study with 430 patients, oral diclofenac (50 mg) before and after

ERCP showed no benefit compared to placebo[66].

Other agents in the prevention of PEP: There are several drugs potentially useful for the prevention of PEP although some drugs are difficult to access and few are used for this purpose.

Topical adrenalineA systematic review with meta-analysis evaluating 6 randomized and 2 observational studies including 4123 patients found that topical adrenaline does not provide any additional advantage in combination with rectal indomethacin in the prevention of PEP in patients who underwent ERCP. However, topical adrenaline alone is associated with a lower risk of PEP compared to placebo and can be considered if rectal indomethacin is not available or if the patient has any contraindications to its use[67].

NitratesIn a systematic review with 2000 patients, the use of nitroglycerin was compared to placebo and it was found that the intervention group demonstrated a 10% reduction in the development of PEP[68].

These data suggest that nitrates combined with rectal NSAIDs may provide more benefits than NSAIDs alone[69,70]. In a randomized trial including 886 patients undergoing ERCP, the risk of PEP was lower in patients treated with diclofenac suppositories and sublingual isosorbide dinitrate compared to patients receiving diclofenac suppositories alone (RR: 0.59, 95%CI: 0.37-0.95)[70].

PANCREATIC SECRETION INHIBITORSSomatostatinSomatostatin leads to a reduction in pancreatic exocrine secretion of basal origin and also when stimulated. A meta-analysis that included 9 studies concluded that somatostatin was ineffective in preventing PEP when administered in the short-term (< 6 h) or long-term (≥ 12 h)[71]. Another meta-analysis, which included 11 RCTs with a total of 2869 patients, found no benefit when somatostatin was administered as a short-term infusion, but showed a benefit when administered as a single bolus or as a long-term infusion[72].

OctreotideTwo systematic reviews with meta-analyses found no benefit of octreotide use in PEP prophylaxis[73,74].

INHIBITORS OF PROTEASE ACTIVATIONThe most studied protease inhibitors include gabexate mesylate, nafamostat mesylate, and ulinastatin. As the activation of proteolytic enzymes can contribute to PEP, protease inhibitors have been investigated in the prevention of PEP. In a meta-analysis of 18 studies involving 4966 patients, there was a small benefit with the use of protease inhibitors[75].

Gabexate mesylateAlthough controversial results have been observed, a meta-analysis of five studies concluded that gabexate mesylate was ineffective in reducing pancreatitis and post-ERCP pain[71].

Nafamostat mesylateAlthough controversial results have been observed, a meta-analysis that included 7 RCTs with 2956 patients found that the incidence of PEP was reduced by 53% compared to patients in the control groups (RR: 0.47, 95%CI: 0.34-0.63)[76].

Another meta-analysis which included 26 studies, found that unlike gabexate mesylate and ulinastatin, nafamostat mesylate and NSAIDs were associated with decreased risk of PEP[77].



UlinastatinA systematic review with a meta-analysis that included 7 RCTs comparing ulinastatin with placebo or gabexate demonstrated a decreased risk of PEP in patients receiving ulinastatin[77].

MONITORING CARE AFTER ERCPMany complications of ERCP are apparent during the first 6 h after the procedure, and others may take days to manifest. We suggest the following recommendations: (1) Serum amylase: Studies have shown that the 4-h serum amylase level is a useful measure in predicting PEP; (2) Clinical monitoring: The immediate post-examination period is critical and the patient must be monitored for signs and symptoms of adverse events; and (3) Diet: We recommend fasting the patient for 6 to 12 h after the examination and discharge only after the serum amylase results and clinical reassessment (patient without complaints of abdominal pain, for example).

CONCLUSIONPancreatitis after ERCP is a feared, potentially fatal, and not entirely preventable complication. The correct and early diagnosis is a turning point in the outcome of the disease. Pre-examination measures such as a correct indication for the procedure, use of rectal NSAIDs, and well-trained staff are necessary. During the examination: Hyperhydration, examination with precision and speed with the correct technique and appropriate material, and prophylactic use of a pancreatic stent. After the examination, maintaining fasting and the appropriate amylase dosage are essential for the clinical and technical success of the procedure.

REFERENCESThiruvengadam NR, Kochman ML. Emerging Therapies to Prevent Post-ERCP Pancreatitis. Curr Gastroenterol Rep 2020; 22: 59 [PMID: 33188441 DOI: 10.1007/s11894-020-00796-w]

1

ASGE Standards of Practice Committee, Chandrasekhara V, Khashab MA, Muthusamy VR, Acosta RD, Agrawal D, Bruining DH, Eloubeidi MA, Fanelli RD, Faulx AL, Gurudu SR, Kothari S, Lightdale JR, Qumseya BJ, Shaukat A, Wang A, Wani SB, Yang J, DeWitt JM. Adverse events associated with ERCP. Gastrointest Endosc 2017; 85: 32-47 [PMID: 27546389 DOI: 10.1016/j.gie.2016.06.051]

2

Feurer ME, Adler DG. Post-ERCP pancreatitis: review of current preventive strategies. Curr Opin Gastroenterol 2012; 28: 280-286 [PMID: 22450899 DOI: 10.1097/MOG.0b013e3283528e68]

3

ASGE Standards of Practice Committee, Anderson MA, Fisher L, Jain R, Evans JA, Appalaneni V, Ben-Menachem T, Cash BD, Decker GA, Early DS, Fanelli RD, Fisher DA, Fukami N, Hwang JH, Ikenberry SO, Jue TL, Khan KM, Krinsky ML, Malpas PM, Maple JT, Sharaf RN, Shergill AK, Dominitz JA. Complications of ERCP. Gastrointest Endosc 2012; 75: 467-473 [PMID: 22341094 DOI: 10.1016/j.gie.2011.07.010]

4

Pieper-Bigelow C, Strocchi A, Levitt MD. Where does serum amylase come from and where does it go? Gastroenterol Clin North Am 1990; 19: 793-810 [PMID: 1702756]

5

Hormati A, Alemi F, Mohammadbeigi A, Sarkeshikian SS, Saeidi M. Prevalence of Endoscopic Retrograde Cholangiopancreatography Complications and Amylase Sensitivity for Predicting Pancreatitis in ERCP Patients. Gastroenterol Nurs 2020; 43: 350-354 [PMID: 32889967 DOI: 10.1097/SGA.0000000000000473]

6

Boškoski I, Costamagna G. How to Prevent Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis. Gastroenterology 2020; 158: 2037-2040 [PMID: 32197979 DOI: 10.1053/j.gastro.2020.03.019]

7

Funatsu E, Masuda A, Takenaka M, Nakagawa T, Shiomi H, Yoshinaka H, Kobayashi T, Sakai A, Yagi Y, Yoshida M, Arisaka Y, Okabe Y, Kutsumi H, Azuma T. History of Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis and Acute Pancreatitis as Risk Factors for Post-ERCP Pancreatitis. Kobe J Med Sci 2017; 63: E1-E8 [PMID: 29434167]

8

Johnson GK, Geenen JE, Johanson JF, Sherman S, Hogan WJ, Cass O. Evaluation of post-ERCP pancreatitis: potential causes noted during controlled study of differing contrast media. Midwest Pancreaticobiliary Study Group. Gastrointest Endosc 1997; 46: 217-222 [PMID: 9378207 DOI: 10.1016/S0016-5107(97)70089-0]

9

Proença IM, dos Santos MEL, de Moura DTH, Ribeiro IB, Matuguma SE, Cheng S, McCarty TR, do Monte Junior ES, Sakai P, de Moura EGH. Role of pancreatography in the endoscopic management of encapsulated pancreatic collections – review and new proposed classification. World J

10


https://dx.doi.org/10.1007/s11894-020-00796-w


https://dx.doi.org/10.1016/j.gie.2016.06.051


https://dx.doi.org/10.1097/MOG.0b013e3283528e68





https://dx.doi.org/10.1097/SGA.0000000000000473





https://dx.doi.org/10.1016/S0016-5107(97)70089-0



Gastroenterol 2020; 26: 7104-7117 [PMID: 33362371 DOI: 10.3748/wjg.v26.i45.7104]Funari MP, Ribeiro IB, de Moura DTH, Bernardo WM, Brunaldi VO, Rezende DT, Resende RH, de Marco MO, Franzini TAP, de Moura EGH. Adverse events after biliary sphincterotomy: Does the electric current mode make a difference? Clin Res Hepatol Gastroenterol 2020; 44: 739-752 [PMID: 32088149 DOI: 10.1016/j.clinre.2019.12.009]

11

Sherman S, Hawes RH, Troiano FP, Lehman GA. Pancreatitis following bile duct sphincter of Oddi manometry: utility of the aspirating catheter. Gastrointest Endosc 1992; 38: 347-350 [PMID: 1376705 DOI: 10.1016/S0016-5107(92)70430-1]

12

George S, Kulkarni AA, Stevens G, Forsmark CE, Draganov P. Role of osmolality of contrast media in the development of post-ERCP pancreatitis: a metanalysis. Dig Dis Sci 2004; 49: 503-508 [PMID: 15139506 DOI: 10.1023/B:DDAS.0000020511.98230.20]

13

Zhao ZH, Hu LH, Ren HB, Zhao AJ, Qian YY, Sun XT, Su S, Zhu SG, Yu J, Zou WB, Guo XR, Wang L, Li ZS, Liao Z. Incidence and risk factors for post-ERCP pancreatitis in chronic pancreatitis. Gastrointest Endosc 2017; 86: 519-524. e1 [PMID: 28062312 DOI: 10.1016/j.gie.2016.12.020]

14

Sun HL, Han B, Zhai HP, Cheng XH, Ma K. Rectal NSAIDs for the prevention of post-ERCP pancreatitis: a meta-analysis of randomized controlled trials. Surgeon 2014; 12: 141-147 [PMID: 24332479 DOI: 10.1016/j.surge.2013.10.010]

15

Vandervoort J, Soetikno RM, Tham TC, Wong RC, Ferrari AP Jr, Montes H, Roston AD, Slivka A, Lichtenstein DR, Ruymann FW, Van Dam J, Hughes M, Carr-Locke DL. Risk factors for complications after performance of ERCP. Gastrointest Endosc 2002; 56: 652-656 [PMID: 12397271 DOI: 10.1067/mge.2002.129086]

16

Ferreira Ade F, Bartelega JA, Urbano HC, de Souza IK. Acute pancreatitis gravity predictive factors: which and when to use them? Arq Bras Cir Dig 2015; 28: 207-211 [PMID: 26537149 DOI: 10.1590/S0102-67202015000300016]

17

Kochar B, Akshintala VS, Afghani E, Elmunzer BJ, Kim KJ, Lennon AM, Khashab MA, Kalloo AN, Singh VK. Incidence, severity, and mortality of post-ERCP pancreatitis: a systematic review by using randomized, controlled trials. Gastrointest Endosc 2015; 81: 143-149. e9 [PMID: 25088919 DOI: 10.1016/j.gie.2014.06.045]

18

Yaghoobi M, Pauls Q, Durkalski V, Romagnuolo J, Fogel EL, Tarnasky PR, Aliperti G, Freeman ML, Kozarek RA, Jamidar PA, Wilcox CM, Elta GH, Hawes RH, Wood-Williams A, Cotton PB. Incidence and predictors of post-ERCP pancreatitis in patients with suspected sphincter of Oddi dysfunction undergoing biliary or dual sphincterotomy: results from the EPISOD prospective multicenter randomized sham-controlled study. Endoscopy 2015; 47: 884-890 [PMID: 26165739 DOI: 10.1055/s-0034-1392418]

19

Dumonceau JM, Andriulli A, Elmunzer BJ, Mariani A, Meister T, Deviere J, Marek T, Baron TH, Hassan C, Testoni PA, Kapral C; European Society of Gastrointestinal Endoscopy. Prophylaxis of post-ERCP pancreatitis: European Society of Gastrointestinal Endoscopy (ESGE) Guideline - updated June 2014. Endoscopy 2014; 46: 799-815 [PMID: 25148137 DOI: 10.1055/s-0034-1377875]

20

Freeman ML, Nelson DB, Sherman S, Haber GB, Herman ME, Dorsher PJ, Moore JP, Fennerty MB, Ryan ME, Shaw MJ, Lande JD, Pheley AM. Complications of endoscopic biliary sphincterotomy. N Engl J Med 1996; 335: 909-918 [PMID: 8782497 DOI: 10.1056/NEJM199609263351301]

21

Freeman ML, DiSario JA, Nelson DB, Fennerty MB, Lee JG, Bjorkman DJ, Overby CS, Aas J, Ryan ME, Bochna GS, Shaw MJ, Snady HW, Erickson RV, Moore JP, Roel JP. Risk factors for post-ERCP pancreatitis: a prospective, multicenter study. Gastrointest Endosc 2001; 54: 425-434 [PMID: 11577302 DOI: 10.1067/mge.2001.117550]

22

Pan Y, Zhao L, Leung J, Zhang R, Luo H, Wang X, Liu Z, Wan B, Tao Q, Yao S, Hui N, Fan D, Wu K, Guo X. Appropriate time for selective biliary cannulation by trainees during ERCP--a randomized trial. Endoscopy 2015; 47: 688-695 [PMID: 25750038 DOI: 10.1055/s-0034-1391564]

23

Morales SJ, Sampath K, Gardner TB. A Review of Prevention of Post-ERCP Pancreatitis. Gastroenterol Hepatol (N Y) 2018; 14: 286-292 [PMID: 29991936]

24

Dumonceau JM, Kapral C, Aabakken L, Papanikolaou IS, Tringali A, Vanbiervliet G, Beyna T, Dinis-Ribeiro M, Hritz I, Mariani A, Paspatis G, Radaelli F, Lakhtakia S, Veitch AM, van Hooft JE. ERCP-related adverse events: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy 2020; 52: 127-149 [PMID: 31863440 DOI: 10.1055/a-1075-4080]

25

Matsubayashi CO, Ribeiro IB, de Moura DTH, Brunaldi VO, Bernardo WM, Hathorn KE, de Moura EGH. Is Endoscopic Balloon Dilation Still Associated With Higher Rates of Pancreatitis? Pancreas 2020; 49: 158-174 [PMID: 32049951 DOI: 10.1097/MPA.0000000000001489]

26

Badalov N, Tenner S, Baillie J. The Prevention, recognition and treatment of post-ERCP pancreatitis. JOP 2009; 10: 88-97 [PMID: 19287099]

27

Testoni PA, Mariani A, Giussani A, Vailati C, Masci E, Macarri G, Ghezzo L, Familiari L, Giardullo N, Mutignani M, Lombardi G, Talamini G, Spadaccini A, Briglia R, Piazzi L; SEIFRED Group. Risk factors for post-ERCP pancreatitis in high- and low-volume centers and among expert and non-expert operators: a prospective multicenter study. Am J Gastroenterol 2010; 105: 1753-1761 [PMID: 20372116 DOI: 10.1038/ajg.2010.136]

28

Mariani A, Giussani A, Di Leo M, Testoni S, Testoni PA. Guidewire biliary cannulation does not reduce post-ERCP pancreatitis compared with the contrast injection technique in low-risk and high-risk patients. Gastrointest Endosc 2012; 75: 339-346 [PMID: 22075192 DOI: 10.1016/j.gie.2011.09.002]

29

Crockett SD, Wani S, Gardner TB, Falck-Ytter Y, Barkun AN; American Gastroenterological 30




https://dx.doi.org/10.1016/j.clinre.2019.12.009


https://dx.doi.org/10.1016/S0016-5107(92)70430-1


https://dx.doi.org/10.1023/B:DDAS.0000020511.98230.20




https://dx.doi.org/10.1016/j.surge.2013.10.010


https://dx.doi.org/10.1067/mge.2002.129086


https://dx.doi.org/10.1590/S0102-67202015000300016




https://dx.doi.org/10.1055/s-0034-1392418


https://dx.doi.org/10.1055/s-0034-1377875






https://dx.doi.org/10.1055/s-0034-1391564



https://dx.doi.org/10.1055/a-1075-4080


https://dx.doi.org/10.1097/MPA.0000000000001489








Association Institute Clinical Guidelines Committee. American Gastroenterological Association Institute Guideline on Initial Management of Acute Pancreatitis. Gastroenterology 2018; 154: 1096-1101 [PMID: 29409760 DOI: 10.1053/j.gastro.2018.01.032]Tenner S, Baillie J, DeWitt J, Vege SS; American College of Gastroenterology. American College of Gastroenterology guideline: management of acute pancreatitis. Am J Gastroenterol 2013; 108: 1400-15; 1416 [PMID: 23896955 DOI: 10.1038/ajg.2013.218]

31

Swaroop VS, Chari ST, Clain JE. Severe acute pancreatitis. JAMA 2004; 291: 2865-2868 [PMID: 15199038 DOI: 10.1001/jama.291.23.2865]

32

Lee PJ, Papachristou GI. Management of Severe Acute Pancreatitis. Curr Treat Options Gastroenterol 2020; 1-12 [PMID: 33230385 DOI: 10.1007/s11938-020-00322-x]

33

Manoharan D, Srivastava DN, Gupta AK, Madhusudhan KS. Complications of endoscopic retrograde cholangiopancreatography: an imaging review. Abdom Radiol (NY) 2019; 44: 2205-2216 [PMID: 30809695 DOI: 10.1007/s00261-019-01953-0]

34

Boxhoorn L, Voermans RP, Bouwense SA, Bruno MJ, Verdonk RC, Boermeester MA, van Santvoort HC, Besselink MG. Acute pancreatitis. Lancet 2020; 396: 726-734 [PMID: 32891214 DOI: 10.1016/S0140-6736(20)31310-6]

35

Freeman ML, Nelson DB, Sherman S, Haber GB, Fennerty MB, DiSario JA, Ryan ME, Kortan PP, Dorsher PJ, Shaw MJ, Herman ME, Cunningham JT, Moore JP, Silverman WB, Imperial JC, Mackie RD, Jamidar PA, Yakshe PN, Logan GM, Pheley AM. Same-day discharge after endoscopic biliary sphincterotomy: observations from a prospective multicenter complication study. The Multicenter Endoscopic Sphincterotomy (MESH) Study Group. Gastrointest Endosc 1999; 49: 580-586 [PMID: 10228255 DOI: 10.1016/S0016-5107(99)70385-8]

36

Ho KY, Montes H, Sossenheimer MJ, Tham TC, Ruymann F, Van Dam J, Carr-Locke DL. Features that may predict hospital admission following outpatient therapeutic ERCP. Gastrointest Endosc 1999; 49: 587-592 [PMID: 10228256 DOI: 10.1016/S0016-5107(99)70386-X]

37

Sutton VR, Hong MK, Thomas PR. Using the 4-hour Post-ERCP amylase level to predict post-ERCP pancreatitis. JOP 2011; 12: 372-376 [PMID: 21737899]

38

Thomas PR, Sengupta S. Prediction of pancreatitis following endoscopic retrograde cholangiopancreatography by the 4-h post procedure amylase level. J Gastroenterol Hepatol 2001; 16: 923-926 [PMID: 11555108 DOI: 10.1046/j.1440-1746.2001.02547.x]

39

Testoni PA, Bagnolo F, Caporuscio S, Lella F. Serum amylase measured four hours after endoscopic sphincterotomy is a reliable predictor of postprocedure pancreatitis. Am J Gastroenterol 1999; 94: 1235-1241 [PMID: 10235200 DOI: 10.1111/j.1572-0241.1999.01072.x]

40

Testoni PA, Bagnolo F. Pain at 24 hours associated with amylase levels greater than 5 times the upper normal limit as the most reliable indicator of post-ERCP pancreatitis. Gastrointest Endosc 2001; 53: 33-39 [PMID: 11154486 DOI: 10.1067/mge.2001.111390]

41

Shi H, Chen S, Swar G, Wang Y, Ying M. Carbon dioxide insufflation during endoscopic retrograde cholangiopancreatography: a review and meta-analysis. Pancreas 2013; 42: 1093-1100 [PMID: 23867366 DOI: 10.1097/MPA.0b013e3182909da5]

42

Bretthauer M, Seip B, Aasen S, Kordal M, Hoff G, Aabakken L. Carbon dioxide insufflation for more comfortable endoscopic retrograde cholangiopancreatography: a randomized, controlled, double-blind trial. Endoscopy 2007; 39: 58-64 [PMID: 17252462 DOI: 10.1055/s-2006-945036]

43

Passos ML, Ribeiro IB, de Moura DTH, Korkischko N, Silva GLR, Franzini TP, Bernando WM, de Moura EGH. Efficacy and safety of carbon dioxide insufflation versus air insufflation during endoscopic retrograde cholangiopancreatography in randomized controlled trials: a systematic review and meta-analysis. Endosc Int Open 2019; 7: E487-E497 [PMID: 31041365 DOI: 10.1055/a-0854-3739]

44

Olaiya B, Adler DG. Intestinal perforations after endoscopic retrograde cholangiopancreatography in the United States: a 16-year study using the national inpatient sample. Minerva Gastroenterol Dietol 2020 [PMID: 32492997 DOI: 10.23736/S1121-421X.20.02718-X]

45

Borazan E, Konduk BT. Comparison of early and delayed diagnosis of mortality in ERCP perforations: A high-volume patient experience. Ulus Travma Acil Cerrahi Derg 2020; 26: 746-753 [PMID: 32946098 DOI: 10.14744/tjtes.2020.61289]

46

Basurto Ona X, Rigau Comas D, Urrútia G. Opioids for acute pancreatitis pain. Cochrane Database Syst Rev 2013; CD009179 [PMID: 23888429 DOI: 10.1002/14651858.CD009179.pub2]

47

Mederos MA, Reber HA, Girgis MD. Acute Pancreatitis: A Review. JAMA 2021; 325: 382-390 [PMID: 33496779 DOI: 10.1001/jama.2020.20317]

48

Messallam AA, Body CB, Berger S, Sakaria SS, Chawla S. Impact of early aggressive fluid resuscitation in acute pancreatitis. Pancreatology 2021; 21: 69-73 [PMID: 33257225 DOI: 10.1016/j.pan.2020.11.006]

49

Zhang J, Zhu S, Tan D, Ma A, Yang Y, Xu J. A meta-analysis of early oral refeeding and quickly increased diet for patients with mild acute pancreatitis. Saudi J Gastroenterol 2019; 25: 14-19 [PMID: 30226482 DOI: 10.4103/sjg.SJG_240_18]

50

Lella F, Bagnolo F, Colombo E, Bonassi U. A simple way of avoiding post-ERCP pancreatitis. Gastrointest Endosc 2004; 59: 830-834 [PMID: 15173796 DOI: 10.1016/S0016-5107(04)00363-3]

51

Artifon EL, Sakai P, Cunha JE, Halwan B, Ishioka S, Kumar A. Guidewire cannulation reduces risk of post-ERCP pancreatitis and facilitates bile duct cannulation. Am J Gastroenterol 2007; 102: 2147-2153 [PMID: 17581267 DOI: 10.1111/j.1572-0241.2007.01378.x]

52

Bailey AA, Bourke MJ, Williams SJ, Walsh PR, Murray MA, Lee EY, Kwan V, Lynch PM. A 53






https://dx.doi.org/10.1001/jama.291.23.2865




https://dx.doi.org/10.1007/s00261-019-01953-0


https://dx.doi.org/10.1016/S0140-6736(20)31310-6


https://dx.doi.org/10.1016/S0016-5107(99)70385-8


https://dx.doi.org/10.1016/S0016-5107(99)70386-X









https://dx.doi.org/10.1097/MPA.0b013e3182909da5


https://dx.doi.org/10.1055/s-2006-945036


https://dx.doi.org/10.1055/a-0854-3739


https://dx.doi.org/10.23736/S1121-421X.20.02718-X


https://dx.doi.org/10.14744/tjtes.2020.61289




https://dx.doi.org/10.1001/jama.2020.20317


https://dx.doi.org/10.1016/j.pan.2020.11.006


https://dx.doi.org/10.4103/sjg.SJG_240_18


https://dx.doi.org/10.1016/S0016-5107(04)00363-3





prospective randomized trial of cannulation technique in ERCP: effects on technical success and post-ERCP pancreatitis. Endoscopy 2008; 40: 296-301 [PMID: 18389448 DOI: 10.1055/s-2007-995566]Testoni PA, Mariani A, Aabakken L, Arvanitakis M, Bories E, Costamagna G, Devière J, Dinis-Ribeiro M, Dumonceau JM, Giovannini M, Gyokeres T, Hafner M, Halttunen J, Hassan C, Lopes L, Papanikolaou IS, Tham TC, Tringali A, van Hooft J, Williams EJ. Papillary cannulation and sphincterotomy techniques at ERCP: European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline. Endoscopy 2016; 48: 657-683 [PMID: 27299638 DOI: 10.1055/s-0042-108641]

54

Tse F, Yuan Y, Moayyedi P, Leontiadis GI. Guide wire-assisted cannulation for the prevention of post-ERCP pancreatitis: a systematic review and meta-analysis. Endoscopy 2013; 45: 605-618 [PMID: 23807804 DOI: 10.1055/s-0032-1326640]

55

Sasahira N, Kawakami H, Isayama H, Uchino R, Nakai Y, Ito Y, Matsubara S, Ishiwatari H, Uebayashi M, Yagioka H, Togawa O, Toda N, Sakamoto N, Kato J, Koike K. Early use of double-guidewire technique to facilitate selective bile duct cannulation: the multicenter randomized controlled EDUCATION trial. Endoscopy 2015; 47: 421-429 [PMID: 25590186 DOI: 10.1055/s-0034-1391228]

56

Madácsy L, Kurucsai G, Joó I, Gódi S, Fejes R, Székely A. Rescue ERCP and insertion of a small-caliber pancreatic stent to prevent the evolution of severe post-ERCP pancreatitis: a case-controlled series. Surg Endosc 2009; 23: 1887-1893 [PMID: 19057957 DOI: 10.1007/s00464-008-0199-z]

57

Kerdsirichairat T, Attam R, Arain M, Bakman Y, Radosevich D, Freeman M. Urgent ERCP with pancreatic stent placement or replacement for salvage of post-ERCP pancreatitis. Endoscopy 2014; 46: 1085-1094 [PMID: 25216326 DOI: 10.1055/s-0034-1377750]

58

Karjula H, Nordblad Schmidt P, Mäkelä J, Liisanantti JH, Ohtonen P, Saarela A. Prophylactic pancreatic duct stenting in severe acute necrotizing pancreatitis: a prospective randomized study. Endoscopy 2019; 51: 1027-1034 [PMID: 30895583 DOI: 10.1055/a-0865-1960]

59

Shaygan-Nejad A, Masjedizadeh AR, Ghavidel A, Ghojazadeh M, Khoshbaten M. Aggressive hydration with Lactated Ringer's solution as the prophylactic intervention for postendoscopic retrograde cholangiopancreatography pancreatitis: A randomized controlled double-blind clinical trial. J Res Med Sci 2015; 20: 838-843 [PMID: 26759569 DOI: 10.4103/1735-1995.170597]

60

Vadalà di Prampero SF, Faleschini G, Panic N, Bulajic M. Endoscopic and pharmacological treatment for prophylaxis against postendoscopic retrograde cholangiopancreatography pancreatitis: a meta-analysis and systematic review. Eur J Gastroenterol Hepatol 2016; 28: 1415-1424 [PMID: 27580214 DOI: 10.1097/MEG.0000000000000734]

61

Kubiliun NM, Adams MA, Akshintala VS, Conte ML, Cote GA, Cotton PB, Dumonceau JM, Elta GH, Fogel EL, Freeman ML, Lehman GA, Naveed M, Romagnuolo J, Scheiman JM, Sherman S, Singh VK, Elmunzer BJ; United States Cooperative for Outcomes Research in Endoscopy (USCORE). Evaluation of Pharmacologic Prevention of Pancreatitis After Endoscopic Retrograde Cholangiopancreatography: A Systematic Review. Clin Gastroenterol Hepatol 2015; 13: 1231-9; quiz e70 [PMID: 25579870 DOI: 10.1016/j.cgh.2014.11.038]

62

Serrano JPR, de Moura DTH, Bernardo WM, Ribeiro IB, Franzini TP, de Moura ETH, Brunaldi VO, Salesse MT, Sakai P, De Moura EGH. Nonsteroidal anti-inflammatory drugs vs placebo for post-endoscopic retrograde cholangiopancreatography pancreatitis: a systematic review and meta-analysis. Endosc Int Open 2019; 07: E477-E486 [PMID: 30957004 DOI: 10.1055/a-0862-0215]

63

Akbar A, Abu Dayyeh BK, Baron TH, Wang Z, Altayar O, Murad MH. Rectal nonsteroidal anti-inflammatory drugs are superior to pancreatic duct stents in preventing pancreatitis after endoscopic retrograde cholangiopancreatography: a network meta-analysis. Clin Gastroenterol Hepatol 2013; 11: 778-783 [PMID: 23376320 DOI: 10.1016/j.cgh.2012.12.043]

64

Ishiwatari H, Urata T, Yasuda I, Matsusaki S, Hisai H, Kawakami H, Ono M, Iwashita T, Doi S, Kawakubo K, Hayashi T, Sonoda T, Sakamoto N, Kato J. No Benefit of Oral Diclofenac on Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis. Dig Dis Sci 2016; 61: 3292-3301 [PMID: 27447477 DOI: 10.1007/s10620-016-4251-x]

65

Iqbal U, Siddique O, Khara HS, Khan MA, Haq KF, Siddiqui MA, Solanki S, Zuchelli TE, Shellenberger MJ, Birk JW. Post-endoscopic retrograde cholangiopancreatography pancreatitis prevention using topical epinephrine: systematic review and meta-analysis. Endosc Int Open 2020; 8: E1061-E1067 [PMID: 32743060 DOI: 10.1055/a-1190-3777]

66

Chen B, Fan T, Wang CH. A meta-analysis for the effect of prophylactic GTN on the incidence of post-ERCP pancreatitis and on the successful rate of cannulation of bile ducts. BMC Gastroenterol 2010; 10: 85 [PMID: 20673365 DOI: 10.1186/1471-230X-10-85]

67

Sotoudehmanesh R, Eloubeidi MA, Asgari AA, Farsinejad M, Khatibian M. A randomized trial of rectal indomethacin and sublingual nitrates to prevent post-ERCP pancreatitis. Am J Gastroenterol 2014; 109: 903-909 [PMID: 24513806 DOI: 10.1038/ajg.2014.9]

68

Tomoda T, Kato H, Ueki T, Akimoto Y, Hata H, Fujii M, Harada R, Ogawa T, Wato M, Takatani M, Matsubara M, Kawai Y, Okada H. Combination of Diclofenac and Sublingual Nitrates Is Superior to Diclofenac Alone in Preventing Pancreatitis After Endoscopic Retrograde Cholangiopancreatography. Gastroenterology 2019; 156: 1753-1760. e1 [PMID: 30772342 DOI: 10.1053/j.gastro.2019.01.267]

69

Andriulli A, Leandro G, Federici T, Ippolito A, Forlano R, Iacobellis A, Annese V. Prophylactic administration of somatostatin or gabexate does not prevent pancreatitis after ERCP: an updated meta-analysis. Gastrointest Endosc 2007; 65: 624-632 [PMID: 17383459 DOI: 10.1016/j.gie.2006.10.030]

70

Qin X, Lei WS, Xing ZX, Shi F. Prophylactic effect of somatostatin in preventing Post-ERCP pancreatitis: an updated meta-analysis. Saudi J Gastroenterol 2015; 21: 372-378 [PMID: 26655132

71


https://dx.doi.org/10.1055/s-2007-995566


https://dx.doi.org/10.1055/s-0042-108641


https://dx.doi.org/10.1055/s-0032-1326640


https://dx.doi.org/10.1055/s-0034-1391228




https://dx.doi.org/10.1055/s-0034-1377750


https://dx.doi.org/10.1055/a-0865-1960


https://dx.doi.org/10.4103/1735-1995.170597






https://dx.doi.org/10.1055/a-0862-0215






https://dx.doi.org/10.1055/a-1190-3777


https://dx.doi.org/10.1186/1471-230X-10-85










DOI: 10.4103/1319-3767.167187]Bai Y, Gao J, Zou DW, Li ZS. Prophylactic octreotide administration does not prevent post-endoscopic retrograde cholangiopancreatography pancreatitis: a meta-analysis of randomized controlled trials. Pancreas 2008; 37: 241-246 [PMID: 18815543 DOI: 10.1097/MPA.0b013e31816c90a1]

72

Omata F, Deshpande G, Tokuda Y, Takahashi O, Ohde S, Carr-Locke DL, Jacobs JL, Mine T, Fukui T. Meta-analysis: somatostatin or its long-acting analogue, octreotide, for prophylaxis against post-ERCP pancreatitis. J Gastroenterol 2010; 45: 885-895 [PMID: 20373114 DOI: 10.1007/s00535-010-0234-4]

73

Seta T, Noguchi Y. Protease inhibitors for preventing complications associated with ERCP: an updated meta-analysis. Gastrointest Endosc 2011; 73: 700-706. e1-2 [PMID: 21145053 DOI: 10.1016/j.gie.2010.09.022]

74

Yu G, Li S, Wan R, Wang X, Hu G. Nafamostat mesilate for prevention of post-ERCP pancreatitis: a meta-analysis of prospective, randomized, controlled trials. Pancreas 2015; 44: 561-569 [PMID: 25822153 DOI: 10.1097/MPA.0000000000000310]

75

Yuhara H, Ogawa M, Kawaguchi Y, Igarashi M, Shimosegawa T, Mine T. Pharmacologic prophylaxis of post-endoscopic retrograde cholangiopancreatography pancreatitis: protease inhibitors and NSAIDs in a meta-analysis. J Gastroenterol 2014; 49: 388-399 [PMID: 23720090 DOI: 10.1007/s00535-013-0834-x]

76

Chen S, Shi H, Zou X, Luo H. Role of ulinastatin in preventing post-endoscopic retrograde cholangiopancreatography pancreatitis: the Emperor's New Clothes or Aladdin's Magic Lamp? Pancreas 2010; 39: 1231-1237 [PMID: 20531245 DOI: 10.1097/MPA.0b013e3181dc67e7]

77

https://dx.doi.org/10.4103/1319-3767.167187


https://dx.doi.org/10.1097/MPA.0b013e31816c90a1


https://dx.doi.org/10.1007/s00535-010-0234-4








https://dx.doi.org/10.1097/MPA.0b013e3181dc67e7


World Journal of



MINIREVIEWS

RON in hepatobiliary and pancreatic cancers: Pathogenesis and potential therapeutic targets

Shao-Long Chen, Guo-Ping Wang, Dan-Rong Shi, Shu-Hao Yao, Ke-Da Chen, Hang-Ping Yao

ORCID number: Shao-Long Chen 0000-0002-2365-7430; Guo-Ping Wang 0000-0002-1300-2430; Dan-Rong Shi 0000-0002-6364-2321; Shu-Hao Yao 0000-0003-3408-5615; Ke-Da Chen 0000-0002-9469-0991; Hang-Ping Yao 0000-0001-6742-7074.

Author contributions: Chen SL, Wang GP, Shi DR, Yao SH, Chen KD, and Yao HP discussed the necessity of writing this manuscript; Chen SL and Yao HP wrote the original draft; Chen SL, Wang GP, Shi DR, Yao SH, Chen KD, and Yao HP reviewed the draft with detailed comments; Chen SL and Yao HP made revisions to the manuscript; all authors read and approved the final manuscript for submission.

Supported by National Natural Sciences Foundation of China, No. 81872883; and Zhejiang Major Medical Health & Sciences Technology Foundation Projects, No. WKJ-ZJ-13.

Conflict-of-interest statement: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

Open-Access: This article is an open-access article that was selected by an in-house editor and

Shao-Long Chen, Ke-Da Chen, Shulan International Medical College, Zhejiang Shuren University, Hangzhou 310000, Zhejiang Province, China

Guo-Ping Wang, Department of Surgical Oncology, The Second Affiliated Hospital of Zhejiang Chinese Medical University, Hangzhou 310000, Zhejiang Province, China

Dan-Rong Shi, Hang-Ping Yao, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310000, Zhejiang Province, China

Shu-Hao Yao, Department of Stomatology, Wenzhou Medical University Renji College, Wenzhou 325035, Zhejiang Province, China

Corresponding author: Hang-Ping Yao, PhD, Professor, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, The First Affiliated Hospital, Zhejiang University School of Medicine, No. 79 Qingchun Road, Shangcheng District, Hangzhou 310000, Zhejiang Province, China. [email protected]

AbstractThe receptor protein tyrosine kinase RON belongs to the c-MET proto-oncogene family. Research has shown that RON has a role in cancer pathogenesis, which places RON on the frontline of the development of novel cancer therapeutic strategies. Hepatobiliary and pancreatic (HBP) cancers have a poor prognosis, being reported as having higher rates of cancer-related death. Therefore, to combat these malignant diseases, the mechanism underlying the aberrant expression and signaling of RON in HBP cancer pathogenesis, and the development of RON as a drug target for therapeutic intervention should be investigated. Abnormal RON expression and signaling have been identified in HBP cancers, and also act as tumorigenic determinants for HBP cancer malignant behaviors. In addition, RON is emerging as an important mediator of the clinical prognosis of HBP cancers. Thus, not only is RON significant in HBP cancers, but also RON-targeted therapeutics could be developed to treat these cancers, for example, therapeutic monoclonal antibodies and small-molecule inhibitors. Among them, antibody-drug conjugates have become increasingly popular in current research and their potential as novel anti-cancer biotherapeutics will be determined in future clinical trials.



http://orcid.org/0000-0002-2365-7430

http://orcid.org/0000-0002-2365-7430

http://orcid.org/0000-0002-1300-2430

http://orcid.org/0000-0002-1300-2430

http://orcid.org/0000-0002-6364-2321

http://orcid.org/0000-0002-6364-2321

http://orcid.org/0000-0003-3408-5615

http://orcid.org/0000-0003-3408-5615

http://orcid.org/0000-0002-9469-0991

http://orcid.org/0000-0002-9469-0991

http://orcid.org/0000-0001-6742-7074

http://orcid.org/0000-0001-6742-7074


Chen SL et al. RON in hepatobiliary and pancreatic cancers


fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/Licenses/by-nc/4.0/




Peer-review report’s scientific quality classificationGrade A (Excellent): 0 Grade B (Very good): B Grade C (Good): C Grade D (Fair): 0 Grade E (Poor): 0

Received: December 31, 2020 Peer-review started: December 31, 2020 First decision: February 23, 2021 Revised: March 4, 2021 Accepted: April 9, 2021 Article in press: April 9, 2021 Published online: May 28, 2021

P-Reviewer: Gomes A, Tangkawattana S S-Editor: Gao CC L-Editor: Wang TQ P-Editor: Liu JH

Key Words: RON; Signal transduction; Hepatobiliary; Pancreatic neoplasms; Molecular targeted therapy


Core Tip: The role of RON in cancer pathogenesis has received increasing research attention. Hepatobiliary and pancreatic (HBP) cancers have a poor prognosis, being reported as having higher rates of cancer-related death because of their high rates of recurrence, metastasis, and invasiveness, and their lack of sensitivity to chemotherapy. In this review, we discuss how RON functions in HBP cancer pathogenesis, as well as its potential role as a therapeutic target in HBP cancers.

Citation: Chen SL, Wang GP, Shi DR, Yao SH, Chen KD, Yao HP. RON in hepatobiliary and pancreatic cancers: Pathogenesis and potential therapeutic targets. World J Gastroenterol 2021; 27(20): 2507-2520URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2507.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2507

INTRODUCTIONRON receptor tyrosine kinase (RTK; also known as MST1R) was first identified in 1993 in a cDNA library from human epithelial cells[1]. RON belongs to the family of c-MET proto-oncogenes[2]. This RTK family only has two members, RON and Met, which share only 34% overall homology; however, the tyrosine kinase region of the receptors is quite similar at 80% homology[3]. In 1994, a mouse cDNA was cloned that encoded a homolog of RON, which was termed stem cell-derived tyrosine kinase receptor[4]. RON is located at human chromosome 3p21 and this gene shows high conservation in different species, including xenopus, zebrafish, chicken, cats, human, and mouse[4-11]. The RON receptor is initially synthesized as a biologically inactive single-chain precursor (pro-RON), then cleaved into a 145 kDa β-chain and a 35 kDa extracellular alpha chain, which are linked by a disulfide bond, forming the mature receptor. In 1994, the physiological ligand of RON was identified as macrophage-stimulating protein (MSP) [also called hepatocyte growth factor (HGF)-like protein], establishing the MSP-RON signaling system[12-15]. MSP is a member of the plasminogen-related kringle protein family[16,17]. The human MSP gene is also located at chromosome 3p21 and is evolutionarily conserved in different species, similar to RON. The main source of MSP is hepatocytes and MSP circulates in the blood as pro-MSP, which is a biologically inactive single-chain precursor. After subsequent proteolytic conversion, the active mature MSP consists of the disulfide-linked alpha subunit and β-chain. The RON receptor high affinity binding site is in the β-chain and RON activity is regulated by the alpha chain[18]. The binding of MSP to RON induces RON dimerization, which activates multiple downstream signaling pathways, leading to RON-mediated cell growth, survival, and invasiveness[19,20].

In the last two decades, increased research has focused on the tumorigenic and therapeutic roles of RON signaling. Although there have been few studies concerning pathology-related changes in MSP expression, numerous studies regarding aberrant RON activation in various types of tumors have been published, including RON protein overexpression[21-28], oncogenic variant generation[29-39], and persistent activation of downstream signaling pathways[21-39]. In addition, tumorigenic progression and malignancy are associated with functional crosstalk between signaling proteins and RON. In clinical application, increased RON expression can be used for prognostic evaluation of patient survival and disease progression. Hepato-biliary and pancreatic (HBP) cancers have a poor outcome, with high rates of cancer-related death because of their high incidences of recurrence, metastasis, and invasiveness, and their lack of sensitivity to chemotherapy[40]. Complete surgical resection remains the most effective treatment for HBP cancers[40]. Among these cancers, the 5-year survival rate of liver cancer is approximately 30%, whereas in biliary tract cancer and pancreatic cancer, it is less than 30% and less than 10%, respectively[41]. The high death rate of pancreatic cancer is caused by the lack of early








diagnosis and effective treatment. In pancreatic cancer, most cases are diagnosed when the disease is already at an advanced stage, and only 20% or less of patients present with potentially curable localized tumors amenable to surgical extirpation[42]. Thus, the identification of a novel potential therapeutic strategy is urgently required. Growing evidence suggests a close relationship between HBP cancers and RON dysregulation[24,43,44]. Thus, the present review primarily focuses on the role of RON in the pathogenesis of cancer, especially HBP cancers. Moreover, we summarize the latest progress in the development of strategies targeting RON as potential HBP cancer therapy.

ROLES OF RON AND C-MET IN CARCINOGENESISRON and c-MET, both of which are members of the semaphorin family of transmembrane receptor tyrosine kinases, share similar structural and biochemical properties[45]. The proteins exist as heterodimers comprising extracellular and transmembrane chains that are linked by disulfide bonds. The RON and c-MET extracellular sequences possess very similar functional domains, including SEMA, which regulates phosphorylation, receptor dimerization, and ligand binding. RON and c-MET are activated by their respective ligands: MSP for RON and HGF for c-MET. c-MET and HGF are expressed in a variety of cell and tissue types. Contra-stingly, RON is restricted tightly to epithelial origin cells, whereas liver cells are the major source of its ligand, MSP[46]. Independent or ligand-dependent activation of RON and c-MET induces matrix invasion, cell migration, and cell proliferation, all of which are crucial for embryogenesis, wound healing, and tumorigenesis.

Increasing evidence has identified the role of RON and c-Met in the pathogenesis of cancer[47]. For example, c-MET and RON overexpression was observed in a variety of primary and metastatic tumors, leading to the activation of aberrant downstream signaling, which contributes to cancer development and progression. Moreover, clinical studies have validated that increased expression of RON and c-MET is a prognostic factor to predict the survival rate and disease progression in certain patients with cancer[48,49]. Moreover, activation of RON and c-MET promotes a cancer cell malignant phenotype. Increased RON and c-MET expression drives tumor cells to undergo epithelial to mesenchymal transition (EMT), which is characterized by epithelial feature loss and the gain of mesenchymal characteristics[12,50]. Increased c-MET and RON expression also contributes to acquired chemoresistance[51]. Given the above role of the increased expression of c-MET and RON in cancer pathogenesis, targeting RON and c-MET represents a promising cancer therapy strategy.

RON ACTIVATION AND SIGNALING PATHWAY MECHANISMSEpithelial cells in the skin, adrenal gland, bone, brain, kidney, gut, lung, and liver express low levels of RON[12]. The action of RON plays a key role in the motility of epithelial cells, enhancement of adhesion, sperm motility in the epididymis, and embryonic development, as well as the regulation of inflammatory responses[52]. Under physiological conditions, the main cause of RON activation is stimulation of its ligand, MSP[12]. Moreover, three other biochemical events activate RON in tumors: RON overexpression, generation of oncogenic RON variants, and RON transactivation (Figure 1). The RON receptor consists of three essential regions: The extracellular domain that recognizes its ligand, the transmembrane domain that anchors the receptor to the membrane, and the intracellular domain that exerts the kinase activity (Figure 1)[53]. The first step for the activation of RON is dimerization at the cell surface, which is caused by the binding of MSP to the extracellular domain containing the specific ligand-binding site, likely resulting in a conformational change in the RON receptor. This activation leads to autophosphorylation at two tyrosine residues (Tyr1238 and Tyr1239) located in the A-loop (Phe1227-Pro1250) of the kinase domain. Phosphorylation of these regulatory residues leads to tyrosine kinase function activation, inducing further phosphorylation of residues Tyr1353 and Tyr1360 located in the C-terminal docking site. This then recruits the cytoplasmic molecules growth factor receptor-bound protein 2 (GRB2) and Son of Sevenless. In addition, the ubiquitin ligase, casitas B-lineage lymphoma (CBL), binds to the docking site to act as a negative modulator.



Figure 1 Mechanisms of RON activation and downstream signaling pathways. Classically, macrophage-stimulating protein (MSP) activates RON. In cancer, RON activation is induced by overexpression, splicing or truncation, and transactivation. The RON receptor consists of three regions including the extracellular domain, the transmembrane domain, as well as the intracellular domain. MSP binding to the extracellular domain leads to autophosphorylation of several tyrosine residues in the kinase activation loop or in the C-terminal tail, resulting in the activation of many biological activities, including increased proliferation/survival, motile-invasive activity, and chemoresistance. MSP: Macrophage-stimulating protein; SOS: Son of Sevenless; GRB2: Growth factor receptor-bound protein 2; CBL: Casitas B-lineage lymphoma; 14-3-3: Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein; PI-3K-AKT: Phosphatidylinositol-4,5-Bisphosphate 3 kinase- protein kinase B; HIF: Hypoxia-inducible factor; RAS-MAPK: RAS-mitogen-activated protein kinase; ERK: Extracellular regulated kinase; RSK: Ribosomal protein S6 kinase; mTOR: Mechanistic target of rapamycin.

The interaction of RON with adaptor proteins, such as β-arrestin-1 and GRB2, represents the first step in the bridging of downstream signaling cascades and RON activation. Via its C-terminal docking site, RON interacts with a variety of cytoplasmic effector molecules, such as phospholipase C gamma, phosphatidylinositol-4,5-Bisphosphate 3-kinase (PI-3 kinase), Src (SRC proto-oncogene, non-receptor tyrosine kinase), tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein (14-3-3), CBL protooncogene (c-Cbl), heat shock protein family A (Hsp70) member 8 (HSC70), integrin-β4, plectin, and protein phosphatase 1. The classical PI-3 kinase- protein kinase B (PI-3K-AKT) and RAS-mitogen-activated protein kinase (RAS-MAPK) pathways are triggered by the interaction of RON’s docking site with downstream signaling proteins. PI-3K-AKT and RAS-MAPK pathways mediate many biological activities, including increased proliferation and survival, EMT, motile-invasive activity, and chemoresistance. The signaling pathways of RON also play a part in regulating tumorigenic activity. Among them, PI-3K-AKT and RAS-MAPK pathway



coordinated activation plays a crucial role in EMT via increased cellular motility[15,22,54,55]. Studies using the MDCK cell model showed that EMT mediated by RON is associated with decreased E-Cadherin expression and unregulated vimentin expression, under mediation by RAS-MAPK signaling[56,57]. The major protein that links EMT to RON signaling is ribosomal protein S6 kinase-2, which is an intermediate in the MAPK pathway. RON-mediated PI-3K-AKT signaling is also involved in invasive growth, including increased epithelial cell matrix invasion, migration, and adhesion in vitro and distant metastasis and tumor cell invasion in vivo[54,55].

ABERRANT RON SIGNALING AND EXPRESSION IN CANCER PATHOGENESISIn general, normal epithelial cells, including those from the colon, lung, and breast, express low levels of RON; however, cells of a mesenchymal origin do not express RON[12,58]. RON activation in tumors is frequently the result of receptor overex-pression, in contrast to classical MSP binding. Dysregulated RON activation and expression were detected in many types of cancers and have prognostic significance for patient survival. Results from the majority of published studies show that RON expression dysregulation is characterized mainly by elevated expression of wild-type RON and the production of active isoforms, ultimately leading to persistent activation of downstream signaling cascades[12]. There have been reports of RON amplification and point mutation; however, this kind of genetic alteration is observed rarely[26]. The relationships between cancer pathogenesis and dysregulated RON signaling and expression were proven via functional studies using immunohistochemical (IHC) staining of tumor specimens and cancer cell lines. The first report of wild-type RON overexpression in cancerous tissue was in primary breast cancer samples. Thereafter, IHC staining has further detected wild-type RON in thyroid, bladder, adrenal gland, head and neck, uterus, skin, lung, kidney, pancreatic, colorectal, and other tumors[59]. These findings are consistent with the results found in other cancers, such as human gliomas, melanoma, and Merkel cell carcinoma, suggesting that aberrant RON expression is also associated with both neurological and skin cancers[18]. In breast tissues, the expression of RON is relatively low in normal breast epithelial cells and even in cells from benign lesions (papilloma and adenoma), whereas its was highly expressed in 47% (35/75 cases) of histologically varied tumor specimens[25]. RON upregulation is associated strongly with its phosphorylation status and invasive activity, suggesting that dysregulated expression of RON functions in human breast carcinoma progression to invasive-metastatic phenotypes. Furthermore, in breast cancer unregulated RON expression was identified as an independent predictor of distant relapse[60]. By contrast, in certain tumors, such as hepatocellular carcinoma (HCC), the frequency of wild-type RON expression was relatively low[21]; however, its importance remains unknown, although this finding indicates that the wild-type RON is not expressed universally in different tumor types. Moreover, RON overex-pression is related to oncogenic RON isoform production, for example RON△160, which comprises the deletion of exons 5 and 6, encoding 109 amino acids in the RON β-chain extracellular sequence[12]. RON variants are detected in primary cancer samples and cell lines relatively frequently, and are detected in 40% to 60% of cases. Cancer pathogenesis and clinical relevance are likely to be affected by the frequencies and levels of RON isoforms.

Increasing evidence has demonstrated the role of RON in regulating cancer cell invasiveness, which is related to the effects of RON activation on a variety of signaling mechanisms. The activation of complex downstream signaling networks including signal transducer and activator of transcription, β-catenin, JUN N-terminal kinase, MAPK, and PI3K/AKT pathways are key contributors to RON-mediated aggressive cancer phenotypes. In breast cancer, several signaling pathways that are vital for stemness, invasiveness, and proliferation are activated by RON. For example, the RON-cellular Abelson murine leukemia viral oncogene homolog (c-Abl)-proliferation cell nuclear antigen (PCNA) pathway was identified to contribute to RON-mediated cell growth in breast cancer. Dysregulated RON signaling results in c-Abl activation, consequently leading to PCNA phosphorylation[61]. Moreover, in breast cancer, RON signaling regulates the invasiveness of cancer cell via the activation of the DEK proto-oncogene (DEK), a DNA-binding non-histone nuclear phosphoprotein that induces closed circular DNA to form positive supercoils[62]. This process appears to function via a paracrine and autocrine canonical β-catenin signaling loop, which ultimately



influences breast cancer stemness. In addition, RON-mediated PI3K-dependent upregulation of methyl-CpG binding domain 4, DNA glycosylase (MBD4) increases the invasive growth and metastasis of breast cancer cell via the reprogramming of the DNA methylation of specific target genes[63]. Clinical data indicated that in patients with breast cancer, poor prognosis is related to the RON-MBD4 epigenetic pathway[63].

RON RECEPTOR AND HEPATOBILIARY CANCERSIn 2030, it is estimated that more than 1 million people will die because of liver cancer worldwide[64]. Primary malignancies of the liver and adjacent biliary tract include HCC, intrahepatic and extrahepatic cholangiocarcinoma (CCA), and gallbladder cancer (GBC). Among them, HCC and intrahepatic CCA account for 85% and 10%, respectively[65]. Abnormal RON expression has been observed in HCC, which may be related to pathological conditions of this cancer[43]. In an HCC cell line study, RON was shown to be associated with oncogenic and invasive phenotypes (e.g., resistance to apoptosis, tumor cell migration, and tumor cell invasion) via AKT, c-Raf, and extracellular regulated kinase (ERK) signaling cascade modulation[66]. Clinically, RON and MET expression in patients with HCC after curative resection suggested no association of RON with overall survival and overall recurrence rates. However, patients with RON+/MET+ disease experienced higher overall recurrence rates compared with those displaying alternative expression patterns[67]. Similar to HCC, RON is emerging as an important mediator of CCA pathogenesis and clinical prognosis. Investigation of RON and MET expression in patients with perihilar CCA who underwent histologically curative resection revealed that patients with RON+/MET+ disease showed a worse overall survival rate than patients with other patterns[44]. In addition, in patients with extrahepatic CCA, the complete loss of MET, RON, or both (and their overexpression) was a factor for poor prognosis, likely due to the high rate of lymph-node metastasis[68]. Recently, Cheng and co-workers indicated that BMS-777607, a MET-RON dual inhibitor, inhibited HuCCT1 and KKU-100 human CCA cell growth, and decreased the growth of tumors in CCA rats. They further found that for patients with CCA who had previously undergone hepatectomies, upregu-lation of RON and MET was a predictor of poor survival[69]. Taken together, these studies suggest that the aberrant RON expression found in human hepatobiliary cancer samples and cell lines is closely related to pathological conditions and clinical outcome.

RON RECEPTOR AND PANCREATIC CANCERThe majority of malignant neoplasms of the pancreas are adenocarcinomas, among which pancreatic ductal adenocarcinoma is the most common malignancy, repres-enting an excess of 95% of all pancreatic malignancies[70]. Pancreatic cancer presents a substantial health problem and is associated with an extremely poor prognosis because of the non-specific symptoms in patients, its aggressive and remarkable resistance to most conventional treatment options, and the fact that it harbors multiple genetic and epigenetic alterations[42]. Therefore, novel therapies to treat pancreatic cancer are urgently required. In recent years, the function of RON in pancreatic cancer has been identified extensively in a variety of model systems, such as animal, cellular, and clinical settings. To date, researchers have reported that RON is expressed in various pancreatic cancer cell lines, such as CFPAC-1, ASPC-1, Hs766.T, L3.6pl, HPAFII, HPAC, Capan-2, and BXPC-3. However, MIA-PACA-2 cells show minimal RON expression[71]. The association of RON with Kras-driven pancreatic carcino-genesis was investigated using genetically engineered mouse models. The results showed that overexpression of RON accelerated pancreatic intraepithelial neoplasia (PanIN) progression, enhanced acinar-ductal metaplasia, and promoted tumor progression towards invasive pancreatic cancer[72]. Moreover, the study proved that the initiation of PanIN was slowed by RON kinase domain genetic inactivation, resulting in smaller tumors, and eventually prolonging tumor-bearing mouse survival[72]. Great progress has been made in our understanding of the clinical relevance of RON in pancreatic cancer, which has focused mainly on RON expression status in pancreatic cancer samples and its possible utility as a prognostic biomarker for patient survival. IHC staining using anti-RON antibodies is a commonly used approach to evaluate RON expression in various experimental settings. Several studies



have identified positive sample rates in pancreatic cancer specimens such as 70%, 88%, 96%, 80%, and 86%, respectively[21,73]. Meanwhile, in pancreatic cancer samples, high RON expression has been detected, whereas minimal levels were detected in their corresponding normal epithelial cells. Notably, during pancreatic cell tumorigenic progression, the frequency and level of RON expression increased[22,74]. Among human pancreatic tissue samples, RON expression was detected in 83% of metastatic lesions, 79% of primary lesions, and 93% of high-grade PanIN using immuno-histochemistry, with low expression being detected in low-grade PanIN and normal ducts (18% and 6%, respectively), suggesting that RON might function in pancreatic carcinogenesis and metastatic progression[22]. Moreover, RON expression levels were significantly related to overall survival in patients with pancreatic cancer, indicating that RON could be an important indicator of prognosis in pancreatic cancer[75]. Conflicting results between RON expression and pancreatic cancer prognosis were found in an early study[76], thus more research is needed to determine the utility of RON as a prognostic biomarker in patients with pancreatic cancer.

Primary and metastatic pancreatic tumor specimens and high grade PanIN lesions show increased RON expression[22]. Accumulating evidence suggests that dysreg-ulated RON signaling and activation might function in tumor formation and metastasis. Generally, activation of RON results in increased pancreatic cancer tumorigenic stemness, chemoresistance, survival capability, angiogenesis, and cell invasiveness[73]. Among them, invasiveness occurs via a phenotype resembling EMT. A study found that MSP treatment of the pancreatic cancer cell line L3.6pl resulted in increased cell invasion, cell migration, and ERK phosphorylation[24]. Activation of RON resulted in decreased levels of membrane-bound E-cadherin together with β-catenin nuclear translocation, which resembled EMT. Treated L3.6pl cells acquired a spindle shape and lost their polarity, their intercellular separation increased, and more pseudopodia were formed[24]. Aberrant RON activation in collaboration with other growth factors, such as transforming growth factor-β, contributes to the phenotypic changes of pancreatic cancer cells towards EMT. Additionally, an investigation of RON signaling-mediated angiogenesis regulation in pancreatic cancer found that RON signaling leads to MAPK-mediated pancreatic cancer cell production of the well-characterized angiogenic protein, vascular endothelial growth factor. RON activation also caused the promotion of microtubule formation[77]. Finally, the RON signaling pathway also plays a part in chemoresistance, which is associated with enhanced survival capability[51,78]. Short hairpin RNA (shRNA)-mediated silencing of RON expression in pancreatic cancer xenografts resulted in increased sensitivity to gemcitabine therapy and susceptibility to apoptosis[51]. In the light of the above findings, it is clear that RON signaling is crucial for pancreatic cancer formation and metastasis.

RON AS A THERAPEUTIC TARGET FOR HBP CANCERSBased on the pathogenic role of RON in cancers, including HBP cancers, efforts have focused predominantly on establishing RON as a drug target for therapeutic intervention[73]. A variety of techniques were proposed to effectively block RON signaling and expression. One approach is to inhibit RON expression using gene silencing with small interfering RNAs (siRNAs). In pancreatic cancer xenografts, RON silencing caused growth inhibition by enhancing their apoptosis susceptibility and via sensitization to gemcitabine therapy[51]. Thus, delivery of RON-specific siRNAs could have therapeutic potential. In addition, small-molecule kinase inhibitors (SMKIs), which block the receptor tyrosine kinase domain either via non-competitive inhibition or via ATP competition, have been proposed[45]. The structural similarities between the kinase domains of MET and RON resulted in the development of selective small molecule inhibitors targeting both the RON and MET kinase domains, with slightly different IC50 values. As described above, BMS-777607, a MET-RON dual inhibitor, has shown its effects in inhibiting the growth of human intra-hepatic CCA cell lines and also decreasing tumor growth in intrahepatic CCA rats[69]. However, preclinical studies to prove RON as a drug target showed unsatisfactory results when using RON-specific SMKIs[46,79-81]. The first reason for the above result is that HBP cancer cell survival does not depend on RON signaling. Second, an SMKI that specifically inhibits only RON kinase activities is not available. Synthetic SMKIs, including Tivantinib, BMS-777607, INCB28060, Compound-1, and PHA665752, all recognize both RON and MET, with similar kinase-binding affinities[73]. Thus, the characterized SMKI RON or MET-specific inhibitors are actually multiple RTK inhibitors and the



Table 1 Tyrosine kinase inhibitors and antibody drug conjugates specific to c-MET and RON

Therapeutic agents Manufacturer Target In vitro effects Effects in animal tumor models Clinical trial information Status Ref.

TKIs

Foretinib GlaxoSmithKline MET, RON, VEGFR2, and PDGFRβ

Inhibits MET and RON signaling and cell growth in various cancer cell lines

Attenuates MET- and RON-mediated tumor growth in mouse tumor xenograft models

Single agent and combination with erlotinib or lapatinib for various types of advanced cancers in Phase II/III clinical trials

Phase I/II/III

Eder et al[82]

MGCD265 MethylGene MET, RON, VEGFR1, VEGFR2, VEGFR3, and TIE2

Inhibits MET and RON signaling and cell growth in cancer cell lines

Attenuates MET- and RON-mediated tumor growth in mouse tumor xenograft models

Single agent and combination with erlotinib or docetaxel for NSCLC in Phase II trials

Phase I/II Belalcazar et al[83]

BMS-777607 Bristol-Myers Squibb

RON and MET Inhibits MET and RON signaling, cell growth, and invasion in cancer cell lines

Inhibits MET- and RON-mediated tumor growth in mouse tumor xenograft models

Multiple ascending doses for metastatic cancers in Phase I trials

Phase I Sharma et al[84]

MK-2461 Merck MET, RON, FLT1, FLT3, FGFR1, FGFR2, and FGFR3

Inhibits MET and RON signaling, cell growth, and migration in cancer cell lines

Inhibits MET- and RON-mediated tumor growth in mouse tumor xenograft models

Antitumor efficacy is under evaluation in Phase II trials

Phase I/II Pan et al[85]

MK-8033 Merck MET and RON Inhibits MET and RON signaling, cell growth, and migration in cancer cell lines

Causes tumor regression in mouse tumor xenograft models

Safety, tolerability, dose, clinical activity and pharmaco-dynamics are under evaluation in Phase I trials

Phase I Northrup et al[86]

PHA665752 Pfizer MET and RON NA NA NA Preclinical Comoglio et al[87]

INC280 Novartis MET NA NA NA Phase I/II Qin et al[88]

Tivantinib ArQule MET NA NA NA Phase II/III Rimassa et al[89]

Antibody drug conjugates

Zt/g4-doxorubicin-immuoliposome

TTUHSC RON Moderately activates RON signaling and strongly induces RON endocytosis

No effect as naked antibody but completely inhibits tumors used as ADCs

NA Preclinical Guin et al[90]

Zt/g4-maytansinoid conjugate

TTUHSC RON Moderately activates RON signaling and strongly induces RON endocytosis


NA Preclinical Feng et al[91]

Zt/g4-MMAE TTUHSC RON Moderately activates RON signaling and strongly induces RON endocytosis


NA Preclinical Yao et al[92]

H5B14-MMAE TTUHSC RON NA NA NA Preclinical Yao et al[59]

SHR-A1403 HengRui MET Highly potent: 0.02 to 1.5 nmol/L for cell proliferation

Xenografts and PDXs, MET over-expressed and amplified

NA Phase I Yang et al[93]



TKIs: Tyrosine kinase inhibitors; VEGFR: Vascular endothelial growth factor receptor; PDGFR: Platelet-derived growth factor receptor; NSCLC: Non-small cell lung cancer; FGFR: Fibroblast growth factor receptor; ADC: Antibody-drug conjugate; MMAE: Monomethyl auristatin E; NA: Not available; PDX: Patient-derived xenografts.

development of SMKIs that exclusively target RON has been a challenge.A more realistic approach is using anti-RON therapeutic monoclonal antibodies

(TPABs) to treat HBP cancers. For instance, anti-RON antibody Zt/c9-directing doxorubicin-immunoliposomes was effective at killing purified pancreatic cancer stem cells in vitro. The underlying mechanism is that Zt/c9-directing doxorubicin-immunol-iposomes specifically interact with pancreatic cancer stem cells and rapidly cause RON internalization, which leads to the uptake of liposome-coated Dox. In addition, preclinical models have been constructed using anti-RON TPABs, such as 7G8, 6D4, 6E6, narnatumab (or IMC-RON8), Zt/f2, and IMC-41A10, which either block MSP binding by recognizing RON’s ligand-binding pocket or affect receptor dimerization by interacting with RON’s extracellular domain (e.g., SEMA), thereby attenuating signaling transduction[73]. However, previous studies concerning TPAB therapy revealed only partial inhibition of tumor growth, and there have been no reports of single anti-RON TPAB administration achieving complete inhibition. Thus, strategies to maximize anti-RON TPABs’ therapeutic activity have moved on to an exciting new area. Anticancer therapeutic agents comprising antibody-drug conjugates (ADCs) combine the specificity of antibodies with the high potency of cytotoxins to enhance cell killing[12]. To generate RON-targeted ADCs, the anti-RON monoclonal antibodies PCM5B14 and Zt/g4 were selected to prepare immunotoxins. To generate Zt/g4 and PCM5B14-based ADCs, cytotoxic payloads with different mechanisms of action were conjugated, including pyrrolobenzodiazepine, duocarmycin (DCM), monomethyl auristatin E, and maytansinoid derivative 1, forming for example, Zt/g4-MME and PCM5B14-DCM[59]. Preclinical studies identified Zt/g4- and PCM5B14-based ADCs as lead candidates for clinical development and increased the chance of their entering into clinical trials (Table 1).

CONCLUSIONRON was identified over two decades ago, and since then, accumulating evidence has indicating RON’s involvement in tumorigenesis, which has resulted in increased momentum for developing RON as a target for therapeutic drug intervention. As outlined in this review, the identification of dysregulated activation and expression of RON in various cancers has expanded our understanding of the mechanisms underlying cancer pathogenesis. Importantly, HBP cancers are characterized patholo-gically by the dysregulated signaling and expression of RON, which also act as tumorigenic determinants for the malignant behavior of HBP cancers. Moreover, abnormal RON expression is important to determine the clinical outcome of patients



with HBP cancers. The growing knowledge concerning the crucial role of RON in HBP cancers can be translated into promising cancer therapeutic strategies. Consequently, a number of clinical trials are underway to assess SMKIs and TPABs targeting RON as a molecular target, some of which have shown promising results. Furthermore, PCM5B14- and Zt/g4-based ADCs, as anti-RON ADCs, are receiving increased research interest and the striking advances in exploiting anti-RON ADCs will hopefully translate into clinical treatments for patients with HBP cancer in the future.

REFERENCESRonsin C, Muscatelli F, Mattei MG, Breathnach R. A novel putative receptor protein tyrosine kinase of the met family. Oncogene 1993; 8: 1195-1202 [PMID: 8386824]

1

Gherardi E, Sharpe M, Lane K, Sirulnik A, Stoker M. Hepatocyte growth factor/scatter factor (HGF/SF), the c-met receptor and the behaviour of epithelial cells. Symp Soc Exp Biol 1993; 47: 163-181 [PMID: 8165564]

2

Wagh PK, Peace BE, Waltz SE. Met-related receptor tyrosine kinase Ron in tumor growth and metastasis. Adv Cancer Res 2008; 100: 1-33 [PMID: 18620091 DOI: 10.1016/S0065-230X(08)00001-8]

3

Iwama A, Okano K, Sudo T, Matsuda Y, Suda T. Molecular cloning of a novel receptor tyrosine kinase gene, STK, derived from enriched hematopoietic stem cells. Blood 1994; 83: 3160-3169 [PMID: 8193352]

4

De Maria R, Maggiora P, Biolatti B, Prat M, Comoglio PM, Castagnaro M, Di Renzo MF. Feline STK gene expression in mammary carcinomas. Oncogene 2002; 21: 1785-1790 [PMID: 11896610 DOI: 10.1038/sj.onc.1205221]

5

Bassett DI. Identification and developmental expression of a macrophage stimulating 1/ hepatocyte growth factor-like 1 orthologue in the zebrafish. Dev Genes Evol 2003; 213: 360-362 [PMID: 12764615 DOI: 10.1007/s00427-003-0339-3]

6

Huitema LF, Renn J, Logister I, Gray JK, Waltz SE, Flik G, Schulte-Merker S. Macrophage-stimulating protein and calcium homeostasis in zebrafish. FASEB J 2012; 26: 4092-4101 [PMID: 22787265 DOI: 10.1096/fj.11-202663]

7

Théry C, Sharpe MJ, Batley SJ, Stern CD, Gherardi E. Expression of HGF/SF, HGF1/MSP, and c-met suggests new functions during early chick development. Dev Genet 1995; 17: 90-101 [PMID: 7554499 DOI: 10.1002/dvg.1020170110]

8

Huff JL, Jelinek MA, Jamieson TA, Parsons JT. Expression and maturation of the cellular sea receptor, a member of the hepatocyte growth factor (HGF) receptor family of protein tyrosine kinases. Oncogene 1996; 12: 299-307 [PMID: 8570207]

9

Wahl RC, Hsu RY, Huff JL, Jelinek MA, Chen K, Courchesne P, Patterson SD, Parsons JT, Welcher AA. Chicken macrophage stimulating protein is a ligand of the receptor protein-tyrosine kinase Sea. J Biol Chem 1999; 274: 26361-26368 [PMID: 10473593 DOI: 10.1074/jbc.274.37.26361]

10

Nakamura T, Aoki S, Takahashi T, Matsumoto K, Kiyohara T, Nakamura T. Cloning and expression of Xenopus HGF-like protein (HLP) and Ron/HLP receptor implicate their involvement in early neural development. Biochem Biophys Res Commun 1996; 224: 564-573 [PMID: 8702427 DOI: 10.1006/bbrc.1996.1065]

11

Yao HP, Zhou YQ, Zhang R, Wang MH. MSP-RON signalling in cancer: pathogenesis and therapeutic potential. Nat Rev Cancer 2013; 13: 466-481 [PMID: 23792360 DOI: 10.1038/nrc3545]

12

Wang MH, Iwama A, Skeel A, Suda T, Leonard EJ. The murine stk gene product, a transmembrane protein tyrosine kinase, is a receptor for macrophage-stimulating protein. Proc Natl Acad Sci U S A 1995; 92: 3933-3937 [PMID: 7732008 DOI: 10.1073/pnas.92.9.3933]

13

Wang MH, Ronsin C, Gesnel MC, Coupey L, Skeel A, Leonard EJ, Breathnach R. Identification of the ron gene product as the receptor for the human macrophage stimulating protein. Science 1994; 266: 117-119 [PMID: 7939629 DOI: 10.1126/science.7939629]

14

Gaudino G, Follenzi A, Naldini L, Collesi C, Santoro M, Gallo KA, Godowski PJ, Comoglio PM. RON is a heterodimeric tyrosine kinase receptor activated by the HGF homologue MSP. EMBO J 1994; 13: 3524-3532 [PMID: 8062829]

15

Han S, Stuart LA, Degen SJ. Characterization of the DNF15S2 Locus on human chromosome 3: identification of a gene coding for four kringle domains with homology to hepatocyte growth factor. Biochemistry 1991; 30: 9768-9780 [PMID: 1655021 DOI: 10.1021/bi00104a029]

16

Yoshimura T, Yuhki N, Wang MH, Skeel A, Leonard EJ. Cloning, sequencing, and expression of human macrophage stimulating protein (MSP, MST1) confirms MSP as a member of the family of kringle proteins and locates the MSP gene on chromosome 3. J Biol Chem 1993; 268: 15461-15468 [PMID: 8393443]

17

Benight NM, Waltz SE. Ron receptor tyrosine kinase signaling as a therapeutic target. Expert Opin Ther Targets 2012; 16: 921-931 [PMID: 22834780 DOI: 10.1517/14728222.2012.710200]

18

Wang MH, Julian FM, Breathnach R, Godowski PJ, Takehara T, Yoshikawa W, Hagiya M, Leonard EJ. Macrophage stimulating protein (MSP) binds to its receptor via the MSP beta chain. J Biol Chem 1997; 272: 16999-17004 [PMID: 9202013 DOI: 10.1074/jbc.272.27.16999]

19




https://dx.doi.org/10.1016/S0065-230X(08)00001-8



https://dx.doi.org/10.1038/sj.onc.1205221


https://dx.doi.org/10.1007/s00427-003-0339-3


https://dx.doi.org/10.1096/fj.11-202663


https://dx.doi.org/10.1002/dvg.1020170110



https://dx.doi.org/10.1074/jbc.274.37.26361


https://dx.doi.org/10.1006/bbrc.1996.1065


https://dx.doi.org/10.1038/nrc3545


https://dx.doi.org/10.1073/pnas.92.9.3933


https://dx.doi.org/10.1126/science.7939629



https://dx.doi.org/10.1021/bi00104a029



https://dx.doi.org/10.1517/14728222.2012.710200





Danilkovitch A, Miller M, Leonard EJ. Interaction of macrophage-stimulating protein with its receptor. Residues critical for beta chain binding and evidence for independent alpha chain binding. J Biol Chem 1999; 274: 29937-29943 [PMID: 10514476 DOI: 10.1074/jbc.274.42.29937]

20

Wang MH, Lee W, Luo YL, Weis MT, Yao HP. Altered expression of the RON receptor tyrosine kinase in various epithelial cancers and its contribution to tumourigenic phenotypes in thyroid cancer cells. J Pathol 2007; 213: 402-411 [PMID: 17955509 DOI: 10.1002/path.2245]

21

Thomas RM, Toney K, Fenoglio-Preiser C, Revelo-Penafiel MP, Hingorani SR, Tuveson DA, Waltz SE, Lowy AM. The RON receptor tyrosine kinase mediates oncogenic phenotypes in pancreatic cancer cells and is increasingly expressed during pancreatic cancer progression. Cancer Res 2007; 67: 6075-6082 [PMID: 17616662 DOI: 10.1158/0008-5472.CAN-06-4128]

22

Kanteti R, Krishnaswamy S, Catenacci D, Tan YH, EL-Hashani E, Cervantes G, Husain AN, Tretiakova M, Vokes EE, Huet H, Salgia R. Differential expression of RON in small and non-small cell lung cancers. Genes Chromosomes Cancer 2012; 51: 841-851 [PMID: 22585712 DOI: 10.1002/gcc.21968]

23

Camp ER, Yang A, Gray MJ, Fan F, Hamilton SR, Evans DB, Hooper AT, Pereira DS, Hicklin DJ, Ellis LM. Tyrosine kinase receptor RON in human pancreatic cancer: expression, function, and validation as a target. Cancer 2007; 109: 1030-1039 [PMID: 17311308 DOI: 10.1002/cncr.22490]

24

Maggiora P, Marchio S, Stella MC, Giai M, Belfiore A, De Bortoli M, Di Renzo MF, Costantino A, Sismondi P, Comoglio PM. Overexpression of the RON gene in human breast carcinoma. Oncogene 1998; 16: 2927-2933 [PMID: 9671413 DOI: 10.1038/sj.onc.1201812]

25

Catenacci DV, Cervantes G, Yala S, Nelson EA, El-Hashani E, Kanteti R, El Dinali M, Hasina R, Brägelmann J, Seiwert T, Sanicola M, Henderson L, Grushko TA, Olopade O, Karrison T, Bang YJ, Kim WH, Tretiakova M, Vokes E, Frank DA, Kindler HL, Huet H, Salgia R. RON (MST1R) is a novel prognostic marker and therapeutic target for gastroesophageal adenocarcinoma. Cancer Biol Ther 2011; 12: 9-46 [PMID: 21543897 DOI: 10.4161/cbt.12.1.15747]

26

Ren X, Daa T, Yada N, Kashima K, Fujitomi Y, Yokoyama S. Expression and mutational status of RON in neoplastic lesions of the breast: analysis of MSP/RON signaling in ductal carcinoma in situ and invasive ductal carcinoma. APMIS 2012; 120: 358-367 [PMID: 22515290 DOI: 10.1111/j.1600-0463.2011.02841.x]

27

Chou YC, Chen CL, Yeh TH, Lin SJ, Chen MR, Doong SL, Lu J, Tsai CH. Involvement of recepteur d'origine nantais receptor tyrosine kinase in Epstein-Barr virus-associated nasopharyngeal carcinoma and its metastasis. Am J Pathol 2012; 181: 1773-1781 [PMID: 22974584 DOI: 10.1016/j.ajpath.2012.07.014]

28

Collesi C, Santoro MM, Gaudino G, Comoglio PM. A splicing variant of the RON transcript induces constitutive tyrosine kinase activity and an invasive phenotype. Mol Cell Biol 1996; 16: 5518-5526 [PMID: 8816464 DOI: 10.1128/mcb.16.10.5518]

29

Santoro MM, Collesi C, Grisendi S, Gaudino G, Comoglio PM. Constitutive activation of the RON gene promotes invasive growth but not transformation. Mol Cell Biol 1996; 16: 7072-7083 [PMID: 8943362 DOI: 10.1128/mcb.16.12.7072]

30

Zhou YQ, He C, Chen YQ, Wang D, Wang MH. Altered expression of the RON receptor tyrosine kinase in primary human colorectal adenocarcinomas: generation of different splicing RON variants and their oncogenic potential. Oncogene 2003; 22: 186-197 [PMID: 12527888 DOI: 10.1038/sj.onc.1206075]

31

Liu X, Zhao L, Derose YS, Lin YC, Bieniasz M, Eyob H, Buys SS, Neumayer L, Welm AL. Short-Form Ron Promotes Spontaneous Breast Cancer Metastasis through Interaction with Phosphoinositide 3-Kinase. Genes Cancer 2011; 2: 753-762 [PMID: 22207901 DOI: 10.1177/1947601911421924]

32

Eckerich C, Schulte A, Martens T, Zapf S, Westphal M, Lamszus K. RON receptor tyrosine kinase in human gliomas: expression, function, and identification of a novel soluble splice variant. J Neurochem 2009; 109: 969-980 [PMID: 19519771 DOI: 10.1111/j.1471-4159.2009.06027.x]

33

Ma Q, Zhang K, Guin S, Zhou YQ, Wang MH. Deletion or insertion in the first immunoglobulin-plexin-transcription (IPT) domain differentially regulates expression and tumorigenic activities of RON receptor Tyrosine Kinase. Mol Cancer 2010; 9: 307 [PMID: 21114864 DOI: 10.1186/1476-4598-9-307]

34

Wei X, Hao L, Ni S, Liu Q, Xu J, Correll PH. Altered exon usage in the juxtamembrane domain of mouse and human RON regulates receptor activity and signaling specificity. J Biol Chem 2005; 280: 40241-40251 [PMID: 16166096 DOI: 10.1074/jbc.M506806200]

35

Ghigna C, Giordano S, Shen H, Benvenuto F, Castiglioni F, Comoglio PM, Green MR, Riva S, Biamonti G. Cell motility is controlled by SF2/ASF through alternative splicing of the Ron protooncogene. Mol Cell 2005; 20: 881-890 [PMID: 16364913 DOI: 10.1016/j.molcel.2005.10.026]

36

Fialin C, Larrue C, Vergez F, Sarry JE, Bertoli S, Mansat-De Mas V, Demur C, Delabesse E, Payrastre B, Manenti S, Roche S, Récher C. The short form of RON is expressed in acute myeloid leukemia and sensitizes leukemic cells to cMET inhibitors. Leukemia 2013; 27: 325-335 [PMID: 22902361 DOI: 10.1038/leu.2012.240]

37

Bardella C, Costa B, Maggiora P, Patane' S, Olivero M, Ranzani GN, De Bortoli M, Comoglio PM, Di Renzo MF. Truncated RON tyrosine kinase drives tumor cell progression and abrogates cell-cell adhesion through E-cadherin transcriptional repression. Cancer Res 2004; 64: 5154-5161 [PMID: 15289319 DOI: 10.1158/0008-5472.CAN-04-0600]

38

Wang MH, Kurtz AL, Chen Y. Identification of a novel splicing product of the RON receptor tyrosine kinase in human colorectal carcinoma cells. Carcinogenesis 2000; 21: 1507-1512 [PMID:

39




https://dx.doi.org/10.1002/path.2245




https://dx.doi.org/10.1002/gcc.21968






https://dx.doi.org/10.4161/cbt.12.1.15747




https://dx.doi.org/10.1016/j.ajpath.2012.07.014


https://dx.doi.org/10.1128/mcb.16.10.5518


https://dx.doi.org/10.1128/mcb.16.12.7072




https://dx.doi.org/10.1177/1947601911421924




https://dx.doi.org/10.1186/1476-4598-9-307




https://dx.doi.org/10.1016/j.molcel.2005.10.026


https://dx.doi.org/10.1038/leu.2012.240





10910951]Wu LM, Zhang LL, Chen XH, Zheng SS. Is irreversible electroporation safe and effective in the treatment of hepatobiliary and pancreatic cancers? Hepatobiliary Pancreat Dis Int 2019; 18: 117-124 [PMID: 30655073 DOI: 10.1016/j.hbpd.2019.01.001]

40

Ueno M, Morizane C, Ikeda M, Okusaka T, Ishii H, Furuse J. A review of changes to and clinical implications of the eighth TNM classification of hepatobiliary and pancreatic cancers. Jpn J Clin Oncol 2019; 49: 1073-1082 [PMID: 31822900 DOI: 10.1093/jjco/hyz150]

41

Kleeff J, Korc M, Apte M, La Vecchia C, Johnson CD, Biankin AV, Neale RE, Tempero M, Tuveson DA, Hruban RH, Neoptolemos JP. Pancreatic cancer. Nat Rev Dis Primers 2016; 2: 16022 [PMID: 27158978 DOI: 10.1038/nrdp.2016.22]

42

Chen Q, Seol DW, Carr B, Zarnegar R. Co-expression and regulation of Met and Ron proto-oncogenes in human hepatocellular carcinoma tissues and cell lines. Hepatology 1997; 26: 59-66 [PMID: 9214452 DOI: 10.1002/hep.510260108]

43

Watanabe H, Yokoyama Y, Kokuryo T, Ebata T, Igami T, Sugawara G, Mizuno T, Shimoyama Y, Nagino M. Prognostic Value of Hepatocyte Growth Factor Receptor Expression in Patients with Perihilar Cholangiocarcinoma. Ann Surg Oncol 2015; 22: 2235-2242 [PMID: 25586241 DOI: 10.1245/s10434-014-4170-z]

44

Wang MH, Padhye SS, Guin S, Ma Q, Zhou YQ. Potential therapeutics specific to c-MET/RON receptor tyrosine kinases for molecular targeting in cancer therapy. Acta Pharmacol Sin 2010; 31: 1181-1188 [PMID: 20694025 DOI: 10.1038/aps.2010.106]

45

Dussault I, Bellon SF. From concept to reality: the long road to c-Met and RON receptor tyrosine kinase inhibitors for the treatment of cancer. Anticancer Agents Med Chem 2009; 9: 221-229 [PMID: 19199866 DOI: 10.2174/187152009787313792]

46

Chang K, Karnad A, Zhao S, Freeman JW. Roles of c-Met and RON kinases in tumor progression and their potential as therapeutic targets. Oncotarget 2015; 6: 3507-3518 [PMID: 25784650 DOI: 10.18632/oncotarget.3420]

47

Lee CT, Chow NH, Su PF, Lin SC, Lin PC, Lee JC. The prognostic significance of RON and MET receptor coexpression in patients with colorectal cancer. Dis Colon Rectum 2008; 51: 1268-1274 [PMID: 18536971 DOI: 10.1007/s10350-008-9297-1]

48

Ponzo MG, Lesurf R, Petkiewicz S, O'Malley FP, Pinnaduwage D, Andrulis IL, Bull SB, Chughtai N, Zuo D, Souleimanova M, Germain D, Omeroglu A, Cardiff RD, Hallett M, Park M. Met induces mammary tumors with diverse histologies and is associated with poor outcome and human basal breast cancer. Proc Natl Acad Sci USA 2009; 106: 12903-12908 [PMID: 19617568 DOI: 10.1073/pnas.0810402106]

49

Comoglio PM, Trusolino L, Boccaccio C. Known and novel roles of the MET oncogene in cancer: a coherent approach to targeted therapy. Nat Rev Cancer 2018; 18: 341-358 [PMID: 29674709 DOI: 10.1038/s41568-018-0002-y]

50

Logan-Collins J, Thomas RM, Yu P, Jaquish D, Mose E, French R, Stuart W, McClaine R, Aronow B, Hoffman RM, Waltz SE, Lowy AM. Silencing of RON receptor signaling promotes apoptosis and gemcitabine sensitivity in pancreatic cancers. Cancer Res 2010; 70: 1130-1140 [PMID: 20103639 DOI: 10.1158/0008-5472.CAN-09-0761]

51

Zarei O, Benvenuti S, Ustun-Alkan F, Hamzeh-Mivehroud M, Dastmalchi S. Strategies of targeting the extracellular domain of RON tyrosine kinase receptor for cancer therapy and drug delivery. J Cancer Res Clin Oncol 2016; 142: 2429-2446 [PMID: 27503093 DOI: 10.1007/s00432-016-2214-4]

52

Wang X, Hankey PA. The ron receptor tyrosine kinase: a key regulator of inflammation and cancer progression. Crit Rev Immunol 2013; 33: 549-574 [PMID: 24266348 DOI: 10.1615/critrevimmunol.2013007953]

53

Xiao ZQ, Chen YQ, Wang MH. Requirement of both tyrosine residues 1330 and 1337 in the C-terminal tail of the RON receptor tyrosine kinase for epithelial cell scattering and migration. Biochem Biophys Res Commun 2000; 267: 669-675 [PMID: 10631120 DOI: 10.1006/bbrc.1999.2011]

54

Wang J, Rajput A, Kan JL, Rose R, Liu XQ, Kuropatwinski K, Hauser J, Beko A, Dominquez I, Sharratt EA, Brattain L, Levea C, Sun FL, Keane DM, Gibson NW, Brattain MG. Knockdown of Ron kinase inhibits mutant phosphatidylinositol 3-kinase and reduces metastasis in human colon carcinoma. J Biol Chem 2009; 284: 10912-10922 [PMID: 19224914 DOI: 10.1074/jbc.M809551200]

55

Wang D, Shen Q, Chen YQ, Wang MH. Collaborative activities of macrophage-stimulating protein and transforming growth factor-beta1 in induction of epithelial to mesenchymal transition: roles of the RON receptor tyrosine kinase. Oncogene 2004; 23: 1668-1680 [PMID: 15001985 DOI: 10.1038/sj.onc.1207282]

56

Ma Q, Guin S, Padhye SS, Zhou YQ, Zhang RW, Wang MH. Ribosomal protein S6 kinase (RSK)-2 as a central effector molecule in RON receptor tyrosine kinase mediated epithelial to mesenchymal transition induced by macrophage-stimulating protein. Mol Cancer 2011; 10: 66 [PMID: 21619683 DOI: 10.1186/1476-4598-10-66]

57

Faham N, Welm AL. RON Signaling Is a Key Mediator of Tumor Progression in Many Human Cancers. Cold Spring Harb Symp Quant Biol 2016; 81: 177-188 [PMID: 28057847 DOI: 10.1101/sqb.2016.81.031377]

58

Yao HP, Suthe SR, Tong XM, Wang MH. Targeting RON receptor tyrosine kinase for treatment of advanced solid cancers: antibody-drug conjugates as lead drug candidates for clinical trials. Ther Adv Med Oncol 2020; 12: 1758835920920069 [PMID: 32426050 DOI: 10.1177/1758835920920069]

59

Lee WY, Chen HH, Chow NH, Su WC, Lin PW, Guo HR. Prognostic significance of co-expression 60



https://dx.doi.org/10.1016/j.hbpd.2019.01.001


https://dx.doi.org/10.1093/jjco/hyz150


https://dx.doi.org/10.1038/nrdp.2016.22






https://dx.doi.org/10.1038/aps.2010.106


https://dx.doi.org/10.2174/187152009787313792




https://dx.doi.org/10.1007/s10350-008-9297-1








https://dx.doi.org/10.1007/s00432-016-2214-4


https://dx.doi.org/10.1615/critrevimmunol.2013007953


https://dx.doi.org/10.1006/bbrc.1999.2011






https://dx.doi.org/10.1186/1476-4598-10-66


https://dx.doi.org/10.1101/sqb.2016.81.031377


https://dx.doi.org/10.1177/1758835920920069



of RON and MET receptors in node-negative breast cancer patients. Clin Cancer Res 2005; 11: 2222-2228 [PMID: 15788670 DOI: 10.1158/1078-0432.CCR-04-1761]Zhao H, Chen MS, Lo YH, Waltz SE, Wang J, Ho PC, Vasiliauskas J, Plattner R, Wang YL, Wang SC. The Ron receptor tyrosine kinase activates c-Abl to promote cell proliferation through tyrosine phosphorylation of PCNA in breast cancer. Oncogene 2014; 33: 1429-1437 [PMID: 23542172 DOI: 10.1038/onc.2013.84]

61

Privette Vinnedge LM, Benight NM, Wagh PK, Pease NA, Nashu MA, Serrano-Lopez J, Adams AK, Cancelas JA, Waltz SE, Wells SI. The DEK oncogene promotes cellular proliferation through paracrine Wnt signaling in Ron receptor-positive breast cancers. Oncogene 2015; 34: 2325-2336 [PMID: 24954505 DOI: 10.1038/onc.2014.173]

62

Bellacosa A. Role of MED1 (MBD4) Gene in DNA repair and human cancer. J Cell Physiol 2001; 187: 137-144 [PMID: 11267993 DOI: 10.1002/jcp.1064]

63

Villanueva A. Hepatocellular Carcinoma. N Engl J Med 2019; 380: 1450-1462 [PMID: 30970190 DOI: 10.1056/NEJMra1713263]

64

Nault JC, Villanueva A. Biomarkers for Hepatobiliary Cancers. Hepatology 2021; 73 Suppl 1: 115-127 [PMID: 32045030 DOI: 10.1002/hep.31175]

65

Cho SB, Park YL, Song YA, Kim KY, Lee GH, Cho DH, Myung DS, Park KJ, Lee WS, Chung IJ, Choi SK, Kim KK, Joo YE. Small interfering RNA-directed targeting of RON alters invasive and oncogenic phenotypes of human hepatocellular carcinoma cells. Oncol Rep 2011; 26: 1581-1586 [PMID: 21874262 DOI: 10.3892/or.2011.1435]

66

Koh YW, Park YS, Kang HJ, Shim JH, Yu E. MET is a predictive factor for late recurrence but not for overall survival of early stage hepatocellular carcinoma. Tumour Biol 2015; 36: 4993-5000 [PMID: 25874493 DOI: 10.1007/s13277-015-3150-7]

67

Hayashi Y, Yamaguchi J, Kokuryo T, Ebata T, Yokoyama Y, Igami T, Sugawara G, Nagino M. The Complete Loss of Tyrosine Kinase Receptors MET and RON Is a Poor Prognostic Factor in Patients with Extrahepatic Cholangiocarcinoma. Anticancer Res 2016; 36: 6585-6592 [PMID: 27919987 DOI: 10.21873/anticanres.11263]

68

Cheng CT, Chen YY, Wu RC, Tsai CY, Chiang KC, Yeh TS, Chen MH, Yeh CN. METRON dual inhibitor, BMS777607, suppresses cholangiocarcinoma cell growth, and METRON upregulation indicates worse prognosis for intrahepatic cholangiocarcinoma patients. Oncol Rep 2018; 40: 1411-1421 [PMID: 30015968 DOI: 10.3892/or.2018.6543]

69

Collisson EA, Bailey P, Chang DK, Biankin AV. Molecular subtypes of pancreatic cancer. Nat Rev Gastroenterol Hepatol 2019; 16: 207-220 [PMID: 30718832 DOI: 10.1038/s41575-019-0109-y]

70

Kang CM, Babicky ML, Lowy AM. The RON receptor tyrosine kinase in pancreatic cancer pathogenesis and its potential implications for future targeted therapies. Pancreas 2014; 43: 183-189 [PMID: 24518495 DOI: 10.1097/MPA.0000000000000088]

71

Babicky ML, Harper MM, Chakedis J, Cazes A, Mose ES, Jaquish DV, French RP, Childers B, Alakus H, Schmid MC, Foubert P, Miyamoto J, Holman PJ, Walterscheid ZJ, Tang CM, Varki N, Sicklick JK, Messer K, Varner JA, Waltz SE, Lowy AM. MST1R kinase accelerates pancreatic cancer progression via effects on both epithelial cells and macrophages. Oncogene 2019; 38: 5599-5611 [PMID: 30967626 DOI: 10.1038/s41388-019-0811-9]

72

Yao HP, Hudson R, Wang MH. RON receptor tyrosine kinase in pancreatic ductal adenocarcinoma: Pathogenic mechanism in malignancy and pharmaceutical target for therapy. Biochim Biophys Acta Rev Cancer 2020; 1873: 188360 [PMID: 32234337 DOI: 10.1016/j.bbcan.2020.188360]

73

Li C, Morvaridi S, Lam G, Chheda C, Kamata Y, Katsumata M, Edderkaoui M, Yuan X, Nissen N, Pandol SJ, Wang Q. MSP-RON Signaling Is Activated in the Transition From Pancreatic Intraepithelial Neoplasia (PanIN) to Pancreatic Ductal Adenocarcinoma (PDAC). Front Physiol 2019; 10: 147 [PMID: 30863319 DOI: 10.3389/fphys.2019.00147]

74

Hu CY, Xu XM, Hong B, Wu ZG, Qian Y, Weng TH, Liu YZ, Tang TM, Wang MH, Yao HP. Aberrant RON and MET Co-overexpression as Novel Prognostic Biomarkers of Shortened Patient Survival and Therapeutic Targets of Tyrosine Kinase Inhibitors in Pancreatic Cancer. Front Oncol 2019; 9: 1377 [PMID: 31867280 DOI: 10.3389/fonc.2019.01377]

75

Tactacan CM, Chang DK, Cowley MJ, Humphrey ES, Wu J, Gill AJ, Chou A, Nones K, Grimmond SM, Sutherland RL, Biankin AV, Daly RJ; Australian Pancreratic Genome Initiative. RON is not a prognostic marker for resectable pancreatic cancer. BMC Cancer 2012; 12: 395 [PMID: 22958871 DOI: 10.1186/1471-2407-12-395]

76

Thomas RM, Jaquish DV, French RP, Lowy AM. The RON tyrosine kinase receptor regulates vascular endothelial growth factor production in pancreatic cancer cells. Pancreas 2010; 39: 301-307 [PMID: 20358644 DOI: 10.1097/mpa.0b013e3181bb9f73]

77

Zou Y, Howell GM, Humphrey LE, Wang J, Brattain MG. Ron knockdown and Ron monoclonal antibody IMC-RON8 sensitize pancreatic cancer to histone deacetylase inhibitors (HDACi). PLoS One 2013; 8: e69992 [PMID: 23922886 DOI: 10.1371/journal.pone.0069992]

78

Kawada I, Hasina R, Arif Q, Mueller J, Smithberger E, Husain AN, Vokes EE, Salgia R. Dramatic antitumor effects of the dual MET/RON small-molecule inhibitor LY2801653 in non-small cell lung cancer. Cancer Res 2014; 74: 884-895 [PMID: 24305878 DOI: 10.1158/0008-5472.CAN-12-3583]

79

Christensen JG, Schreck R, Burrows J, Kuruganti P, Chan E, Le P, Chen J, Wang X, Ruslim L, Blake R, Lipson KE, Ramphal J, Do S, Cui JJ, Cherrington JM, Mendel DB. A selective small molecule inhibitor of c-Met kinase inhibits c-Met-dependent phenotypes in vitro and exhibits cytoreductive antitumor activity in vivo. Cancer Res 2003; 63: 7345-7355 [PMID: 14612533]

80




https://dx.doi.org/10.1038/onc.2013.84


https://dx.doi.org/10.1038/onc.2014.173


https://dx.doi.org/10.1002/jcp.1064


https://dx.doi.org/10.1056/NEJMra1713263




https://dx.doi.org/10.3892/or.2011.1435


https://dx.doi.org/10.1007/s13277-015-3150-7


https://dx.doi.org/10.21873/anticanres.11263


https://dx.doi.org/10.3892/or.2018.6543






https://dx.doi.org/10.1038/s41388-019-0811-9


https://dx.doi.org/10.1016/j.bbcan.2020.188360


https://dx.doi.org/10.3389/fphys.2019.00147




https://dx.doi.org/10.1186/1471-2407-12-395


https://dx.doi.org/10.1097/mpa.0b013e3181bb9f73








Liu X, Wang Q, Yang G, Marando C, Koblish HK, Hall LM, Fridman JS, Behshad E, Wynn R, Li Y, Boer J, Diamond S, He C, Xu M, Zhuo J, Yao W, Newton RC, Scherle PA. A novel kinase inhibitor, INCB28060, blocks c-MET-dependent signaling, neoplastic activities, and cross-talk with EGFR and HER-3. Clin Cancer Res 2011; 17: 7127-7138 [PMID: 21918175 DOI: 10.1158/1078-0432.CCR-11-1157]

81

Eder JP, Shapiro GI, Appleman LJ, Zhu AX, Miles D, Keer H, Cancilla B, Chu F, Hitchcock-Bryan S, Sherman L, McCallum S, Heath EI, Boerner SA, LoRusso PM. A phase I study of foretinib, a multi-targeted inhibitor of c-Met and vascular endothelial growth factor receptor 2. Clin Cancer Res 2010; 16: 3507-3516 [PMID: 20472683 DOI: 10.1158/1078-0432.CCR-10-0574]

82

Belalcazar A, Azaña D, Perez CA, Raez LE, Santos ES. Targeting the Met pathway in lung cancer. Expert Rev Anticancer Ther 2012; 12: 519-528 [PMID: 22500688 DOI: 10.1586/era.12.16]

83

Sharma S, Zeng JY, Zhuang CM, Zhou YQ, Yao HP, Hu X, Zhang R, Wang MH. Small-molecule inhibitor BMS-777607 induces breast cancer cell polyploidy with increased resistance to cytotoxic chemotherapy agents. Mol Cancer Ther 2013; 12: 725-736 [PMID: 23468529 DOI: 10.1158/1535-7163.MCT-12-1079]

84

Pan BS, Chan GK, Chenard M, Chi A, Davis LJ, Deshmukh SV, Gibbs JB, Gil S, Hang G, Hatch H, Jewell JP, Kariv I, Katz JD, Kunii K, Lu W, Lutterbach BA, Paweletz CP, Qu X, Reilly JF, Szewczak AA, Zeng Q, Kohl NE, Dinsmore CJ. MK-2461, a novel multitargeted kinase inhibitor, preferentially inhibits the activated c-Met receptor. Cancer Res 2010; 70: 1524-1533 [PMID: 20145145 DOI: 10.1158/0008-5472.CAN-09-2541]

85

Northrup AB, Katcher MH, Altman MD, Chenard M, Daniels MH, Deshmukh SV, Falcone D, Guerin DJ, Hatch H, Li C, Lu W, Lutterbach B, Allison TJ, Patel SB, Reilly JF, Reutershan M, Rickert KW, Rosenstein C, Soisson SM, Szewczak AA, Walker D, Wilson K, Young JR, Pan BS, Dinsmore CJ. Discovery of 1-[3-(1-methyl-1H-pyrazol-4-yl)-5-oxo-5H-benzo[4,5]cyclohepta[1,2-b]pyridin-7-yl]-N-(pyridin-2-ylmethyl)methanesulfonamide (MK-8033): A Specific c-Met/Ron dual kinase inhibitor with preferential affinity for the activated state of c-Met. J Med Chem 2013; 56: 2294-2310 [PMID: 23379595 DOI: 10.1021/jm301619u]

86

Comoglio PM, Giordano S, Trusolino L. Drug development of MET inhibitors: targeting oncogene addiction and expedience. Nat Rev Drug Discov 2008; 7: 504-516 [PMID: 18511928 DOI: 10.1038/nrd2530]

87

Qin S, Chan SL, Sukeepaisarnjaroen W, Han G, Choo SP, Sriuranpong V, Pan H, Yau T, Guo Y, Chen M, Ren Z, Xu J, Yen CJ, Lin ZZ, Manenti L, Gu Y, Sun Y, Tiedt R, Hao L, Song W, Tanwandee T. A phase II study of the efficacy and safety of the MET inhibitor capmatinib (INC280) in patients with advanced hepatocellular carcinoma. Ther Adv Med Oncol 2019; 11: 1758835919889001 [PMID: 31853265 DOI: 10.1177/1758835919889001]

88

Rimassa L, Assenat E, Peck-Radosavljevic M, Pracht M, Zagonel V, Mathurin P, Rota Caremoli E, Porta C, Daniele B, Bolondi L, Mazzaferro V, Harris W, Damjanov N, Pastorelli D, Reig M, Knox J, Negri F, Trojan J, López López C, Personeni N, Decaens T, Dupuy M, Sieghart W, Abbadessa G, Schwartz B, Lamar M, Goldberg T, Shuster D, Santoro A, Bruix J. Tivantinib for second-line treatment of MET-high, advanced hepatocellular carcinoma (METIV-HCC): a final analysis of a phase 3, randomised, placebo-controlled study. Lancet Oncol 2018; 19: 682-693 [PMID: 29625879 DOI: 10.1016/S1470-2045(18)30146-3]

89

Guin S, Ma Q, Padhye S, Zhou YQ, Yao HP, Wang MH. Targeting acute hypoxic cancer cells by doxorubicin-immunoliposomes directed by monoclonal antibodies specific to RON receptor tyrosine kinase. Cancer Chemother Pharmacol 2011; 67: 1073-1083 [PMID: 20658288 DOI: 10.1007/s00280-010-1408-8]

90

Feng L, Yao HP, Wang W, Zhou YQ, Zhou J, Zhang R, Wang MH. Efficacy of anti-RON antibody Zt/g4-drug maytansinoid conjugation (Anti-RON ADC) as a novel therapeutics for targeted colorectal cancer therapy. Clin Cancer Res 2014; 20: 6045-6058 [PMID: 25294907 DOI: 10.1158/1078-0432.CCR-14-0898]

91

Yao HP, Feng L, Suthe SR, Chen LH, Weng TH, Hu CY, Jun ES, Wu ZG, Wang WL, Kim SC, Tong XM, Wang MH. Therapeutic efficacy, pharmacokinetic profiles, and toxicological activities of humanized antibody-drug conjugate Zt/g4-MMAE targeting RON receptor tyrosine kinase for cancer therapy. J Immunother Cancer 2019; 7: 75 [PMID: 30871619 DOI: 10.1186/s40425-019-0525-0]

92

Yang CY, Wang L, Sun X, Tang M, Quan HT, Zhang LS, Lou LG, Gou SH. SHR-A1403, a novel c-Met antibody-drug conjugate, exerts encouraging anti-tumor activity in c-Met-overexpressing models. Acta Pharmacol Sin 2019; 40: 971-979 [PMID: 30643210 DOI: 10.1038/s41401-018-0198-0]

93






https://dx.doi.org/10.1586/era.12.16


https://dx.doi.org/10.1158/1535-7163.MCT-12-1079




https://dx.doi.org/10.1021/jm301619u


https://dx.doi.org/10.1038/nrd2530


https://dx.doi.org/10.1177/1758835919889001


https://dx.doi.org/10.1016/S1470-2045(18)30146-3


https://dx.doi.org/10.1007/s00280-010-1408-8




https://dx.doi.org/10.1186/s40425-019-0525-0


https://dx.doi.org/10.1038/s41401-018-0198-0


World Journal of



MINIREVIEWS

Evolving role of endoscopy in inflammatory bowel disease: Going beyond diagnosis

Paulina Núñez F, Noa Krugliak Cleveland, Rodrigo Quera, David T Rubin

ORCID number: Paulina Núñez F 0000-0003-3727-1851; Noa Krugliak Cleveland 0000-0003-1707-8560; Rodrigo Quera 0000-0001-5854-0526; David T Rubin 0000-0001-5647-1723.

Author contributions: All authors equally contributed to this review with the conception and design of the study, literature review and analysis, drafting and critical revision and editing, and approval of the final version.

Conflict-of-interest statement: Rubin DT has received grant support from Takeda; and has served as a consultant for Abbvie, Altrubio, Allergan Inc., Arena Pharmaceuticals, Bellatrix Pharmaceuticals, Boehringer Ingelheim Ltd., Bristol-Myers Squibb, Celgene Corp/Syneos, Connect BioPharma, GalenPharma/Atlantica, Genentech/Roche, Gilead Sciences, InDex Pharmaceuticals, Ironwood Pharmaceuticals, Iterative Scopes, Janssen Pharmaceuticals, Lilly, Materia Prima, Pfizer, Prometheus Biosciences, Reistone, Takeda, and Techlab Inc. Quera R and Nuñez F Phave received support for attending meetings from Janssen.

Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external

Paulina Núñez F, Rodrigo Quera, Department of Gastroenterology, Inflammatory Bowel Disease Program, Clinica Universidad de los Andes, Santiago 7620157, RM, Chile

Paulina Núñez F, Department of Gastroenterology, Hospital San Juan de Dios, Santiago 8350488, RM, Chile

Noa Krugliak Cleveland, David T Rubin, University of Chicago Medicine Inflammatory Bowel Disease Center, Chicago, IL 60637, United States

Corresponding author: David T Rubin, MD Chief, Section of Gastroenterology, Hepatology and Nutrition.University of Chicago Medicine Inflammatory Bowel Disease Center, 5841 S. Maryland Ave., MC4076, Room M410, Chicago, IL 60637, United States. [email protected]

AbstractInflammatory bowel disease, encompassing Crohn’s disease (CD) and ulcerative colitis, are chronic immune-mediated inflammatory bowel diseases (IBD) that primarily affect the gastrointestinal tract with periods of activity and remission. Large body of evidence exist to strengthen the prognostic role of endoscopic evaluation for both disease activity and severity and it remains the gold standard for the assessment of mucosal healing. Mucosal healing has been associated with improved clinical outcomes with prolonged remission, decreased hospitalization, IBD-related surgeries and colorectal cancer risk. Therefore, endoscopic objectives in IBD have been incorporated as part of standard care. With the known increased risk of colorectal cancer in IBD, although prevention strategies continue to develop, regular surveillance for early detection of neoplasia continue to be paramount in IBD patients’ care. It is thanks to evolving technology and visual-ization techniques that surveillance strategies are continuously advancing. Therapeutic endoscopic options in IBD have also been expanding, from surgery sparing therapies such as balloon dilation of fibrostenotic strictures in CD to endoscopic mucosal resection of neoplastic lesions. In this review article, we discuss the current evidence on the use of endoscopy as part of standard of care of IBD, its role in surveillance of neoplasia, and the role of interventional endoscopic therapies.

Key Words: Inflammatory bowel disease; Endoscopy; Crohn’s disease; Ulcerative colitis; Therapeutic endoscopy; Surveillance



http://orcid.org/0000-0003-3727-1851

http://orcid.org/0000-0003-3727-1851

http://orcid.org/0000-0003-1707-8560

http://orcid.org/0000-0003-1707-8560

http://orcid.org/0000-0001-5854-0526

http://orcid.org/0000-0001-5854-0526

http://orcid.org/0000-0001-5647-1723

http://orcid.org/0000-0001-5647-1723

http://orcid.org/0000-0001-5647-1723


Núñez F P et al. Role of endoscopy in IBD


reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/Licenses/by-nc/4.0/



Country/Territory of origin: United States

Peer-review report’s scientific quality classificationGrade A (Excellent): A Grade B (Very good): B Grade C (Good): 0 Grade D (Fair): 0 Grade E (Poor): 0

Received: January 22, 2021 Peer-review started: January 22, 2021 First decision: March 29, 2021 Revised: April 11, 2021 Accepted: April 26, 2021 Article in press: April 26, 2021 Published online: May 28, 2021

P-Reviewer: Fan RY, Joshi H S-Editor: Gao CC L-Editor: A P-Editor: Liu JH


Core Tip: Endoscopy is critical to the diagnosis of inflammatory bowel diseases (IBD) and is increasingly being used for disease monitoring and management to achieve the therapeutic goal of mucosa healing. In this review, we focus on the utility of endoscopy as a therapeutic objective, in disease monitoring, and in surveillance to detect and prevent neoplasia. We will also discuss the evolving role of endoscopic therapeutic interventions in IBD.

Citation: Núñez F P, Krugliak Cleveland N, Quera R, Rubin DT. Evolving role of endoscopy in inflammatory bowel disease: Going beyond diagnosis. World J Gastroenterol 2021; 27(20): 2521-2530URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2521.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2521

INTRODUCTIONInflammatory bowel diseases (IBD), including ulcerative colitis (UC) and Crohn's disease (CD) are immune-mediated inflammatory diseases that primarily affect the gastrointestinal tract with periods of activity and remission[1,2]. The diagnosis of IBD requires endoscopic and histological evaluation of the intestinal mucosa, combined with the proper clinical presentation and setting[3]. Endoscopy allows us to differ-entiate CD from UC and define its severity by visualization of the colonic mucosa, its vascular pattern, mucosal lesions, as well as patterns locations of inflammation including assessment of involvement of the terminal ileum and perianal region[4].

In the absence of effective treatment, intestinal inflammation is often progressive and cumulative, leading to complications such as strictures, fistulas that require surgery, and in the long term, the risk of dysplasia or colorectal cancer (CRC)[5]. There is also evidence that such cumulative damage may result in mucosal dysfunction, with dysmotility and hypersensitivity[6]. Therefore, endoscopic control has become part of the objective treatment pillar of IBD and mucosa healing is a preferred goal of treatment[7].

In this review we will focus on the usefulness of endoscopy as a therapeutic objective, in follow-up and surveillance to detect and to prevent the development of dysplasia and CRC, and its evolving utility in interventional therapies.

ENDOSCOPY TO ASSESS DISEASE ACTIVITYMucosal healing is recommended as a therapeutic objective in patients with IBD. This objective is associated with a better prognosis, lower rates of hospitalizations, lower risk of relapse and the need for surgery[8,9]. After the STRIDE consensus (selecting therapeutic targets in inflammatory bowel disease), it is recommended to evaluate the colonic mucosa looking for the resolution of ulcers and the friability of the mucosa 6-9 mo after starting therapy in CD and 3-6 mo in UC[10].

Other reviews have supported this approach as well[11,12]. It has been suggested that endoscopic remission in UC should be defined as a Ulcerative Colitis Endoscopic Index of Severity (UCEIS) of 0 or a Mayo Endoscopic Score (MES) of 0 or 1. However, in follow-up studies, a higher percentage/risk of relapse has been observed in patients who had reached a MES 1 index compared to the group with a MES 0 index (36.6% vs 9.4% with P < 0.001)[13]. On the other hand, an endoscopic response has been defined as a decrease in ≥ 1 degrees of the MES or a decrease in ≥ 2 points of the UCEIS[14,15].

The Crohn's Disease Endoscopic Index of Severity (CDEIS) and Simple Endoscopic Score for Crohn's Disease (SES-CD) are validated and reproducible indices in CD[16]. However, remission thresholds have been arbitrarily determined considering endoscopic remission with a SES-CD 0-2 index and after surgery a Rutgeerts i0-i1[17,18]. Endoscopic response is defined as a decrease greater than 50% in SES-CD or CDEIS[19] (Figure 1). STRIDE-II maintained endoscopic criteria for UC and CD[20].








Figure 1 Endoscopic index in inflammatory bowel diseases[13-22]. The Simple endoscopic score Crohn´s disease: Sum of values of the four variables for the five bowel segments (rectum, left colon, transverse colon, right colon and ileum). UCEIS: Ulcerative colitis endoscopic index of severity; SES CD: Simple endoscopic score Crohn´s disease; IBD: Inflammatory bowel diseases.

One of the most important postoperative endoscopy studies was the multicenter POCER (postoperative Crohn's endoscopic recurrence), which included 17 centers in Australia and New Zealand[21]. Authors recommend performing a control endoscopy six months after surgery, regardless of the risk factors for recurrence, since approx-imately 60% may recur (80% if they have high risk factors vs 30% in those with low risk). In this control endoscopy the anastomosis and the terminal ileum are to be evaluated and assessed for recurrence. High risk factors for recurrence include the following: disease in those under 30 years of age, a penetrating phenotype, presence of perianal disease or having two or more surgeries[22]. The therapeutic objectives in endoscopy are fundamental to avoid future complications of the disease, and are summarized in Figure 2.

ENDOSCOPY FOR SCREENING AND SURVEILLANCE FOR COLORECTAL NEOPLASIACRC is the third most frequent neoplasm in the general population, with an incidence of 1.8 million new cases per year, being the second cause of cancer mortality in the world[23]. IBD presents a higher risk of CRC, and traditionally it was considered that patients with IBD had a 2% increase in CRC risk at 10 years of disease, 8% at 20 years and 18% after 30 years of evolution[24]. However, the estimated risk of neoplasia in recent years has been lower, possibly due to more effective therapies[25] as well as better prevention strategies. A subsequent meta-analysis identified a 2.4-fold risk of CRC in UC patients, with a higher risk in male patients, younger age at diagnosis and extensive disease. Cumulative absolute risk of developing CRC was 0.4 % at 10 years and 1.1%- 5.3% at 20 years[26].

Other risk factors to consider are a family history of CRC or its association with primary sclerosing cholangitis[27]. Studies in CD have identified risk factors similar to those in UC, including diagnosis at an early age, prolonged disease, distal location, and penetrating phenotype[28]. Perianal fistulas have also been identified to have a



Figure 2 Endoscopic objectives in inflammatory bowel diseases treatment[8-22]. CCR: Colorectal cancer; UC: Ulcerative colitis; CD: Crohn’s disease; IPAA: Ileal-pouch-anal anastomoses; SES-CD: Simple endoscopic score Crohn´s disease; UCEIS: Ulcerative colitis endoscopic index of severity.

risk of neoplastic transformation[29] (Table 1).A screening endoscopic evaluation for dysplasia or colon cancer has been

recommended 8 years after diagnosis in UC patients with an extension beyond the rectum or in CD patients with an extension over 30% of the colon or that compromises more than one segment[30]. In 2017, a Cochrane systematic review reported that those patients who were in endoscopic surveillance had a lower mortality associated with CRC compared to those who were not followed up [8.5% vs 22.3%, respectively; odds ratio (OR) 0.36, 95% confidence interval (CI): 0.19-0.69][31].

Endoscopic surveillance requires optimal control of the disease, including mucosal healing. This will allow for early recognition of neoplasia[3]. It has even been recommended that surveillance should be performed with fecal calprotectin levels under 100 μg/g[32] to improve visualization. It should not be forgotten that optimal bowel preparation is essential for the detection of lesions[33].

Dysplasia is defined as a neoplastic alteration of the intestinal epithelium that remains restricted within the basement membrane, without invasion of the lamina propria. These lesions can be low (LGD) or high grade (HGD). The distinction between these two depends on the distribution of nuclei within the mucosa. LGD maintains hyperchromatic nuclei located in the basal half of cells, whereas HGD presents nuclear stratification and loss of cell polarity[34].

Lesions found can be polypoid or flat, the latter being the most frequent. Flat lesions are associated with higher rate of neoplasia, common in regions that are or have been involved by IBD[35]. These lesions tend to be subtle and can be multifocal, requiring meticulous and trained endoscopists[36]. Furthermore, since dysplasia can be difficult to distinguish from epithelial regeneration secondary to inflammation[37], it is recommended that biopsies are evaluated by two expert pathologists[38].

DYSPLASIA DETECTION METHODSDye-based chromoendoscopy (DCE) is an image-enhanced endoscopic technique in which topical dyes such as methylene blue or indigo carmine are applied allowing a detailed view of the mucosa and a targeted evaluation of suspicious lesions[39]. It has reported a higher performance in detecting dysplasia in the analysis per patient OR 2.05 (95%CI: 1.26-3.35) and analysis by type of lesion OR 2.79 (95%CI: 2.08-3.73)[40]. On the other hand, Feuerstein et al[41] observed that DCE was more effective in identifying dysplasia compared to standard white light endoscopy (WLE), but without reaching significant differences compared to high-definition WLE (HD-WLE). Recently, a retrospective analysis, has shown no difference in the detection of dysplasia using DCE compared to HD-WLE, although withdrawal times were longer



Table 1 Risk factors for colitis associated neoplasia[26-29]

CRC risk

Disease related factorsPatient related factors

UC CD

Male sex; Longer duration of the disease or younger age of diagnosis; Prior History of colon neoplasia; History of CRC in first degree relative; Especially if < 50 yr old; Primary sclerosing colangitis

Extensive colitis; Severity of inflammation; Early onset of disease; Mass/stenosis; Pseudopolyps

Early onset of disease; Younger age of diagnosis; Perianal disease; Colonic extension; Bypassed segments of bowel strictures

UC: Ulcerative colitis; CD: Crohn’s disease; CCR: Colorectal cancer.

with DCE (24.6 min vs 15.4 min, P < 0.001)[42].SCENIC (Surveillance for Colorectal Endoscopic Neoplasia Detection and

Management in Inflammatory Bowel Disease Patients: International Consensus) recommends a surveillance study with high-definition colonoscopy or chromoen-doscopy when standard white-light exam are performed[30]. The American College of Gastroenterology 2019 guidelines suggest targeted biopsies as an option with high-definition scopes. This approach is evolving as the technology is advancing. In the absence of augmenting imaging by high-definition or chromo, systematic sampling of the mucosa with random biopsies are recommended for every 10 cm of colon[3,37].

Virtual chromoendoscopy (VCE) [which includes narrow banding imaging (Oympus NBI), i-SCAN (Pentax) and Fujinon Fuji Intelligent Chromo-Endoscopy] is an optical imaging technique that uses filters to enhance the contrast of both the mucosa and the superficial vasculature, allowing a better evaluation of the mucosa. Bisschops et al[43], conducted a multicenter study with UC patients comparing DCE vs NBI. No significant difference was reported between these techniques in detecting neoplastic lesions (OR 1.02, 95%CI: 0.44-2.35, P = 0.964). The SCENIC consensus guidelines state that VCE should not replace DCE. Undoubtedly, larger studies are lacking to evaluate the usefulness of these techniques. The 2019 ACG guidelines recommend the use of DCE or NBI for the surveillance of dysplasia (conditional recommendation, low quality of evidence)[44].

MANAGEMENT OF DYSPLASIAOnce dysplasia has been identified, its respectability is essential to interrupt the carcinogenic sequence and thus reduce the incidence of CRC. The different management guidelines indicate that active surveillance and endoscopic follow-up should be performed according to the type of dysplasia found and the patient's risk factors. In Figures 3 and 4 the subsequent management is summarized. In selected patients, segmental resection without proctocolectomy is possible[45].

ENDOSCOPY FOR THERAPEUTIC INTERVENTIONSCurrently, therapeutic endoscopic interventions are considered in four areas: stenosis, fistulas, complications associated with surgery and neoplasms associated with colitis[46]. We will review the latter in a separate chapter.

StenosisFor its diagnosis, an endoscopic evaluation and radiologic exams are required, which may be a magnetic resonance enterography or an abdomen-pelvis computed tomographic enterography[47]. It is important to record the number of stenoses, their location, their composition (inflammatory, fibrotic), morphology, size, as well as the detection of complications such as abscesses or fistulas[48].

There are three endoscopic therapeutic options in those stenoses smaller than 5 cm; endoscopic balloon dilation, endoscopic stricturotomy or stent placement, the latter with a very low level of evidence[49]. Endoscopic balloon dilation would have a lower risk of bleeding, but a higher risk of perforation. However, retrospective studies have shown it is a safe technique in patients with CD, with more than 40% of patients asymptomatic and without requiring surgery in a subsequent follow-up[50].



Figure 3 Inflammatory bowel diseases surveillance[3,24-28]. CCR: Colorectal cancer; IBD: Inflammatory bowel diseases; UC: Ulcerative colitis.

Figure 4 Algorithm for the management of dysplasia[3,37,42-46]. LGD: Low grade dysplasia; HGD: High grade dysplasia; CE: Chromoendoscopy.

Endoscopic stricturotomy is in evolution and may be an effective procedure in treating fibrotic, distal or anastomotic stenosis. The technique is based on an electro-incision, allowing control of the depth and location of the cut, and with a lower risk of perforation. In a small reported experience surgery free survival in IBD to non-IBD patients undergoing endoscopic stricturotomy, there was no statistically significant difference between the groups[51].

Endoscopic injections of steroids or anti-tumor necrosis factor agents in conjunction with endoscopic balloon dilation have been reported in series of cases with the intention of reducing the need for future dilation with inconsistent results[52,53].

In recurrent or refractory stenoses, self-expanding metal stents, covered metal stents or biodegradable metal stents have been used with good results[54,55]. These can be placed endoscopically with or without fluoroscopic guidance but must be maintained for at least 4 wk[56].



FistulasThe penetrating phenotype may be primary or a result of a long-standing CD. The goals of endoscopic treatment are drainage, closure of the fistulas and preventing them from becoming complex[48]. In a cohort of 29 patients, about 90% achieved resolution of their fistulas by endoscopic fistulotomy[57]. This type of therapy might be performed in superficial, short and enteroenteric fistulas[56]. In addition, it is possible to close the fistula endoscopically by means of a clip, avoiding the formation of abscesses. This has been reported to be achieved with through-the-scope clips or over-the-scope clips, even with reports in perianal fistulas[48]. Further work is needed in this area, with clarification of risks, benefits, and approach to combination with medical interventions.

Post-operative complicationsPost-surgery complications may present as dehiscence of the suture/staple line or present later with stenosis of the anastomosis, causing obstructive problems[58]. Endoscopic management of suture dehiscence has been described in clinical cases or case series, reporting lumen integrity in over 80%[59]. Approximately 11% of patients with UC require an ileal pouch anal anastomoses for their management[60], where strictures may develop in anastomosis, in the pouch or in the afferent loop. Postoperative strictures can be managed endoscopically with balloon dilations or stricturotomy. Endoscopic interventions should be avoided in periods of increased inflammation given the increased risk of perforation[21].

CONCLUSIONEndoscopic evaluation is essential to achieve current therapeutic goals in IBD, which focus on mucosal healing[61,62]. In cases in which the objectives are not achieved, endoscopic assessments direct therapeutic optimization in order to achieve them. Importantly, endoscopic assessment has a key role in prognostication and disease monitoring to control modifiable risk factors for worse clinical outcomes, with a focus on control of inflammation and prevention of complications. Among these, the incidence of CRC has a significant impact on patient morbidity and mortality, and its early diagnosis is a crucial element in IBD management. The evolution of endoscopic visualization techniques, such as the use of VCE and DCE, allow a better detection and characterization of premalignant endoscopic lesions, avoiding the risks of advanced stages[63]. Additionally, evolving endoscopic therapies allow greater minimally invasive therapeutic options for the treatment of stenosis and fistulas as an alternative to surgical management with lower morbidity.

REFERENCESUngaro R, Mehandru S, Allen PB, Peyrin-Biroulet L, Colombel JF. Ulcerative colitis. Lancet 2017; 389: 1756-1770 [PMID: 27914657 DOI: 10.1016/S0140-6736(16)32126-2]

1

Loftus EV Jr. Crohn's Disease: Etiology, Complications, Assessment, Therapy, and Management. Gastroenterol Clin North Am 2017; 46: xiii-xixv [PMID: 28838422 DOI: 10.1016/j.gtc.2017.06.001]

2

Maaser C, Sturm A, Vavricka SR, Kucharzik T, Fiorino G, Annese V, Calabrese E, Baumgart DC, Bettenworth D, Borralho Nunes P, Burisch J, Castiglione F, Eliakim R, Ellul P, González-Lama Y, Gordon H, Halligan S, Katsanos K, Kopylov U, Kotze PG, Krustinš E, Laghi A, Limdi JK, Rieder F, Rimola J, Taylor SA, Tolan D, van Rheenen P, Verstockt B, Stoker J; European Crohn’s and Colitis Organisation [ECCO] and the European Society of Gastrointestinal and Abdominal Radiology [ESGAR]. ECCO-ESGAR Guideline for Diagnostic Assessment in IBD Part 1: Initial diagnosis, monitoring of known IBD, detection of complications. J Crohns Colitis 2019; 13: 144-164 [PMID: 30137275 DOI: 10.1093/ecco-jcc/jjy113]

3

Zilli A, Capogreco A, Furfaro F, Allocca M, Roda G, Loy L, Fiorino G, Danese S. Improving quality of care in endoscopy of inflammatory bowel disease: can we do better? Expert Rev Gastroenterol Hepatol 2020; 14: 819-828 [PMID: 32543983 DOI: 10.1080/17474124.2020.1780913]

4

Vatn MH. Natural history and complications of IBD. Curr Gastroenterol Rep 2009; 11: 481-487 [PMID: 19903424 DOI: 10.1007/s11894-009-0073-8]

5

Torres J, Billioud V, Sachar DB, Peyrin-Biroulet L, Colombel JF. Ulcerative colitis as a progressive disease: the forgotten evidence. Inflamm Bowel Dis 2012; 18: 1356-1363 [PMID: 22162423 DOI: 10.1002/ibd.22839]

6

Nuñez F P, Mahadevan U, Quera R, Bay C, Ibañez P. Treat-to-target approach in the management of inflammatory Bowel disease. Gastroenterol Hepatol 2021; 44: 312-319 [PMID: 33070988 DOI:

7


https://dx.doi.org/10.1016/S0140-6736(16)32126-2


https://dx.doi.org/10.1016/j.gtc.2017.06.001


https://dx.doi.org/10.1093/ecco-jcc/jjy113


https://dx.doi.org/10.1080/17474124.2020.1780913


https://dx.doi.org/10.1007/s11894-009-0073-8


https://dx.doi.org/10.1002/ibd.22839




10.1016/j.gastrohep.2020.06.032]Colombel JF, Rutgeerts P, Reinisch W, Esser D, Wang Y, Lang Y, Marano CW, Strauss R, Oddens BJ, Feagan BG, Hanauer SB, Lichtenstein GR, Present D, Sands BE, Sandborn WJ. Early mucosal healing with infliximab is associated with improved long-term clinical outcomes in ulcerative colitis. Gastroenterology 2011; 141: 1194-1201 [PMID: 21723220 DOI: 10.1053/j.gastro.2011.06.054]

8

Colombel JF, Rutgeerts PJ, Sandborn WJ, Yang M, Camez A, Pollack PF, Thakkar RB, Robinson AM, Chen N, Mulani PM, Chao J. Adalimumab induces deep remission in patients with Crohn's disease. Clin Gastroenterol Hepatol 2014; 12: 414-22. e5 [PMID: 23856361 DOI: 10.1016/j.cgh.2013.06.019]

9

Peyrin-Biroulet L, Sandborn W, Sands BE, Reinisch W, Bemelman W, Bryant RV, D'Haens G, Dotan I, Dubinsky M, Feagan B, Fiorino G, Gearry R, Krishnareddy S, Lakatos PL, Loftus EV Jr, Marteau P, Munkholm P, Murdoch TB, Ordás I, Panaccione R, Riddell RH, Ruel J, Rubin DT, Samaan M, Siegel CA, Silverberg MS, Stoker J, Schreiber S, Travis S, Van Assche G, Danese S, Panes J, Bouguen G, O'Donnell S, Pariente B, Winer S, Hanauer S, Colombel JF. Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE): Determining Therapeutic Goals for Treat-to-Target. Am J Gastroenterol 2015; 110: 1324-1338 [PMID: 26303131 DOI: 10.1038/ajg.2015.233]

10

Ungaro R, Colombel JF, Lissoos T, Peyrin-Biroulet L. A Treat-to-Target Update in Ulcerative Colitis: A Systematic Review. Am J Gastroenterol 2019; 114: 874-883 [PMID: 30908297 DOI: 10.14309/ajg.0000000000000183]

11

Rutgeerts P, Diamond RH, Bala M, Olson A, Lichtenstein GR, Bao W, Patel K, Wolf DC, Safdi M, Colombel JF, Lashner B, Hanauer SB. Scheduled maintenance treatment with infliximab is superior to episodic treatment for the healing of mucosal ulceration associated with Crohn's disease. Gastrointest Endosc 2006; 63: 433-442; quiz 464 [PMID: 16500392 DOI: 10.1016/j.gie.2005.08.011]

12

Barreiro-de Acosta M, Vallejo N, de la Iglesia D, Uribarri L, Bastón I, Ferreiro-Iglesias R, Lorenzo A, Domínguez-Muñoz JE. Evaluation of the Risk of Relapse in Ulcerative Colitis According to the Degree of Mucosal Healing (Mayo 0 vs 1): A Longitudinal Cohort Study. J Crohns Colitis 2016; 10: 13-19 [PMID: 26351390 DOI: 10.1093/ecco-jcc/jjv158]

13

Vuitton L, Peyrin-Biroulet L, Colombel JF, Pariente B, Pineton de Chambrun G, Walsh AJ, Panes J, Travis SP, Mary JY, Marteau P. Defining endoscopic response and remission in ulcerative colitis clinical trials: an international consensus. Aliment Pharmacol Ther 2017; 45: 801-813 [PMID: 28112419 DOI: 10.1111/apt.13948]

14

Negreanu L, Voiosu T, State M, Voiosu A, Bengus A, Mateescu BR. Endoscopy in inflammatory bowel disease: from guidelines to real life. Therap Adv Gastroenterol 2019; 12: 1756284819865153 [PMID: 31384307 DOI: 10.1177/1756284819865153]

15

Daperno M, Comberlato M, Bossa F, Biancone L, Bonanomi AG, Cassinotti A, Cosintino R, Lombardi G, Mangiarotti R, Papa A, Pica R, Rizzello F, D'Incà R, Orlando A. Inter-observer agreement in endoscopic scoring systems: preliminary report of an ongoing study from the Italian Group for Inflammatory Bowel Disease (IG-IBD). Dig Liver Dis 2014; 46: 969-973 [PMID: 25154049 DOI: 10.1016/j.dld.2014.07.010]

16

Rutgeerts P, Geboes K, Vantrappen G, Beyls J, Kerremans R, Hiele M. Predictability of the postoperative course of Crohn's disease. Gastroenterology 1990; 99: 956-963 [PMID: 2394349 DOI: 10.1016/0016-5085(90)90613-6]

17

Rivière P, Vermeire S, Irles-Depe M, Van Assche G, Rutgeerts P, de Buck van Overstraeten A, Denost Q, Wolthuis A, D'Hoore A, Laharie D, Ferrante M. No Change in Determining Crohn's Disease Recurrence or Need for Endoscopic or Surgical Intervention With Modification of the Rutgeerts' Scoring System. Clin Gastroenterol Hepatol 2019; 17: 1643-1645 [PMID: 30291910 DOI: 10.1016/j.cgh.2018.09.047]

18

Vuitton L, Marteau P, Sandborn WJ, Levesque BG, Feagan B, Vermeire S, Danese S, D'Haens G, Lowenberg M, Khanna R, Fiorino G, Travis S, Mary JY, Peyrin-Biroulet L. IOIBD technical review on endoscopic indices for Crohn's disease clinical trials. Gut 2016; 65: 1447-1455 [PMID: 26353983 DOI: 10.1136/gutjnl-2015-309903]

19

Turner D, Ricciuto A, Lewis A, D'Amico F, Dhaliwal J, Griffiths AM, Bettenworth D, Sandborn WJ, Sands BE, Reinisch W, Schölmerich J, Bemelman W, Danese S, Mary JY, Rubin D, Colombel JF, Peyrin-Biroulet L, Dotan I, Abreu MT, Dignass A; International Organization for the Study of IBD. STRIDE-II: An Update on the Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE) Initiative of the International Organization for the Study of IBD (IOIBD): Determining Therapeutic Goals for Treat-to-Target strategies in IBD. Gastroenterology 2021; 160: 1570-1583 [PMID: 33359090 DOI: 10.1053/j.gastro.2020.12.031]

20

De Cruz P, Kamm MA, Hamilton AL, Ritchie KJ, Krejany EO, Gorelik A, Liew D, Prideaux L, Lawrance IC, Andrews JM, Bampton PA, Gibson PR, Sparrow M, Leong RW, Florin TH, Gearry RB, Radford-Smith G, Macrae FA, Debinski H, Selby W, Kronborg I, Johnston MJ, Woods R, Elliott PR, Bell SJ, Brown SJ, Connell WR, Desmond PV. Crohn's disease management after intestinal resection: a randomised trial. Lancet 2015; 385: 1406-1417 [PMID: 25542620 DOI: 10.1016/S0140-6736(14)61908-5]

21

Bhattacharya A, Shen B, Regueiro M. Endoscopy in Postoperative Patients with Crohn's Disease or Ulcerative Colitis. Does It Translate to Better Outcomes? Gastrointest Endosc Clin N Am 2019; 29: 487-514 [PMID: 31078249 DOI: 10.1016/j.giec.2019.02.013]

22

Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: 23

https://dx.doi.org/10.1016/j.gastrohep.2020.06.032








https://dx.doi.org/10.14309/ajg.0000000000000183




https://dx.doi.org/10.1093/ecco-jcc/jjv158




https://dx.doi.org/10.1177/1756284819865153


https://dx.doi.org/10.1016/j.dld.2014.07.010


https://dx.doi.org/10.1016/0016-5085(90)90613-6








https://dx.doi.org/10.1016/S0140-6736(14)61908-5


https://dx.doi.org/10.1016/j.giec.2019.02.013



GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018; 68: 394-424 [PMID: 30207593 DOI: 10.3322/caac.21492]Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gut 2001; 48: 526-535 [PMID: 11247898 DOI: 10.1136/gut.48.4.526]

24

Herrinton LJ, Liu L, Levin TR, Allison JE, Lewis JD, Velayos F. Incidence and mortality of colorectal adenocarcinoma in persons with inflammatory bowel disease from 1998 to 2010. Gastroenterology 2012; 143: 382-389 [PMID: 22609382 DOI: 10.1053/j.gastro.2012.04.054]

25

Jess T, Rungoe C, Peyrin-Biroulet L. Risk of colorectal cancer in patients with ulcerative colitis: a meta-analysis of population-based cohort studies. Clin Gastroenterol Hepatol 2012; 10: 639-645 [PMID: 22289873 DOI: 10.1016/j.cgh.2012.01.010]

26

Beaugerie L, Itzkowitz SH. Cancers complicating inflammatory bowel disease. N Engl J Med 2015; 372: 1441-1452 [PMID: 25853748 DOI: 10.1056/NEJMra1403718]

27

Scaringi S, Di Martino C, Zambonin D, Fazi M, Canonico G, Leo F, Ficari F, Tonelli F. Colorectal cancer and Crohn's colitis: clinical implications from 313 surgical patients. World J Surg 2013; 37: 902-910 [PMID: 23381673 DOI: 10.1007/s00268-013-1922-z]

28

Connell WR, Sheffield JP, Kamm MA, Ritchie JK, Hawley PR, Lennard-Jones JE. Lower gastrointestinal malignancy in Crohn's disease. Gut 1994; 35: 347-352 [PMID: 8150345 DOI: 10.1136/gut.35.3.347]

29

Laine L, Kaltenbach T, Barkun A, McQuaid KR, Subramanian V, Soetikno R; SCENIC Guideline Development Panel. SCENIC international consensus statement on surveillance and management of dysplasia in inflammatory bowel disease. Gastroenterology 2015; 148: 639-651. e28 [PMID: 25702852 DOI: 10.1053/j.gastro.2015.01.031]

30

Bye WA, Nguyen TM, Parker CE, Jairath V, East JE. Strategies for detecting colon cancer in patients with inflammatory bowel disease. Cochrane Database Syst Rev 2017; 9: CD000279 [PMID: 28922695 DOI: 10.1002/14651858.CD000279.pub4]

31

Iacucci M, Cannatelli R, Tontini GE, Panaccione R, Danese S, Fiorino G, Matsumoto T, Kochhar GS, Shen B, Kiesslich R, Ghosh S. Improving the quality of surveillance colonoscopy in inflammatory bowel disease. Lancet Gastroenterol Hepatol 2019; 4: 971-983 [PMID: 31696831 DOI: 10.1016/S2468-1253(19)30194-3]

32

Megna B, Weiss J, Ley D, Saha S, Pfau P, Grimes I, Li Z, Caldera F. Clear liquid diet before bowel preparation predicts successful chromoendoscopy in patients with inflammatory bowel disease. Gastrointest Endosc 2019; 89: 373-379. e2 [PMID: 30339950 DOI: 10.1016/j.gie.2018.09.039]

33

Itzkowitz SH, Harpaz N. Diagnosis and management of dysplasia in patients with inflammatory bowel diseases. Gastroenterology 2004; 126: 1634-1648 [PMID: 15168373 DOI: 10.1053/j.gastro.2004.03.025]

34

Vieth M, Behrens H, Stolte M. Sporadic adenoma in ulcerative colitis: endoscopic resection is an adequate treatment. Gut 2006; 55: 1151-1155 [PMID: 16423892 DOI: 10.1136/gut.2005.075531]

35

Lakatos PL, Lakatos L. Risk for colorectal cancer in ulcerative colitis: changes, causes and management strategies. World J Gastroenterol 2008; 14: 3937-3947 [PMID: 18609676 DOI: 10.3748/wjg.14.3937]

36

Mattar MC, Lough D, Pishvaian MJ, Charabaty A. Current management of inflammatory bowel disease and colorectal cancer. Gastrointest Cancer Res 2011; 4: 53-61 [PMID: 21673876]

37

Cairns SR, Scholefield JH, Steele RJ, Dunlop MG, Thomas HJ, Evans GD, Eaden JA, Rutter MD, Atkin WP, Saunders BP, Lucassen A, Jenkins P, Fairclough PD, Woodhouse CR; British Society of Gastroenterology; Association of Coloproctology for Great Britain and Ireland. Guidelines for colorectal cancer screening and surveillance in moderate and high risk groups (update from 2002). Gut 2010; 59: 666-689 [PMID: 20427401 DOI: 10.1136/gut.2009.179804]

38

American Society for Gastrointestinal Endoscopy Standards of Practice Committee. , Shergill AK, Lightdale JR, Bruining DH, Acosta RD, Chandrasekhara V, Chathadi KV, Decker GA, Early DS, Evans JA, Fanelli RD, Fisher DA, Fonkalsrud L, Foley K, Hwang JH, Jue TL, Khashab MA, Muthusamy VR, Pasha SF, Saltzman JR, Sharaf R, Cash BD, DeWitt JM. The role of endoscopy in inflammatory bowel disease. Gastrointest Endosc 2015; 81: 1101-21. e1-13 [PMID: 25800660 DOI: 10.1016/j.gie.2014.10.030]

39

Har-Noy O, Katz L, Avni T, Battat R, Bessissow T, Yung DE, Engel T, Koulaouzidis A, Eliakim R, Ben-Horin S, Kopylov U. Chromoendoscopy, Narrow-Band Imaging or White Light Endoscopy for Neoplasia Detection in Inflammatory Bowel Diseases. Dig Dis Sci 2017; 62: 2982-2990 [PMID: 28965228 DOI: 10.1007/s10620-017-4772-y]

40

Feuerstein JD, Rakowsky S, Sattler L, Yadav A, Foromera J, Grossberg L, Cheifetz AS. Meta-analysis of dye-based chromoendoscopy compared with standard- and high-definition white-light endoscopy in patients with inflammatory bowel disease at increased risk of colon cancer. Gastrointest Endosc 2019; 90: 186-195. e1 [PMID: 31009609 DOI: 10.1016/j.gie.2019.04.219]

41

Clarke K, Kang M, Gorrepati VS, Stine JG, Tinsley A, Williams E, Moyer M, Coates M. Dysplasia detection is similar between chromoendoscopy and high-definition white-light colonoscopy in inflammatory bowel disease patients: a US-matched case-control study. Int J Colorectal Dis 2020; 35: 2301-2307 [PMID: 32812090 DOI: 10.1007/s00384-020-03719-3]

42

Bisschops R, Bessissow T, Joseph JA, Baert F, Ferrante M, Ballet V, Willekens H, Demedts I, Geboes K, De Hertogh G, Vermeire S, Rutgeerts P, Van Assche G. Chromoendoscopy vs narrow band imaging in UC: a prospective randomised controlled trial. Gut 2018; 67: 1087-1094 [PMID: 28698230 DOI: 10.1136/gutjnl-2016-313213]

43


https://dx.doi.org/10.3322/caac.21492


https://dx.doi.org/10.1136/gut.48.4.526
















https://dx.doi.org/10.1016/S2468-1253(19)30194-3






https://dx.doi.org/10.1136/gut.2005.075531


https://dx.doi.org/10.3748/wjg.14.3937



https://dx.doi.org/10.1136/gut.2009.179804








https://dx.doi.org/10.1007/s00384-020-03719-3





Rubin DT, Ananthakrishnan AN, Siegel CA, Sauer BG, Long MD. ACG Clinical Guideline: Ulcerative Colitis in Adults. Am J Gastroenterol 2019; 114: 384-413 [PMID: 30840605 DOI: 10.14309/ajg.0000000000000152]

44

Krugliak Cleveland N, Ollech JE, Colman RJ, Rodriquez D, Hirsch A, Cohen RD, Hanauer SB, Hart J, Hurst R, Rubin DT. Efficacy and Follow-up of Segmental or Subtotal Colectomy in Patients With Colitis-Associated Neoplasia. Clin Gastroenterol Hepatol 2019; 17: 205-206 [PMID: 29751167 DOI: 10.1016/j.cgh.2018.04.061]

45

Shen B, Kochhar G, Hull TL. Bridging Medical and Surgical Treatment of Inflammatory Bowel Disease: The Role of Interventional IBD. Am J Gastroenterol 2019; 114: 539-540 [PMID: 30464306 DOI: 10.1038/s41395-018-0416-x]

46

Fletcher JG, Fidler JL, Bruining DH, Huprich JE. New concepts in intestinal imaging for inflammatory bowel diseases. Gastroenterology 2011; 140: 1795-1806 [PMID: 21530746 DOI: 10.1053/j.gastro.2011.02.013]

47

Gu YB, Zhong J; Chinese IBD Endoscopic Club. Endoscopic management of stricturing Crohn's disease. J Dig Dis 2020; 21: 351-354 [PMID: 32537911 DOI: 10.1111/1751-2980.12914]

48

Shen B, Kochhar G, Navaneethan U, Liu X, Farraye FA, Gonzalez-Lama Y, Bruining D, Pardi DS, Lukas M, Bortlik M, Wu K, Sood A, Schwartz DA, Sandborn WJ; Global Interventional Inflammatory Bowel Disease Group. Role of interventional inflammatory bowel disease in the era of biologic therapy: a position statement from the Global Interventional IBD Group. Gastrointest Endosc 2019; 89: 215-237 [PMID: 30365985 DOI: 10.1016/j.gie.2018.09.045]

49

Tilmant M, Serrero M, Poullenot F, Bouguen G, Pariente B, Altwegg R, Basile P, Filippi J, Vanelslander P, Buisson A, Desjeux A, Laharie D, Le Balch E, Nachury M, Boivineau L, Savoye G, Hebuterne X, Poincloux L, Vuitton L, Brazier F, Yzet C, Lamrani A, Peyrin-Biroulet L, Fumery M. Endoscopic balloon dilation of colorectal strictures complicating Crohn's disease: a multicenter study. Clin Res Hepatol Gastroenterol 2020; 101561 [PMID: 33214090 DOI: 10.1016/j.clinre.2020.10.006]

50

Zhang LJ, Lan N, Wu XR, Shen B. Endoscopic stricturotomy in the treatment of anastomotic strictures in inflammatory bowel disease (IBD) and non-IBD patients. Gastroenterol Rep (Oxf) 2020; 8: 143-150 [PMID: 32280474 DOI: 10.1093/gastro/goz051]

51

Swaminath A, Lichtiger S. Dilation of colonic strictures by intralesional injection of infliximab in patients with Crohn's colitis. Inflamm Bowel Dis 2008; 14: 213-216 [PMID: 18022870 DOI: 10.1002/ibd.20318]

52

Di Nardo G, Oliva S, Passariello M, Pallotta N, Civitelli F, Frediani S, Gualdi G, Gandullia P, Mallardo S, Cucchiara S. Intralesional steroid injection after endoscopic balloon dilation in pediatric Crohn's disease with stricture: a prospective, randomized, double-blind, controlled trial. Gastrointest Endosc 2010; 72: 1201-1208 [PMID: 20951986 DOI: 10.1016/j.gie.2010.08.003]

53

Matsuhashi N, Nakajima A, Suzuki A, Yazaki Y, Takazoe M. Long-term outcome of non-surgical strictureplasty using metallic stents for intestinal strictures in Crohn's disease. Gastrointest Endosc 2000; 51: 343-345 [PMID: 10699786 DOI: 10.1016/s0016-5107(00)70366-x]

54

Rodrigues C, Oliveira A, Santos L, Pires E, Deus J. Biodegradable stent for the treatment of a colonic stricture in Crohn's disease. World J Gastrointest Endosc 2013; 5: 265-269 [PMID: 23678382 DOI: 10.4253/wjge.v5.i5.265]

55

Chen M, Shen B. Endoscopic therapy for Kock pouch strictures in patients with inflammatory bowel disease. Gastrointest Endosc 2014; 80: 353-359 [PMID: 24852104 DOI: 10.1016/j.gie.2014.03.039]

56

Kochhar G, Shen B. Endoscopic fistulotomy in inflammatory bowel disease (with video). Gastrointest Endosc 2018; 88: 87-94 [PMID: 29501418 DOI: 10.1016/j.gie.2018.02.034]

57

Chen M, Shen B. Endoscopic Therapy in Crohn's Disease: Principle, Preparation, and Technique. Inflamm Bowel Dis 2015; 21: 2222-2240 [PMID: 26284298 DOI: 10.1097/MIB.0000000000000433]

58

Verlaan T, Bartels SA, van Berge Henegouwen MI, Tanis PJ, Fockens P, Bemelman WA. Early, minimally invasive closure of anastomotic leaks: a new concept. Colorectal Dis 2011; 13 Suppl 7: 18-22 [PMID: 22098512 DOI: 10.1111/j.1463-1318.2011.02775.x]

59

Prudhomme M, Dozois RR, Godlewski G, Mathison S, Fabbro-Peray P. Anal canal strictures after ileal pouch-anal anastomosis. Dis Colon Rectum 2003; 46: 20-23 [PMID: 12544517 DOI: 10.1007/s10350-004-6491-7]

60

Agrawal M, Colombel JF. Treat-to-Target in Inflammatory Bowel Diseases, What Is the Target and How Do We Treat? Gastrointest Endosc Clin N Am 2019; 29: 421-436 [PMID: 31078245 DOI: 10.1016/j.giec.2019.02.004]

61

Gupta A, Yu A, Peyrin-Biroulet L, Ananthakrishnan AN. Treat to Target: The Role of Histologic Healing in Inflammatory Bowel Diseases: A Systematic Review and Meta-analysis. Clin Gastroenterol Hepatol 2020 [PMID: 33010406 DOI: 10.1016/j.cgh.2020.09.046]

62

Scoville EA, Schwartz DA. Endoscopy in inflammatory bowel disease: advances in disease management. Gastrointest Endosc 2017; 86: 952-961 [PMID: 28886962 DOI: 10.1016/j.gie.2017.08.034]

63


https://dx.doi.org/10.14309/ajg.0000000000000152








https://dx.doi.org/10.1111/1751-2980.12914




https://dx.doi.org/10.1016/j.clinre.2020.10.006


https://dx.doi.org/10.1093/gastro/goz051








https://dx.doi.org/10.4253/wjge.v5.i5.265






https://dx.doi.org/10.1097/MIB.0000000000000433




https://dx.doi.org/10.1007/s10350-004-6491-7


https://dx.doi.org/10.1016/j.giec.2019.02.004






World Journal of



MINIREVIEWS

Deep learning for diagnosis of precancerous lesions in upper gastrointestinal endoscopy: A review

Tao Yan, Pak Kin Wong, Ye-Ying Qin

ORCID number: Tao Yan 0000-0002-8929-015X; Pak Kin Wong 0000-0002-7623-6904; Ye-Ying Qin 0000-0002-7779-1045.

Author contributions: Wong PK and Yan T contributed to concept design and drafted the manuscript; Yan T and Qin YY collected the data; All the authors have approved the final version of the manuscript.

Supported by The Science and Technology Development Fund, Macau SAR, No. 0021/2019/A.

Conflict-of-interest statement: All authors declare no conflict of interest.


Manuscript source: Invited

Tao Yan, School of Mechanical Engineering, Hubei University of Arts and Science, Xiangyang 441053, Hubei Province, China

Tao Yan, Pak Kin Wong, Ye-Ying Qin, Department of Electromechanical Engineering, University of Macau, Taipa 999078, Macau, China

Corresponding author: Pak Kin Wong, PhD, Professor, Department of Electromechanical Engineering, University of Macau, Avenida da Universidade, Taipa 999078, Macau, China. [email protected]

AbstractUpper gastrointestinal (GI) cancers are the leading cause of cancer-related deaths worldwide. Early identification of precancerous lesions has been shown to minimize the incidence of GI cancers and substantiate the vital role of screening endoscopy. However, unlike GI cancers, precancerous lesions in the upper GI tract can be subtle and difficult to detect. Artificial intelligence techniques, especially deep learning algorithms with convolutional neural networks, might help endoscopists identify the precancerous lesions and reduce interobserver variability. In this review, a systematic literature search was undertaken of the Web of Science, PubMed, Cochrane Library and Embase, with an emphasis on the deep learning-based diagnosis of precancerous lesions in the upper GI tract. The status of deep learning algorithms in upper GI precancerous lesions has been systematically summarized. The challenges and recommendations targeting this field are comprehensively analyzed for future research.

Key Words: Artificial intelligence; Deep learning; Convolutional neural network; Precancerous lesions; Endoscopy


Core Tip: Artificial intelligence techniques, especially deep learning algorithms with convolutional neural networks, have revolutionized upper gastrointestinal endoscopy. In recent years, several deep learning-based artificial intelligence systems have emerged in the gastrointestinal community for endoscopic detection of precancerous lesions. The current review provides an analysis and status of the deep learning-based



http://orcid.org/0000-0002-8929-015X

http://orcid.org/0000-0002-8929-015X

http://orcid.org/0000-0002-8929-015X

http://orcid.org/0000-0002-7623-6904

http://orcid.org/0000-0002-7623-6904

http://orcid.org/0000-0002-7623-6904

http://orcid.org/0000-0002-7779-1045

http://orcid.org/0000-0002-7779-1045

http://orcid.org/0000-0002-7779-1045





Yan T et al. Deep learning for precancerous lesions


manuscript



Peer-review report’s scientific quality classificationGrade A (Excellent): 0 Grade B (Very good): B Grade C (Good): C Grade D (Fair): 0 Grade E (Poor): 0

Received: January 24, 2021 Peer-review started: January 24, 2021 First decision: March 14, 2021 Revised: March 27, 2021 Accepted: April 9, 2021 Article in press: April 9, 2021 Published online: May 28, 2021

P-Reviewer: Amornyotin S, Haruma K S-Editor: Fan JR L-Editor: Filipodia P-Editor: Liu JH

diagnosis of precancerous lesions in the upper gastrointestinal tract and identifies future challenges and recommendations.

Citation: Yan T, Wong PK, Qin YY. Deep learning for diagnosis of precancerous lesions in upper gastrointestinal endoscopy: A review. World J Gastroenterol 2021; 27(20): 2531-2544URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2531.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2531

INTRODUCTIONUpper gastrointestinal (GI) cancers, mainly including gastric cancer (8.2% of total cancer deaths) and esophageal cancer (5.3% of total cancer deaths), are the leading cause of cancer-related deaths worldwide[1]. Previous studies have shown that upper GI cancers always go through the stages of precancerous lesions, which can be defined as common conditions associated with a higher risk of developing cancers over time[2-4]. The detection of the precancerous lesions before cancer occurs could significantly reduce morbidity and mortality rates[5,6]. Currently, the main approach for the diagnosis of disorders or issues in the upper GI tract is endoscopy[7,8]. Compared with GI cancers, which usually show typical morphological characteristics, the precancerous lesions often appear in flat mucosa and exhibit few morphological changes. Manual screening through endoscopy is labor-intensive, time-consuming and relies heavily on clinical experience. Computer-assisted diagnosis based on artificial intelligence (AI) can overcome these dilemmas.

Over the past few decades, AI techniques such as machine learning (ML) and deep learning (DL) have been widely used in endoscopic imaging to improve the diagnostic accuracy and efficiency of various GI lesions[9-13]. The exact definition of AI, ML and DL can be misunderstood by physicians. AI, ML and DL are overlapping disciplines (Figure 1). AI is a hierarchy that encompasses ML and DL; it describes a computerized solution to address the issues of human cognitive defined by McCarthy in 1956[14]. ML is a subset of AI in which algorithms can execute complex tasks, but it needs handcrafted feature extraction. ML originated around the 1980s and focuses on patterns and inference[15]. DL is a subset of ML and became feasible in the 2010s; it is focused specifically on deep neural networks. A convolutional neural network (CNN) is the primary DL algorithm for image processing[16,17].

The diagnostic process of an AI model is similar to the human brain. We take our previous research as an example to illustrate the diagnostic process of ML, DL and human experts. When an input image with gastric intestinal metaplasia (GIM), a precancerous lesion of gastric cancer, is fed into the ML system, it usually needs a manual feature extraction step, while the handcrafted features are unable to discern slight variations in the endoscopic image (Figure 2)[15]. Unlike conventional ML algorithms, CNNs can automatically learn representative features from the endoscopic images[17]. When we apply a CNN model to detect GIM, it performs better than conventional ML models and is comparable to experienced endoscopists[18]. For a broad variety of image processing activities in endoscopy, CNNs also show excellent results, and some CNN-based algorithms have been used in clinical practice[19-21]. However, DL, especially CNN, has some limitations. First, DL requires a lot of data and easily leads to overfitting. Second, the diagnostic accuracy of DL relies on the training data, but the clinical data of different types of diseases are always imbalanced, which easily causes diagnosis bias. In addition, a DL model is complex and requires a huge calculation, so most researchers can only use the ready-made model.

Despite the above limitations, DL-based AI systems are revolutionizing GI endoscopy. While there are several surveys on DL for GI cancers[9-13], no specific review on the application of DL in the endoscopic diagnosis of precancerous lesions is available in the literature. Therefore, the performance of DL on gastroenterology is summarized in this review, with an emphasis on the automatic diagnosis of precan-cerous lesions in the upper GI tract. GI cancers are out of the scope of this review. Specifically, we review the status of intelligent diagnoses of esophageal and gastric precancerous lesions. The challenges and recommendations based on the findings of the review are comprehensively analyzed to advance the field.





Figure 1 Infographic with icons and timeline for artificial intelligence, machine learning and deep learning.

Figure 2 Illustration of the diagnostic process of physician, machine learning and deep learning. A: Physician diagnostic process; B: Machine learning; C: Deep learning. Conv: Convolutional layer; FC: Fully connected layer; GIM: gastric intestinal metaplasia.

DL IN ENDOSCOPIC DETECTION OF PRECANCEROUS LESIONS IN ESOPHAGEAL MUCOSAEsophageal cancer is the eighth most prevalent form of cancer and the sixth most lethal cancer globally[1]. There are two major subtypes of esophageal cancer: Esophageal squamous cell carcinoma (ESCC) and esophageal adenocarcinoma (EAC)[22]. Esophageal squamous dysplasia (ESD) is believed to be the precancerous



lesion of ESCC[23-25], and Barrett’s esophagus (BE) is the identifiable precancerous lesion associated with EAC[4,26]. Endoscopic surveillance is recommended by GI societies to enable early detection of the two precancerous lesions of esophageal cancer[25,27].

However, the current endoscopic diagnosis methods for patients with BE, such as random 4-quadrant biopsy, laser-induced endomicroscopy, image enhanced endoscopy, etc., have disadvantages concerning the learning curve, cost, interobserver variability and time-consuming problems[28]. The current standard for identifying ESD is Lugol’s chromoendoscopy, but it shows poor specificity[29]. Besides, iodine staining often presents a risk of allergic reactions. To overcome these challenges, DL-based AI systems have been established to help endoscopists identify ESD and BE.

DL in ESDLow-grade and high-grade intraepithelial neoplasms, collectively referred to as ESD, are deemed as precancerous lesions of ESCC. Early and accurate detection of ESD is essential but also full of challenges[23-25]. DL is reliably able to depict ESD in real-time upper endoscopy. Cai et al[30] designed a novel computer-assisted diagnosis system to localize and identify early ESCC, including low-grade and high-grade intraepithelial neoplasia, through real-time white light endoscopy (WLE). The system achieved a sensitivity, specificity and accuracy of 97.8%, 85.4% and 91.4%, respectively. They also demonstrated that when referring to the results of the system, the overall diagnostic capability of the endoscopist has been increased. This research paved the way for the real-time diagnosis of ESD and ESCC. Following this work, Guo et al[31] applied 6473 narrow band (NB) images to train a real-time automated computer-assisted diagnosis system to support non-experts in the detection of ESD and ESCC. The system serves as a “second observer” in an endoscopic examination and achieves a sensitivity of 98.04% and specificity of 95.30% on NB images. The per-frame sensitivity was 96.10% for magnifying narrow band imaging (M-NBI) videos and 60.80% for non-M-NBI videos. The per lesion sensitivity was 100% in M-NBI videos.

DL in BEBE is a disorder in which the lining of the esophagus is damaged by gastric acid. The critical purpose of endoscopic Barrett’s surveillance is early detection of BE-related dysplasia[4,26-28]. Recently, there have been many studies on the DL-based diagnosis of BE, and we review some representative studies. de Groof et al[32] performed one of the first pilot studies to assess the performance of a DL-based system during live endoscopic procedures of patients with or without BE dysplasia. The system demonstrated 90% accuracy, 91% sensitivity and 89% specificity in a per-level analysis. Following up this work, they improved this system using stepwise transfer learning and five independent endoscopy data sets. The enhanced system obtained higher accuracy than non-expert endoscopists and with comparable delineation performance[33]. Furthermore, their team also demonstrated the feasibility of a DL-based system for tissue characterization of NBI endoscopy in BE, and the system achieved a promising diagnostic accuracy[34].

Hashimoto et al[35] borrowed from the Inception-ResNet-v2 algorithm to develop a model for real-time classification of early esophageal neoplasia in BE, and they also applied YOLO-v2 to draw localization boxes around regions classified as dysplasia. For detection of neoplasia, the system achieved a sensitivity of 96.4%, specificity of 94.2% and accuracy of 95.4%. Hussein et al[36] built a CNN model to diagnose dysplastic BE mucosa with a sensitivity of 88.3% and specificity of 80.0%. The results preliminarily indicated that the diagnostic performance of the CNN model was close to that of experienced endoscopists. Ebigbo et al[37] exploited the use of a CNN-based system to classify and segment cancer in BE. The system achieved an accuracy of 89.9% in 14 patients with neoplastic BE.

DL has also achieved excellent results in distinguishing multiple types of esophageal lesions, including BE. Liu et al[38] explored the use of a CNN model to distinguish esophageal cancers from BE. The model was trained and evaluated on 1272 images captured by WLE. After pre-processing and data augmentation, the average sensitivity, specificity and accuracy of the CNN model were 94.23%, 94.67% and 85.83%, respectively. Wu et al[39] developed a CNN-based framework named ELNet for automatic esophageal lesion (i.e. EAC, BE and inflammation) classification and segmentation, the ELNet achieved a classification sensitivity of 90.34%, specificity of 97.18% and accuracy of 96.28%. The segmentation sensitivity, specificity and accuracy were 80.18%, 96.55% and 94.62%, respectively. A similar study was proposed by Ghatwary et al[40], who applied a CNN algorithm to detect BE, EAC and ESCC from



endoscopic videos and obtained a high sensitivity of 93.7% and a high F-measure of 93.2%.

The studies exploring the creation of DL algorithms for the diagnosis of precan-cerous lesions in esophageal mucosa are summarized in Table 1.

DL IN ENDOSCOPIC DETECTION OF PRECANCEROUS LESIONS IN GASTRIC MUCOSAGastric cancer is the fifth most prevalent form of cancer and the third most lethal cancer globally[1]. Even though the prevalence of gastric cancer has declined during the last few decades, gastric cancer remains a significant clinical problem, especially in developing countries. This is because most patients are diagnosed in late stages with poor prognosis and restricted therapeutic choices[41]. The pathogenesis of gastric cancer involves a series of events starting with Helicobacter pylori-induced (H. pylori-induced) chronic inflammation, progressing towards atrophic gastritis, GIM, dysplasia and eventually gastric cancer[42]. Patients with the precancerous lesions (e.g., H. pylori-induced chronic inflammation, atrophic gastritis, GIM and dysplasia) are at consid-erable risk of gastric cancer[3,6,43]. It has been argued that the detection of such precancerous lesions may significantly reduce the incidence of gastric cancer. However, endoscopic examination is difficult to identify these precancerous lesions, and the diagnostic result also has high interobserver variability due to their subtle morphological changes in the mucosa and lack of experienced endoscopists[44,45]. Currently, many researchers are trying to use DL-based methods to detect gastric precancerous lesions; here, we review these studies in detail.

DL in H. pylori infectionMost of the gastric precancerous lesions are correlated with long-term infections with H. pylori[46]. Shichijo et al[47] performed one of the pioneering studies to apply CNNs in the diagnosis of H. pylori infection. The CNNs were built on GoogLeNet and trained on 32208 WLE images. One of their CNN models has higher accuracy than endoscopists. The study showed the feasibility of using CNN to diagnose H. pylori from endoscopic images. After this study, Itoh et al[48] developed a CNN model to detect H. pylori infection in WLE images and showed a sensitivity of 86.7% and specificity of 86.7% in the test dataset. A similar model was developed by Zheng et al[49] to evaluate H. pylori infection status, and the per-patient sensitivity, specificity and accuracy of the model were 91.6%, 98.6% and 93.8%, respectively. Besides WLE, blue laser imaging-bright and linked color imaging systems were prospectively applied by Nakashima et al[50] to collect endoscopic images. With these images, they fine-tuned a pre-trained GoogLeNet to predict H. pylori infection status. As compared with linked color imaging, the model achieved the highest sensitivity (96.7%) and specificity (86.7%) when using blue laser imaging-bright. Nakashima et al[51] also did a single center prospective study to build a CNN model to identify the status of H. pylori in uninfected, currently infected and post-eradication patients. The area under the receiver operating characteristic curve for the uninfected, currently infected and post-eradication categories was 0.90, 0.82 and 0.77, respectively.

DL in atrophic gastritisAtrophic gastritis is a form of chronic inflammation of the gastric mucosa; accurate endoscopic diagnosis is difficult[52]. Guimarães et al[53] reported the application of CNN to detect atrophic gastritis; the system achieved an accuracy of 93% and performed better than expert endoscopists. Recently, another CNN-based system for detecting atrophic gastritis was reported by Zhang et al[54]. The CNN model was trained and tested on a dataset containing 3042 images with atrophic gastritis and 2428 without atrophic gastritis. The diagnostic accuracy, sensitivity and specificity of the model were 94.2%, 94.5% and 94.0%, respectively, which were better than those of the experts. More recently, Horiuchi et al[55] explored the diagnostic ability of the CNN model to distinguish early gastric cancer and gastritis through M-NBI; the 22-layer CNN was built on GoogleNet and pretrained using 2570 endoscopic images, and the sensitivity, specificity and accuracy on 258 images were 95.4%, 71.0% and 85.3%, respectively. Except for high sensitivity, the CNN model also showed an overall test speed of 0.02 s per image, which was faster than human experts.



Table 1 Summary of studies using deep learning for detection of esophageal precancerous lesions

Ref. Year Imaging Study design Study aim DL model Dataset Outcomes

Cai et al[30] 2019 WLE Retrospective Detection of precancerous lesions and early ESCC

-- 2615 images Sensitivity: 97.8%. Specificity: 85.4%. Accuracy: 91.4%

Guo et al[31] 2020 NBI, M-NBI

Retrospective Detection of precancerous lesions and early ESCC

SegNet 13144 images and 168865 video frames

Sensitivity: 96.10% for M-NBI videos, 60.80% for non-M-NBI videos, 98.04% for images. Specificity: 99.90% for non-M-NBI/M-NBI videos, 95.30% for images

de Groof et al[32]

2020 WLE Retrospective Detection of Barrett’s neoplasia

ResNet/U-Ne 1544 images Sensitivity: 91%. Specificity: 89%. Accuracy: 90%

de Groof et al[33]

2020 WLE Retrospective Detection of Barrett’s neoplasia

ResNet/U-Ne 494364 unlabeled images and 1704 labeled images

Sensitivity: 90%. Specificity: 88%. Accuracy: 89%

Struyvenberg et al[34]

2021 NBI Retrospective Detection of Barrett’s neoplasia

ResNet/U-Ne 2677 images Sensitivity: 88%. Specificity: 78%. Accuracy: 84%

Hashimoto et al[35]

2020 WLE, NBI

Retrospective Recognition of early neoplasia in BE

Inception-ResNet-v2, YOLO-v2

2290 images Sensitivity: 96.4%. Specificity: 94.2%. Accuracy: 95.4%

Hussein et al[36]

2020 WLE Retrospective Diagnosis of early neoplasia in BE

Resnet101 266930 video frames

Sensitivity: 88.26%. Specificity: 80.13%

Ebigbo et al[37]

2020 WLE Retrospective Diagnosis of early EAC in BE

DeepLab V.3+, Resnet101

191 images Sensitivity: 83.7%. Specificity: 100%. Accuracy: 89.9%

Liu et al[38] 2020 WLE Retrospective Detection of esophageal cancer from precancerous lesions

Inception-ResNet 1272 images Sensitivity: 94.23%. Specificity: 94.67%. Accuracy: 85.83%

Wu et al[39] 2021 WLE Retrospective Automatic classification and segmentation for esophageal lesions

ELNet 1051 images Classification sensitivity: 90.34%. Classification specificity: 97.18%. Classification accuracy: 96.28%. Segmentation sensitivity: 80.18%. Segmentation Specificity: 96.55%, Segmentation accuracy: 94.62%

Ghatwary et al[40]

2021 WLE Retrospective Detection of esophageal abnormalities from endoscopic videos

DenseConvLstm, Faster R-CNN

42425 video frames

Sensitivity: 93.7%. F-measure: 93.2%

BE: Barrett’s esophagus; DL: Deep learning; EAC: Esophageal adenocarcinoma; ESCC: Esophageal squamous cell carcinoma; M-NBI: Magnifying narrow band imaging; NBI: Narrow band imaging; WLE: White light endoscopy.

DL in GIMGIM is the replacement of gastric-type mucinous epithelial cells with intestinal-type cells, which is a precancerous lesion with a worldwide prevalence of 25%[56]. The morphological characteristics of GIM are subtle and difficult to observe, so the manual diagnosis of GIM is full of challenges. Wang et al[57] reported the first instance of an AI system for localizing and identifying GIM from WLE images. The system achieved a high classification accuracy and a satisfactory segmentation result. A recent study developed a CNN-based diagnosis system that can detect atrophic gastritis and GIM from WLE images[58], and the detection sensitivity and specificity for atrophic gastritis were 87.2% and 91.1%, respectively. For detection of GIM, the system also achieved a sensitivity of 90.3% and a specificity of 93.7%. Recently, our team also developed a novel DL-based diagnostic system for detection of GIM in endoscopic images[18]. The difference from the previous research is that our system is composed of three independent CNNs, which can identify GIM from either NBI or M-NBI. The per-patient sensitivity, specificity and accuracy of the system were 91.9%, 86.0% and 88.8%, respectively. The diagnostic performance showed no significant differences as compared with human experts. Our research showed that the integration of NBI and M-NBI into the DL system could achieve satisfactory diagnostic performance for GIM.



DL in gastric dysplasiaGastric dysplasia is the penultimate step of gastric carcinogenesis, and accurate diagnosis of this lesion remains controversial[59]. To accurately classify advanced low-grade dysplasia, high-grade dysplasia, early gastric cancer and gastric cancer, Cho et al[60] established three CNN models based on 5017 endoscopic images. They found that the Inception-Resnet-v2 model performed the best, while it showed lower five-class accuracy compared with the endoscopists (76.4% vs 87.6%). Inoue et al[61] constructed a detection system using the Single-Shot Multibox Detector, which can automatically detect duodenal adenomas and high-grade dysplasia from WLE or NBI. The system detected 94.7% adenomas and 100% high-grade dysplasia on a dataset containing 1080 endoscopic images within only 31 s. Although most of the AI-assisted system can achieve high accuracy on endoscopic diagnosis, no study has investigated the role of AI in the training of junior endoscopists. To evaluate the role of AI in the training of junior endoscopists in predicting histology of endoscopic gastric lesions, including dysplasia, Lui et al[62] designed and validated a CNN classifier based on 3000 NB images. The classifier achieved an overall accuracy of 91.0%, sensitivity of 97.1% and specificity of 85.9%, which was superior to all junior endoscopists. They also demonstrated that with the feedback from the CNN classifier, the learning curve of junior endoscopists was improved in predicting histology of gastric lesions.

The studies exploring the creation of DL algorithms for the diagnosis of precan-cerous lesions in gastric mucosa are summarized in Table 2.

CHALLENGES AND RECOMMENDATIONSAI has gained much attention in recent years. In the field of GI endoscopy, DL is also a promising innovation in the identification and characterization of lesions[9-13]. Many successful studies have focused on GI cancers. Accurate detection of precancerous lesions such as ESD, BE, H. pylori-induced chronic inflammation, atrophic gastritis, GIM and gastric dysplasia can greatly reduce the incidence of cancers and require less cancer treatment. DL-assisted detection of these precancerous lesions has increasingly emerged in the last 5 yrs. To perform a systematic review of the status of DL for diagnosis of precancerous lesions of the upper GI tract, we conducted a compre-hensive search for all original publications on this target between January 1, 2017 and December 30, 2020. A variety of published papers has verified the outstanding performance of DL-assisted systems, several challenges remain from the viewpoint of physicians and algorithm engineers. The challenges and our recommendations on future research directions are outlined below.

Prospective studies and clinical verificationThe current literature reveals that most studies were designed in a retrospective manner with a strong probability of bias. In these retrospective studies, researchers tended to collect high-quality endoscopic images that showed typical characteristics of the detected lesions from a single medical center, while they excluded common low-quality images. This kind of selection bias may jeopardize the precision and lead to lower generalization of the DL models. Thus, data collected from multicenter studies with uninformative frames are necessary to build robust DL models, and prospective studies are needed to properly verify the accuracy of AI in clinical practice.

Handling of overfittingOverfitting means an AI model performs well on the training set but has high error on unseen data. The deep CNN architectures usually contain several convolutional layers and fully connected layers, which produce millions of parameters that easily lead to strong overfitting[16,17]. Training these parameters needs large-scale well-annotated data. However, well-annotated data are costly and hard to obtain in the clinical community. Possible solutions for overcoming the lack of well-annotated data to avoid overfitting mainly include data augmentation[63], transfer learning[17,64], semi-supervised learning[65] and data synthesis using generative adversarial networks[66].

Data augmentation is a common method to train CNNs to reduce overfitting[63]. According to current literature, almost all studies use data augmentation. Data augmentation is performed by using several image transformations such as random image rotation, flipping, shifting, scaling and their combinations are shown in Figure 3. Transfer learning involves transfer knowledge learned from a large source domain to a target domain[17,64]. This technique is usually performed by initializing the CNN using the weights pretrained on ImageNet dataset. As there are many



Table 2 Summary of studies using deep learning for detection of gastric precancerous lesions

Ref. Year Imaging Study design Study aim DL model Dataset Outcomes

Shichijo et al[47]

2017 WLE Retrospective Diagnosis of H. pylori infection

GoogLeNet 43689 images

Sensitivity: 88.9%; Specificity: 87.4%; Accuracy: 87.7%

Itoh et al[48] 2018 WLE Retrospective Analysis of H. pylori infection


Sensitivity: 86.7%; Specificity: 86.7%

Zheng et al[49]

2019 WLE Retrospective Evaluation of H. pylori infection status

ResNet-50 15484 images


Nakashima et al[50]

2018 BLI-bright, LCI

Prospective Prediction of H. pylori infection status


Sensitivity: 96.7%; Specificity: 86.7%

Nakashima et al[51]

2020 WLE, LCI Prospective Diagnosis of H. pylori infection

-- 13127 images

For currently infected patients, the sensitivity and specificity are 62.5% and 92.5%, respectively

Guimarães et al[53]

2020 WLE Retrospective Diagnosis of atrophic gastritis

VGG16 270 images

Accuracy: 93%

Zhang et al[54]

2020 WLE Retrospective Diagnosis of atrophic gastritis

DenseNet121 5470 images


Horiuchi et al[55]

2020 M-NBI Retrospective Differentiation between early gastric cancer and gastritis



Wang et al[57]

2019 WLE Retrospective Localization and identification of GIM

DeepLab V.3+ 200 images

Accuracy: 89.51%

Zheng et al[58]

2020 WLE Retrospective Detection of atrophic gastritis and GIM

ResNet-50 3759 images

Sensitivity for atrophic gastritis: 87.2%; Specificity for atrophic gastritis: 91.1%; Sensitivity for GIM: 90.3%; Specificity for GIM: 93.7%

Yan et al[18] 2020 NBI, M-NBI

Retrospective Diagnosis of GIM EfficientNetB4 2357 images


Cho et al[60] 2019 WLE Prospective Classification of multiclass gastric neoplasms

Inception-Resnet-v2

5217 images

Accuracy: 84.6%

Inoue et al[61]

2020 WLE, NBI Retrospective Detection of duodenal adenomas and high-grade dysplasias

Single-Shot Multibox Detector

1511 images

For high-grade dysplasia, the sensitivity and specificity are all 100%

Lui et al[62] 2020 NBI Retrospective Classification of gastric lesions

ResNet 3000 images


BLI-bright: Blue laser imaging-bright; DL: Deep learning; GIM: Gastric intestinal metaplasia; H. pylori: Helicobacter pylori; LCI: Linked color imaging; M-NBI: Magnifying narrow band imaging; NBI: Narrow band imaging; WLE: White light endoscopy.

imaging modalities such as WLE, NBI and M-NBI and the images share the common features of the detected lesions, Struyvenberg et al[34] applied a stepwise transfer learning approach to tackle the shortage of NB images. Their CNN model was first trained on many WLE images, which are easy to acquire as compared with NB images. Then, the weights were further trained and optimized using NB images. In the upper endoscopy, although well-annotated data are limited, the unlabeled data are abundant and easily available in most situations. Semi-supervised learning, which utilizes limited labeled data and large-scale unlabeled data to train and improve the CNN model, is useful in the GI tract[65,67,68]. Generative adversarial networks are widely used in the field of medical image synthesis; the synthetic images can be used to provide auxiliary training to improve the performance of the CNN models[66]. de Souza et al[69] introduced generative adversarial networks to generate high-quality endoscopic images for BE, and these synthetic images were used to provide auxiliary training. The detection results suggested that with the help of these synthetic images the CNN model outperformed the ones over the original datasets.

Although the above techniques were used to reduce overfitting to a certain extent, the single use of one type did not guarantee resolution of the overfitting problem. Integration of these techniques is a promising strategy, and it was verified in our research[18,70].



Figure 3 Data augmentation for a typical magnifying narrow band image for training a convolutional neural network model. This is performed by using a variety of image transformations and their combinations. A: Original image; B: Flip horizontal and random rotation; C: Flip vertical and magnification; D: Random rotation and shift; E: Flip horizontal, minification and shift; F: Flip vertical, rotation and shift.

Improvement on interpretabilityLack of interpretability (i.e. the “black box” nature), which is the nature of DL technology, is another gap between studies and clinical applications in the field of precancerous lesion detection from endoscopic images. The black box nature means that the decision-making process by the DL model is not clearly demonstrated, which may reduce the willingness of doctors to use it. Although attention maps can help explain the dominant areas by highlighting them, they are constrained in that they do not thoroughly explain how the algorithm comes to its final decision[71]. The attention maps are displayed as heat maps overlaid upon the original images, where warmer colors mean higher contributions to the decision making, which usually correspond to lesions. However, the attention maps also have some defects such as inaccurate display of lesions as shown in Figure 4, where the attention maps only cover partial areas associated with BE and GIM. This is the inherent shortcoming of attention maps. Therefore, understanding the mechanism used by the DL model for prediction is a hot research topic. The network dissection[72], an empirical method to identify the semantics of individual hidden nodes in the DL model, may be a feasible solution to improve interpretability.

Network designIn this review, we analyzed the DL model used in the detection of precancerous lesions in the upper GI tract. The literature shows that almost all the DL-based AI systems are developed based on state-of-the-art CNN architectures. These CNNs can only handle a single task, such as GoogLeNet for disease classification[47,48,50,55], YOLO for lesion identification[36] and SegNet for lesion segmentation[31]. Few networks can handle multiple tasks simultaneously. Thus, networks must be designed for multitask learning, which is valuable in clinical applications. Networks designed for handling high-resolution images can help detect micropatterns, which is profitable for small precancerous lesions. Moreover, attempts should be guided to exploit the use of videos rather than images to minimize the processing time and keep DL algorithms working at almost real-time level. Therefore, as suggested by Mori et al[73] and Thakkar et al[74], the AI systems may be treated as an extra pair of eyes to prevent the absence of subtle lesions.



Figure 4 Informative features (partially related to lesions areas) acquired by the convolutional neural networks, where warmer colors mean higher contributions to decision making. A: Original endoscopic images; B: Corresponding attention. BE: Barrett’s esophagus; GIM: Gastric intestinal metaplasia.

CONCLUSIONUpper GI cancers are a major cause of cancer-related deaths worldwide. Early detection of precancerous lesions could significantly reduce cancer incidence. Upper GI endoscopy is a gold standard procedure for identifying precancerous lesions in the upper GI tract. DL-based endoscopic systems can provide an easier, faster and more reliable endoscopic method. We have conducted a thorough review of detection of precancerous lesions of the upper GI tract using DL approaches since 2017. This is the first review on the DL-based diagnosis of precancerous lesions of the upper GI tract. The status, challenges and recommendations summarized in this review can provide guidance for intelligent diagnosis of other GI tract diseases, which can help engineers develop perfect AI products to assist clinical decision making. Despite the success of DL algorithms in upper GI endoscopy, prospective studies and clinical validation are still needed. Creation of large public databases, adoption of comprehensive overfitting prevention strategies and application of more advanced interpretable methods and networks are also necessary to encourage clinical application of AI for medical diagnosis.

ACKNOWLEDGEMENTSThe authors would like to thank Dr. I Cheong Choi, Dr. Hon Ho Yu and Dr. Mo Fong Li from the Department of Gastroenterology, Kiang Wu Hospital, Macau for their advice on this manuscript.

REFERENCESBray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018; 68: 394-424 [PMID: 30207593 DOI: 10.3322/caac.21492]

1

Morson BC. Precancerous lesions of upper gastrointestinal tract. JAMA 1962; 179: 311-315 [PMID: 14476362 DOI: 10.1001/jama.1962.03050050001001]

2

Rugge M, Capelle LG, Cappellesso R, Nitti D, Kuipers EJ. Precancerous lesions in the stomach: from biology to clinical patient management. Best Pract Res Clin Gastroenterol 2013; 27: 205-223 [PMID: 23809241 DOI: 10.1016/j.bpg.2012.12.007]

3

Coron E, Robaszkiewicz M, Chatelain D, Svrcek M, Fléjou JF. Advanced precancerous lesions in the lower oesophageal mucosa: high-grade dysplasia and intramucosal carcinoma in Barrett's oesophagus. Best Pract Res Clin Gastroenterol 2013; 27: 187-204 [PMID: 23809240 DOI:

4






https://dx.doi.org/10.1016/j.bpg.2012.12.007




10.1016/j.bpg.2013.03.011]Fassan M, Croce CM, Rugge M. miRNAs in precancerous lesions of the gastrointestinal tract. World J Gastroenterol 2011; 17: 5231-5239 [PMID: 22219591 DOI: 10.3748/wjg.v17.i48.5231]

5

Pimentel-Nunes P, Libânio D, Marcos-Pinto R, Areia M, Leja M, Esposito G, Garrido M, Kikuste I, Megraud F, Matysiak-Budnik T, Annibale B, Dumonceau JM, Barros R, Fléjou JF, Carneiro F, van Hooft JE, Kuipers EJ, Dinis-Ribeiro M. Management of epithelial precancerous conditions and lesions in the stomach (MAPS II): European Society of Gastrointestinal Endoscopy (ESGE), European Helicobacter and Microbiota Study Group (EHMSG), European Society of Pathology (ESP), and Sociedade Portuguesa de Endoscopia Digestiva (SPED) guideline update 2019. Endoscopy 2019; 51: 365-388 [PMID: 30841008 DOI: 10.1055/a-0859-1883]

6

Berci G, Forde KA. History of endoscopy: what lessons have we learned from the past? Surg Endosc 2000; 14: 5-15 [PMID: 10653228 DOI: 10.1007/s004649900002]

7

Kavic SM, Basson MD. Complications of endoscopy. Am J Surg 2001; 181: 319-332 [PMID: 11438266 DOI: 10.1016/s0002-9610(01)00589-x]

8

Niu PH, Zhao LL, Wu HL, Zhao DB, Chen YT. Artificial intelligence in gastric cancer: Application and future perspectives. World J Gastroenterol 2020; 26: 5408-5419 [PMID: 33024393 DOI: 10.3748/wjg.v26.i36.5408]

9

Parasher G, Wong M, Rawat M. Evolving role of artificial intelligence in gastrointestinal endoscopy. World J Gastroenterol 2020; 26: 7287-7298 [PMID: 33362384 DOI: 10.3748/wjg.v26.i46.7287]

10

Hussein M, González-Bueno Puyal J, Mountney P, Lovat LB, Haidry R. Role of artificial intelligence in the diagnosis of oesophageal neoplasia: 2020 an endoscopic odyssey. World J Gastroenterol 2020; 26: 5784-5796 [PMID: 33132634 DOI: 10.3748/wjg.v26.i38.5784]

11

Huang LM, Yang WJ, Huang ZY, Tang CW, Li J. Artificial intelligence technique in detection of early esophageal cancer. World J Gastroenterol 2020; 26: 5959-5969 [PMID: 33132647 DOI: 10.3748/wjg.v26.i39.5959]

12

Yang YJ, Bang CS. Application of artificial intelligence in gastroenterology. World J Gastroenterol 2019; 25: 1666-1683 [PMID: 31011253 DOI: 10.3748/wjg.v25.i14.1666]

13

McCarthy J, Minsky ML, Rochester N, Shannon CE. A proposal for the dartmouth summer research project on artificial intelligence, August 31, 1955. AI Mag 2006; 27: 12-14 [DOI: 10.1609/aimag.v27i4.1904]

14

Jordan MI, Mitchell TM. Machine learning: Trends, perspectives, and prospects. Science 2015; 349: 255-260 [PMID: 26185243 DOI: 10.1126/science.aaa8415]

15

LeCun Y, Bengio Y, Hinton G. Deep learning. Nature 2015; 521: 436-444 [PMID: 26017442 DOI: 10.1038/nature14539]

16

Shin HC, Roth HR, Gao M, Lu L, Xu Z, Nogues I, Yao J, Mollura D, Summers RM. Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning. IEEE Trans Med Imaging 2016; 35: 1285-1298 [PMID: 26886976 DOI: 10.1109/TMI.2016.2528162]

17

Yan T, Wong PK, Choi IC, Vong CM, Yu HH. Intelligent diagnosis of gastric intestinal metaplasia based on convolutional neural network and limited number of endoscopic images. Comput Biol Med 2020; 126: 104026 [PMID: 33059237 DOI: 10.1016/j.compbiomed.2020.104026]

18

Wang P, Berzin TM, Glissen Brown JR, Bharadwaj S, Becq A, Xiao X, Liu P, Li L, Song Y, Zhang D, Li Y, Xu G, Tu M, Liu X. Real-time automatic detection system increases colonoscopic polyp and adenoma detection rates: a prospective randomised controlled study. Gut 2019; 68: 1813-1819 [PMID: 30814121 DOI: 10.1136/gutjnl-2018-317500]

19

Gong D, Wu L, Zhang J, Mu G, Shen L, Liu J, Wang Z, Zhou W, An P, Huang X, Jiang X, Li Y, Wan X, Hu S, Chen Y, Hu X, Xu Y, Zhu X, Li S, Yao L, He X, Chen D, Huang L, Wei X, Wang X, Yu H. Detection of colorectal adenomas with a real-time computer-aided system (ENDOANGEL): a randomised controlled study. Lancet Gastroenterol Hepatol 2020; 5: 352-361 [PMID: 31981518 DOI: 10.1016/S2468-1253(19)30413-3]

20

An P, Yang D, Wang J, Wu L, Zhou J, Zeng Z, Huang X, Xiao Y, Hu S, Chen Y, Yao F, Guo M, Wu Q, Yang Y, Yu H. A deep learning method for delineating early gastric cancer resection margin under chromoendoscopy and white light endoscopy. Gastric Cancer 2020; 23: 884-892 [PMID: 32356118 DOI: 10.1007/s10120-020-01071-7]

21

Rustgi AK, El-Serag HB. Esophageal carcinoma. N Engl J Med 2014; 371: 2499-2509 [PMID: 25539106 DOI: 10.1056/NEJMra1314530]

22

Taylor PR, Abnet CC, Dawsey SM. Squamous dysplasia--the precursor lesion for esophageal squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev 2013; 22: 540-552 [PMID: 23549398 DOI: 10.1158/1055-9965.EPI-12-1347]

23

Raghu Subramanian C, Triadafilopoulos G. Diagnosis and therapy of esophageal squamous cell dysplasia and early esophageal squamous cell cancer. Gastroenterol Rep 2017; 5: 247-257 [DOI: 10.1093/gastro/gox022]

24

Auld M, Srinath H, Jeyarajan E. Oesophageal Squamous Dysplasia. J Gastrointest Cancer 2018; 49: 385-388 [PMID: 29806061 DOI: 10.1007/s12029-018-0122-3]

25

Hvid-Jensen F, Pedersen L, Drewes AM, Sørensen HT, Funch-Jensen P. Incidence of adenocarcinoma among patients with Barrett's esophagus. N Engl J Med 2011; 365: 1375-1383 [PMID: 21995385 DOI: 10.1056/NEJMoa1103042]

26

Eluri S, Shaheen NJ. Barrett's esophagus: diagnosis and management. Gastrointest Endosc 2017; 85: 889-903 [PMID: 28109913 DOI: 10.1016/j.gie.2017.01.007]

27

https://dx.doi.org/10.1016/j.bpg.2013.03.011




https://dx.doi.org/10.1055/a-0859-1883


https://dx.doi.org/10.1007/s004649900002













https://dx.doi.org/10.1609/aimag.v27i4.1904


https://dx.doi.org/10.1126/science.aaa8415




https://dx.doi.org/10.1109/TMI.2016.2528162


https://dx.doi.org/10.1016/j.compbiomed.2020.104026




https://dx.doi.org/10.1016/S2468-1253(19)30413-3


https://dx.doi.org/10.1007/s10120-020-01071-7




https://dx.doi.org/10.1158/1055-9965.EPI-12-1347

https://dx.doi.org/10.1093/gastro/gox022


https://dx.doi.org/10.1007/s12029-018-0122-3







Tavakkoli A, Appelman HD, Beer DG, Madiyal C, Khodadost M, Nofz K, Metko V, Elta G, Wang T, Rubenstein JH. Use of Appropriate Surveillance for Patients With Nondysplastic Barrett's Esophagus. Clin Gastroenterol Hepatol 2018; 16: 862-869. e3 [PMID: 29432922 DOI: 10.1016/j.cgh.2018.01.052]

28

Dawsey SM, Fleischer DE, Wang GQ, Zhou B, Kidwell JA, Lu N, Lewin KJ, Roth MJ, Tio TL, Taylor PR. Mucosal iodine staining improves endoscopic visualization of squamous dysplasia and squamous cell carcinoma of the esophagus in Linxian, China. Cancer 1998; 83: 220-231 [PMID: 9669803]

29

Cai SL, Li B, Tan WM, Niu XJ, Yu HH, Yao LQ, Zhou PH, Yan B, Zhong YS. Using a deep learning system in endoscopy for screening of early esophageal squamous cell carcinoma (with video). Gastrointest Endosc 2019; 90: 745-753. e2 [PMID: 31302091 DOI: 10.1016/j.gie.2019.06.044]

30

Guo L, Xiao X, Wu C, Zeng X, Zhang Y, Du J, Bai S, Xie J, Zhang Z, Li Y, Wang X, Cheung O, Sharma M, Liu J, Hu B. Real-time automated diagnosis of precancerous lesions and early esophageal squamous cell carcinoma using a deep learning model (with videos). Gastrointest Endosc 2020; 91: 41-51 [PMID: 31445040 DOI: 10.1016/j.gie.2019.08.018]

31

de Groof AJ, Struyvenberg MR, Fockens KN, van der Putten J, van der Sommen F, Boers TG, Zinger S, Bisschops R, de With PH, Pouw RE, Curvers WL, Schoon EJ, Bergman JJGHM. Deep learning algorithm detection of Barrett's neoplasia with high accuracy during live endoscopic procedures: a pilot study (with video). Gastrointest Endosc 2020; 91: 1242-1250 [PMID: 31926965 DOI: 10.1016/j.gie.2019.12.048]

32

de Groof AJ, Struyvenberg MR, van der Putten J, van der Sommen F, Fockens KN, Curvers WL, Zinger S, Pouw RE, Coron E, Baldaque-Silva F, Pech O, Weusten B, Meining A, Neuhaus H, Bisschops R, Dent J, Schoon EJ, de With PH, Bergman JJ. Deep-Learning System Detects Neoplasia in Patients With Barrett's Esophagus With Higher Accuracy Than Endoscopists in a Multistep Training and Validation Study With Benchmarking. Gastroenterology 2020; 158: 915-929. e4 [PMID: 31759929 DOI: 10.1053/j.gastro.2019.11.030]

33

Struyvenberg MR, de Groof AJ, van der Putten J, van der Sommen F, Baldaque-Silva F, Omae M, Pouw R, Bisschops R, Vieth M, Schoon EJ, Curvers WL, de With PH, Bergman JJ. A computer-assisted algorithm for narrow-band imaging-based tissue characterization in Barrett's esophagus. Gastrointest Endosc 2021; 93: 89-98 [PMID: 32504696 DOI: 10.1016/j.gie.2020.05.050]

34

Hashimoto R, Requa J, Dao T, Ninh A, Tran E, Mai D, Lugo M, El-Hage Chehade N, Chang KJ, Karnes WE, Samarasena JB. Artificial intelligence using convolutional neural networks for real-time detection of early esophageal neoplasia in Barrett's esophagus (with video). Gastrointest Endosc 2020; 91: 1264-1271. e1 [PMID: 31930967 DOI: 10.1016/j.gie.2019.12.049]

35

Hussein M, Juana Gonzales-Bueno P, Brandao P, Toth D, Seghal V, Everson MA, Lipman G, Ahmad OF, Kader R, Esteban JM, Bisschops R, Banks M, Mountney P, Stoyanov D, Lovat L, Haidry R. Deep Neural Network for the detection of early neoplasia in Barrett’s oesophagus. Gastrointest Endosc 2020; 91: AB250 [DOI: 10.1016/j.gie.2020.03.1826]

36

Ebigbo A, Mendel R, Probst A, Manzeneder J, Prinz F, de Souza LA Jr, Papa J, Palm C, Messmann H. Real-time use of artificial intelligence in the evaluation of cancer in Barrett's oesophagus. Gut 2020; 69: 615-616 [PMID: 31541004 DOI: 10.1136/gutjnl-2019-319460]

37

Liu G, Hua J, Wu Z, Meng T, Sun M, Huang P, He X, Sun W, Li X, Chen Y. Automatic classification of esophageal lesions in endoscopic images using a convolutional neural network. Ann Transl Med 2020; 8: 486 [PMID: 32395530 DOI: 10.21037/atm.2020.03.24]

38

Wu Z, Ge R, Wen M, Liu G, Chen Y, Zhang P, He X, Hua J, Luo L, Li S. ELNet:Automatic classification and segmentation for esophageal lesions using convolutional neural network. Med Image Anal 2021; 67: 101838 [PMID: 33129148 DOI: 10.1016/j.media.2020.101838]

39

Ghatwary N, Zolgharni M, Janan F, Ye X. Learning Spatiotemporal Features for Esophageal Abnormality Detection From Endoscopic Videos. IEEE J Biomed Health Inform 2021; 25: 131-142 [PMID: 32750901 DOI: 10.1109/JBHI.2020.2995193]

40

Van Cutsem E, Sagaert X, Topal B, Haustermans K, Prenen H. Gastric cancer. Lancet 2016; 388: 2654-2664 [PMID: 27156933 DOI: 10.1016/S0140-6736(16)30354-3]

41

Correa P, Piazuelo MB. The gastric precancerous cascade. J Dig Dis 2012; 13: 2-9 [PMID: 22188910 DOI: 10.1111/j.1751-2980.2011.00550.x]

42

de Vries AC, van Grieken NC, Looman CW, Casparie MK, de Vries E, Meijer GA, Kuipers EJ. Gastric cancer risk in patients with premalignant gastric lesions: a nationwide cohort study in the Netherlands. Gastroenterology 2008; 134: 945-952 [PMID: 18395075 DOI: 10.1053/j.gastro.2008.01.071]

43

Chung CS, Wang HP. Screening for precancerous lesions of upper gastrointestinal tract: from the endoscopists' viewpoint. Gastroenterol Res Pract 2013; 2013: 681439 [PMID: 23573079 DOI: 10.1155/2013/681439]

44

Menon S, Trudgill N. How commonly is upper gastrointestinal cancer missed at endoscopy? Endosc Int Open 2014; 2: E46-E50 [PMID: 26135259 DOI: 10.1055/s-0034-1365524]

45

Matysiak-Budnik T, Mégraud F. Helicobacter pylori infection and gastric cancer. Eur J Cancer 2006; 42: 708-716 [PMID: 16556496 DOI: 10.1016/j.ejca.2006.01.020]

46

Shichijo S, Nomura S, Aoyama K, Nishikawa Y, Miura M, Shinagawa T, Takiyama H, Tanimoto T, Ishihara S, Matsuo K, Tada T. Application of Convolutional Neural Networks in the Diagnosis of Helicobacter pylori Infection Based on Endoscopic Images. EBioMedicine 2017; 25: 106-111 [PMID:

47




















https://dx.doi.org/10.21037/atm.2020.03.24


https://dx.doi.org/10.1016/j.media.2020.101838


https://dx.doi.org/10.1109/JBHI.2020.2995193


https://dx.doi.org/10.1016/S0140-6736(16)30354-3






https://dx.doi.org/10.1155/2013/681439


https://dx.doi.org/10.1055/s-0034-1365524


https://dx.doi.org/10.1016/j.ejca.2006.01.020



29056541 DOI: 10.1016/j.ebiom.2017.10.014]Itoh T, Kawahira H, Nakashima H, Yata N. Deep learning analyzes Helicobacter pylori infection by upper gastrointestinal endoscopy images. Endosc Int Open 2018; 6: E139-E144 [PMID: 29399610 DOI: 10.1055/s-0043-120830]

48

Zheng W, Zhang X, Kim JJ, Zhu X, Ye G, Ye B, Wang J, Luo S, Li J, Yu T, Liu J, Hu W, Si J. High Accuracy of Convolutional Neural Network for Evaluation of Helicobacter pylori Infection Based on Endoscopic Images: Preliminary Experience. Clin Transl Gastroenterol 2019; 10: e00109 [PMID: 31833862 DOI: 10.14309/ctg.0000000000000109]

49

Nakashima H, Kawahira H, Kawachi H, Sakaki N. Artificial intelligence diagnosis of Helicobacter pylori infection using blue laser imaging-bright and linked color imaging: a single-center prospective study. Ann Gastroenterol 2018; 31: 462-468 [PMID: 29991891 DOI: 10.20524/aog.2018.0269]

50

Nakashima H, Kawahira H, Kawachi H, Sakaki N. Endoscopic three-categorical diagnosis of Helicobacter pylori infection using linked color imaging and deep learning: a single-center prospective study (with video). Gastric Cancer 2020; 23: 1033-1040 [PMID: 32382973 DOI: 10.1007/s10120-020-01077-1]

51

Okamura T, Iwaya Y, Kitahara K, Suga T, Tanaka E. Accuracy of Endoscopic Diagnosis for Mild Atrophic Gastritis Infected with Helicobacter pylori. Clin Endosc 2018; 51: 362-367 [PMID: 29695148 DOI: 10.5946/ce.2017.177]

52

Guimarães P, Keller A, Fehlmann T, Lammert F, Casper M. Deep-learning based detection of gastric precancerous conditions. Gut 2020; 69: 4-6 [PMID: 31375599 DOI: 10.1136/gutjnl-2019-319347]

53

Zhang Y, Li F, Yuan F, Zhang K, Huo L, Dong Z, Lang Y, Zhang Y, Wang M, Gao Z, Qin Z, Shen L. Diagnosing chronic atrophic gastritis by gastroscopy using artificial intelligence. Dig Liver Dis 2020; 52: 566-572 [PMID: 32061504 DOI: 10.1016/j.dld.2019.12.146]

54

Horiuchi Y, Aoyama K, Tokai Y, Hirasawa T, Yoshimizu S, Ishiyama A, Yoshio T, Tsuchida T, Fujisaki J, Tada T. Convolutional Neural Network for Differentiating Gastric Cancer from Gastritis Using Magnified Endoscopy with Narrow Band Imaging. Dig Dis Sci 2020; 65: 1355-1363 [PMID: 31584138 DOI: 10.1007/s10620-019-05862-6]

55

Lam SK, Lau G. Novel treatment for gastric intestinal metaplasia, a precursor to cancer. JGH Open 2020; 4: 569-573 [PMID: 32782940 DOI: 10.1002/jgh3.12318]

56

Wang C, Li Y, Yao J, Chen B, Song J, Yang X. Localizing and Identifying Intestinal Metaplasia Based on Deep Learning in Oesophagoscope. 2019 8th International Symposium on Next Generation 2019; 1-4 [DOI: 10.1109/ISNE.2019.8896546]

57

Zheng W, Yu T, Lin N, Ye G, Zhu X, Shen Y, Zhang X, Liu J, Hu W, Cao Q, Si J. Tu1075 deep convolutional neural networks for recognition of atrophic gastritis and intestinal metaplasia based on endoscopy images. Gastrointest Endosc 2020; 91: AB533-AB534 [DOI: 10.1016/j.gie.2020.03.3247]

58

Sung JK. Diagnosis and management of gastric dysplasia. Korean J Intern Med 2016; 31: 201-209 [PMID: 26932397 DOI: 10.3904/kjim.2016.021]

59

Cho BJ, Bang CS, Park SW, Yang YJ, Seo SI, Lim H, Shin WG, Hong JT, Yoo YT, Hong SH, Choi JH, Lee JJ, Baik GH. Automated classification of gastric neoplasms in endoscopic images using a convolutional neural network. Endoscopy 2019; 51: 1121-1129 [PMID: 31443108 DOI: 10.1055/a-0981-6133]

60

Inoue S, Shichijo S, Aoyama K, Kono M, Fukuda H, Shimamoto Y, Nakagawa K, Ohmori M, Iwagami H, Matsuno K, Iwatsubo T, Nakahira H, Matsuura N, Maekawa A, Kanesaka T, Yamamoto S, Takeuchi Y, Higashino K, Uedo N, Ishihara R, Tada T. Application of Convolutional Neural Networks for Detection of Superficial Nonampullary Duodenal Epithelial Tumors in Esophagogastroduodenoscopic Images. Clin Transl Gastroenterol 2020; 11: e00154 [PMID: 32352719 DOI: 10.14309/ctg.0000000000000154]

61

Lui TKL, Wong KKY, Mak LLY, To EWP, Tsui VWM, Deng Z, Guo J, Ni L, Cheung MKS, Leung WK. Feedback from artificial intelligence improved the learning of junior endoscopists on histology prediction of gastric lesions. Endosc Int Open 2020; 8: E139-E146 [PMID: 32010746 DOI: 10.1055/a-1036-6114]

62

Shorten C, Khoshgoftaar TM. A survey on image data augmentation for deep learning. J Big Data 2019; 6: 60 [DOI: 10.1186/s40537-019-0197-0]

63

Pan SJ, Yang Q. A survey on transfer learning. IEEE 2009; 22: 1345-1359 [DOI: 10.1109/TKDE.2009.191]

64

Zhu X, Goldberg AB. Introduction to Semi-Supervised Learning. Synthesis Lectures on Artificial Intelligence and Machine Learning 2009; 3: 1-130 [DOI: 10.2200/S00196ED1V01Y200906AIM006]

65

Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio, Y. Generative adversarial nets. Advances in neural information processing systems 2014; 27: 2672-2680.

66

Guo X, Yuan Y. Semi-supervised WCE image classification with adaptive aggregated attention. Med Image Anal 2020; 64: 101733 [PMID: 32574987 DOI: 10.1016/j.media.2020.101733]

67

Borgli H, Thambawita V, Smedsrud PH, Hicks S, Jha D, Eskeland SL, Randel KR, Pogorelov K, Lux M, Nguyen DTD, Johansen D, Griwodz C, Stensland HK, Garcia-Ceja E, Schmidt PT, Hammer HL, Riegler MA, Halvorsen P, de Lange T. HyperKvasir, a comprehensive multi-class image and video dataset for gastrointestinal endoscopy. Sci Data 2020; 7: 283 [PMID: 32859981 DOI: 10.1038/s41597-020-00622-y]

68

de Souza LA Jr, Passos LA, Mendel R, Ebigbo A, Probst A, Messmann H, Palm C, Papa JP. Assisting Barrett's esophagus identification using endoscopic data augmentation based on Generative Adversarial Networks. Comput Biol Med 2020; 126: 104029 [PMID: 33059236 DOI:

69


https://dx.doi.org/10.1016/j.ebiom.2017.10.014


https://dx.doi.org/10.1055/s-0043-120830


https://dx.doi.org/10.14309/ctg.0000000000000109


https://dx.doi.org/10.20524/aog.2018.0269


https://dx.doi.org/10.1007/s10120-020-01077-1


https://dx.doi.org/10.5946/ce.2017.177






https://dx.doi.org/10.1007/s10620-019-05862-6


https://dx.doi.org/10.1002/jgh3.12318

https://dx.doi.org/10.1109/ISNE.2019.8896546



https://dx.doi.org/10.3904/kjim.2016.021


https://dx.doi.org/10.1055/a-0981-6133


https://dx.doi.org/10.14309/ctg.0000000000000154


https://dx.doi.org/10.1055/a-1036-6114

https://dx.doi.org/10.1186/s40537-019-0197-0

https://dx.doi.org/10.1109/TKDE.2009.191

https://dx.doi.org/10.2200/S00196ED1V01Y200906AIM006








10.1016/j.compbiomed.2020.104029]Yan T, Wong PK, Ren H, Wang H, Wang J, Li Y. Automatic distinction between COVID-19 and common pneumonia using multi-scale convolutional neural network on chest CT scans. Chaos Solitons Fractals 2020; 140: 110153 [PMID: 32834641 DOI: 10.1016/j.chaos.2020.110153]

70

Selvaraju, RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-cam: Visual explanations from deep networks via gradient-based localization. In: Proceedings of the IEEE International Conference on Computer Vision 2017; 618-626 [DOI: 10.1109/ICCV.2017.74]

71

Bau D, Zhu JY, Strobelt H, Lapedriza A, Zhou B, Torralba A. Understanding the role of individual units in a deep neural network. Proc Natl Acad Sci USA 2020; 117: 30071-30078 [PMID: 32873639 DOI: 10.1073/pnas.1907375117]

72

Mori Y, Kudo SE, Mohmed HEN, Misawa M, Ogata N, Itoh H, Oda M, Mori K. Artificial intelligence and upper gastrointestinal endoscopy: Current status and future perspective. Dig Endosc 2019; 31: 378-388 [PMID: 30549317 DOI: 10.1111/den.13317]

73

Thakkar SJ, Kochhar GS. Artificial intelligence for real-time detection of early esophageal cancer: another set of eyes to better visualize. Gastrointest Endosc 2020; 91: 52-54 [PMID: 31865996 DOI: 10.1016/j.gie.2019.09.036]

74



https://dx.doi.org/10.1016/j.chaos.2020.110153

https://dx.doi.org/10.1109/ICCV.2017.74




https://dx.doi.org/10.1111/den.13317




World Journal of



MINIREVIEWS

State of machine and deep learning in histopathological applications in digestive diseases

Soma Kobayashi, Joel H Saltz, Vincent W Yang

ORCID number: Soma Kobayashi 0000-0002-0470-4027; Joel H Saltz 0000-0002-3451-2165; Vincent W Yang 0000-0002-6981-3558.

Author contributions: Kobayashi S organized and drafted the manuscript; Saltz JH and Yang VW reviewed and performed critical revisions of the manuscript.

Supported by National Institutes of Health, No. GM008444 (to Kobayashi S), No. CA225021 (to Saltz JH), and No. DK052230 (to Yang VW).

Conflict-of-interest statement: The authors have no conflicts of interest to report.


Manuscript source: Invited

Soma Kobayashi, Joel H Saltz, Department of Biomedical Informatics, Renaissance School of Medicine, Stony Brook University, Stony Brook, NY 11794, United States

Vincent W Yang, Department of Medicine, Renaissance School of Medicine, Stony Brook University, Stony Brook, NY 11794, United States

Vincent W Yang, Department of Physiology and Biophysics, Renaissance School of Medicine, Stony Brook University, Stony Brook , NY 11794, United States

Corresponding author: Vincent W Yang, MD, PhD, Chairman, Full Professor, Department of Medicine, Renaissance School of Medicine, Stony Brook University, HSC T-16, Rm 040, 101 Nicolls Road, Stony Brook, NY 11794, United States. [email protected]

AbstractMachine learning (ML)- and deep learning (DL)-based imaging modalities have exhibited the capacity to handle extremely high dimensional data for a number of computer vision tasks. While these approaches have been applied to numerous data types, this capacity can be especially leveraged by application on histopatho-logical images, which capture cellular and structural features with their high-resolution, microscopic perspectives. Already, these methodologies have demonstrated promising performance in a variety of applications like disease classification, cancer grading, structure and cellular localizations, and prognostic predictions. A wide range of pathologies requiring histopathological evaluation exist in gastroenterology and hepatology, indicating these as disciplines highly targetable for integration of these technologies. Gastroenterologists have also already been primed to consider the impact of these algorithms, as development of real-time endoscopic video analysis software has been an active and popular field of research. This heightened clinical awareness will likely be important for future integration of these methods and to drive interdisciplinary collaborations on emerging studies. To provide an overview on the application of these method-ologies for gastrointestinal and hepatological histopathological slides, this review will discuss general ML and DL concepts, introduce recent and emerging literature using these methods, and cover challenges moving forward to further advance the field.

Key Words: Artificial intelligence; Machine learning; Deep learning; Gastroenterology; Hepatology; Histopathology



http://orcid.org/0000-0002-0470-4027

http://orcid.org/0000-0002-0470-4027

http://orcid.org/0000-0002-3451-2165

http://orcid.org/0000-0002-3451-2165

http://orcid.org/0000-0002-6981-3558

http://orcid.org/0000-0002-6981-3558





Kobayashi S et al. ML/DL in digestive diseases


manuscript


Country/Territory of origin: United States

Peer-review report’s scientific quality classificationGrade A (Excellent): 0 Grade B (Very good): 0 Grade C (Good): C, C Grade D (Fair): 0 Grade E (Poor): 0


P-Reviewer: Cabezuelo AS S-Editor: Gao CC L-Editor: A P-Editor: Wang LL


Core Tip: Machine learning- and deep learning-based imaging approaches have been increasingly applied to histopathological slides and hold much potential in areas spanning diagnosis, disease grading and characterizations, academic research, and clinical decision support mechanisms. As these studies have entered into translational applications, tracking the current state of these methodologies and the clinical areas in which impact is most likely is of high importance. This review will thus provide a background of major concepts and terminologies while highlighting emerging literature regarding histopathological applications of these techniques and challenges and opportunities moving forward.

Citation: Kobayashi S, Saltz JH, Yang VW. State of machine and deep learning in histopathological applications in digestive diseases. World J Gastroenterol 2021; 27(20): 2545-2575URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2545.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2545

INTRODUCTIONThe past decade has seen the growing popularity of machine learning (ML) and deep learning (DL) applications across numerous domains, and the medical field has been no exception. A search for DL publications in the domains of Medical Informatics, Sensing, Bioinformatics, Imaging, and Public Health shows a 5-fold to 6-fold increase in publication counts from 2010-2015[1], and this trend continues today. Applications of DL in healthcare have been particularly wide ranged, covering proteomics, genomics and expression data, electronic health records for patient characterizations, as well as image analysis for histopathology, magnetic resonance images (MRI) scans, positron emission topography scans, computerized topography (CT) scans, and endoscopy videos. DL image analysis methodologies have the potential to automate and speed up pathologists’ tasks with high accuracy and precision. Recent applications have also illustrated the capacity for DL methodologies to extract information from histopathological images unseen to the human eye, such as expression data. Importantly, these ML and DL image analysis applications have the benefit of requiring no additional sample collection from patients, as inputs are typically biomedical images already collected within the clinical workflow.

The majority of ML and DL focus in the gastroenterology and hepatology communities has been in endoscopy, and this is highlighted by the recent Breakthrough Device Designation granted by the United States Food and Drug Administration (FDA) for a DL-based endoscopic, real-time diagnostic software for gastric cancer[2]. However, the application of ML and DL methodologies on histopath-ological images is a blossoming field with significant potential for clinical impact. Imaging modalities like hematoxylin and eosin (HE)- or immunohistochemistry (IHC)-stained slides, unlike others such as CT, MRIs, or endoscopies, provide microscopic perspectives into tissue sections, allowing for the algorithms to utilize cellular and nuclear features like shape, size, color, and texture. Hence, the goal of this review is: (1) To cover major terminology and trainable tasks by ML and DL; (2) To briefly review the history of digital pathology; (3) To provide an overview of the current ML and DL histopathological imaging-based approaches in gastroenterology and hepatology; and (4) To discuss challenges and opportunities moving forward.

ML AND DL OVERVIEWThe FDA defines ML as “an artificial intelligence technique that can be used to design and train software algorithms to learn from and act on data”[3], where artificial intelligence is the development of computer systems capable of tasks deemed to require human intelligence. ML involves representation of samples or inputs by a fixed, user-determined set of features, then the application of a classifier that can





distinguish and separate classes or types within the set of samples based off those selected features. There have historically been a range of popular ML techniques. Some examples include logic-based approaches, such as decision tree-based methods like Random Forest (RF) classifiers, statistic-based approaches, such as Bayesian networks or nearest neighbor algorithms, and support vector machines (SVMs), which aim to find optimal hyperplanes to separate classes on high dimensional data feature spaces[4]. DL represents a modern, specific subset of ML that uses deep neural network architectures for feature extraction and predictions. A schematic of a deep neural network is shown in Figure 1.

The general goal of a DL algorithm is to connect an input, such as an image, to a desired output. The hidden layers in the network act as feature extractors, and a final layer aggregates and utilizes these extracted features to generate the desired output. Specifically, deep neural networks have an input layer that is followed by successive hidden layers, each containing nodes. Starting at the input layer containing data, nodes in each hidden layer compute weighted sums from outputs in the previous layer. Within each node, these weighted sums are then passed into activation functions, which are critical for neural networks as they introduce non-linear transformations onto data. Each hidden layer thus introduces additional mathematical complexity in an effort to transform the input into new, informative representations within a new feature space. This process of defining representations for inputs in this new feature space is called embedding, and the representations are deemed informative when they can be effectively utilized by the final output layer in the network to carry out desired predictions. Some popular activation functions include the sigmoid, tanh, and Rectified Linear Unit functions.

To train the model, gradient descent, a popular optimization method, is utilized to minimize the “loss function”, which quantifies model performance. Specifically, gradient descent minimizes the “loss function” by adjusting algorithmic weights at layer nodes, which directly affect the weighted sum calculations. As a result, the embedding process is iteratively improved to gradually tune and train the model for the task at hand. More detail is provided in the Model Training and Gradient Descent subsection below.

Common trainable tasksThe three most common tasks for DL approaches in imaging applications is in classi-fication, segmentation, and detection (Figure 2). Classification involves the prediction of a label for an input image, such as “Normal” vs “Cancer”. Segmentation involves the identification and localization of objects within a single image and outputs pixel-level designation of classes. Therefore, output segmentation maps will commonly have objects in the image colored or shaded based on their predicted class type. Lastly, detection, which is not a focus in this review, involves the identification of object classes in an image with a bounding box placed around it, such as in facial detection.

Levels of supervisionAn important aspect of these studies is image annotation of correct class labels. Due to the tremendous file size of these high-resolution images upon digitization of histological slides into whole-slide images (WSIs), analysis over an entire WSI at once is computationally infeasible. As such, WSIs are typically broken up into equally sized patches and require training patch-level models. Labeling therefore can occur at the level of the WSI and at the level of the patches.

When labelling the classes of individual patches from a WSI, this can be done in a fully supervised, weakly supervised, or unsupervised manner. This section will cover these levels of supervision in the context of classification tasks. An overview of these approaches is covered in Figure 3A.

The fully supervised approach involves dataset-wide annotation at the patch-level. For example, this may involve a dataset of patches extracted from WSIs with pathologist-annotated labels for each patch as “cancer” or “normal”. Thus, cancer positive WSIs will likely contain both types of patch classes. Though training iterations, the model will eventually learn to correctly predict the patch labels from just input patch images.

Weakly supervised methods concern annotations provided only at the WSI-level. Extracted patches from these WSIs are then run through algorithms that determine which patches were most important for the WSI-level label. Some possible approaches involve expectation-maximization methods[5] or multiple instance learning (MIL)[6-8]. In the context of the cancer positive WSIs, this would mean that the model eventually learns that the cancerous patches were most responsible for the WSI label, while healthy patches were not.



Figure 1 Example of general deep neural network architecture. Circles indicate nodes. Lines indicate feeding of layer node outputs into next layer nodes.

Unsupervised methods require no annotations to generate patch classes. While WSI-level labels may be provided, they are not utilized in patch class definition. These methods typically involve feature extraction from patches across the training dataset, followed by clustering approaches to define patch classes. For example, analysis of a dataset of cancer positive and negative WSIs may reveal dataset-wide patch-level classes for tumor, healthy, and fibrosis, although not all types may be present in each WSI. These approaches identify implicit patterns in the data to define these classes.

Fully supervised approaches require a tedious annotation process to provide correct output labels. As such, weakly supervised and unsupervised approaches hold the additional benefit of circumventing this labeling process and may be important in increasing throughput by decreasing annotation-related load. Another possible solution that is an active field of research is the generation of synthetic data that is indistinguishable from real world data. An example of this is general adversarial networks (GANs). GANs create synthetic data then have a ‘discriminator’ module that attempts to determine whether generated data is synthetic or real. The worse this discriminator performs, the better the GAN is at generating synthetic data. As such, computational approaches that effectively generate synthetic data across different classes may help develop labeled training datasets at high throughput.

Although the above methods discuss patch-level classifications, many biomedical imaging studies require a prediction at the WSI-level, such as a diagnosis. Often, this patch level information is aggregated by an additional classifier, and an overview of general approaches is provided in Figure 3B. This can be a ML classifier, such as an SVM or RF classifier that takes as input the relative counts of the different patch types per WSI to output a WSI-level prediction. This classifier can also be in the form of neural networks like recurrent neural networks (RNNs) and long short-term memory (LSTM) networks, covered later in this section, that take in variable length sequences of patches or patch representations as inputs to generate a WSI-level prediction. As the WSI-level label is typically clinically or biologically-informed, such as a diagnosis, prognosis, or grading, this part of the process typically receives supervision.

A variety of studies encompassing these approaches will be covered in this review in the two sections “Emulating the Pathologist” and “Beyond the Pathologist – Features Invisible to the Human Eye?”. A general diagrammatic overview of the approaches used in provided in Figure 4.

Model training and gradient descentIn practice, DL is performed in response to the quantification by a “loss function” of how well the neural network performed across the training dataset. As loss functions quantify model performance, they require knowledge of the correct output for each sample and are easiest to introduce with supervised learning concepts. For classi-fication, the output involves patch-level labels, and, for segmentation, the output will be images of the same dimensions as the inputs, where object classes in the image are distinguished by pixel-level color designation of classes (e.g., shading cancerous areas with one color and shading healthy areas with another).

For classification, on each epoch, or iteration, of training, the algorithm attempts to predict the label of every image, then calculates, from the loss function, a scalar loss value that captures the degree to which the model-predicted output labels were



Figure 2 Common trainable tasks by deep learning. A: Classification involves designation of a class label to an image input. Image patches for the figure were taken from colorectal cancer and normal adjacent intestinal samples obtained via an IRB-approved protocol; B: Segmentation tasks output a mask with pixel-level color designation of classes. Here, white indicates nuclei and black represents non-nuclear areas; C: Detection tasks generate bounding boxes with object classifications. Immunofluorescence images of mouse-derived organoids with manually inserted classifications and bounding boxes in yellow are included for illustrative purposes.

different from the correct, user-designated output labels across the dataset. A lower loss value would therefore indicate better performance of the model in predicting the correct image labels. As segmentation involves correct, human-designated outputs at the pixel-level, the loss function quantifies correct class predictions across every pixel in a segmentation map output.

Gradient descent is an optimization method that iteratively moves in the direction of the steepest slope to approach minimums and is utilized in DL to minimize the loss functions. Gradient descent starts from the loss function and propagates through previous layers to the first, identifying the gradients for each algorithmic weight at every network layer node, then incrementally adjusting these weights according to the gradients. This process occurs every epoch with the overall goal of improving model performance by minimizing the loss function. These weights affect the non-linear mathematical operations performed at each hidden layer node and thus serves as a way for the network to tweak these operations to eventually determine a feature space and sample representations most effective for the task at hand. As opposed to ML techniques that depend on human-designated features for classification, the DL-based sample representations can be interpreted as an extraction of features deemed best and driven by the neural network’s gradient descent optimization with respect to the loss function. This therefore introduces the common “black box” issue, where the meanings of these final representations, or extracted features, cannot be defined due to the high amount of mathematical complexity introduced onto the input image tensor by each of the layers in the network.



Figure 3 General deep learning training and prediction approaches. A: Examples pipelines for fully supervised, weakly supervised, and unsupervised learning methods for training patch classifiers are shown; B: Two pipelines translating patch-level information into whole-slide image-level predictions are shown. The top approach utilizes a patch classifier trained by one of the approaches in (A). The bottom approach uses a convolutional neural network feature extractor to generate patch representations that are fed into a long short-term memory or recurrent neural network. H&E: Hematoxylin and eosin; WSI: Whole-slide image; CNN: Convolutional neural network; RNN: Recurrent neural network; LSTM: Long short-term memory; CAE: Convolutional autoencoder.

Although gradient descent and loss functions are covered here, these are basic descriptions. For instance, the introduction of more complex loss functions that incorporate different learning constraints, study of approaches to take in multimodal inputs, and the development of novel network layers are all examples of highly active fields of research that add complexity to these concepts. Additionally, subcategories of gradient descent exist based on batch size, the number of images inputted before updating weights, such as stochastic gradient descent and mini-batch gradient descent. Other optimization algorithms like Adam optimization exist as well. Finally, various hyperparameters that affect model learning typically need to be tested over a range of values and each can affect different portions of training. Some major hyperparameters include learning rate, momentum, batch size, and number of epochs.



Figure 4 General overview of approaches. Overview of general machine learning - and deep learning-based approaches covered in the sections of this review are presented here. Whole-slide image (WSI) hematoxylin and eosin and Synthetic immunofluorescence (IF) WSI images in the Synthetic immunohistochemistry/IF Generation pipeline. Citation: Burlingame EA, McDonnell M, Schau GF, Thibault G, Lanciault C, Morgan T, Johnson BE, Corless C, Gray JW, Chang YH. SHIFT: speedy histological-to-immunofluorescent translation of a tumor signature enabled by deep learning. Sci Rep 2020; 10: 17507. Copyright© The Authors 2020. Published by Springer Nature. TIL: Tumor-infiltrating lymphocyte; H. pylori: Helicobacter pylori; NASH CRN: Nonalcoholic Steatohepatitis Clinical Research Network; IHC: Immunohistochemistry; IF: Immunofluorescence; H&E: Hematoxylin and eosin; WSI: Whole-slide image; GAN: General adversarial network.

Many of the common alternatives and hyperparameters are covered in this review by Shrestha and Mahmood[9].

Imaging data and convolutional neural networksFor imaging data, a specific type of neural network, called convolutional neural networks (CNNs), need to be utilized as inputs are in the form of 3-dimensional matrices, or tensors. A key concept is that any image can be represented by its numerical, pixel intensity values. For example, a 224 × 224 pixels grayscale image will be a 224 × 224 tensor with all values on a range from 0-1. A 224 × 224 RGB image, however, will be a 224 × 224 × 3 tensor with all values on a range from 0-255, though values are typically normalized to a range from 0-1.



To work with these tensors, convolutions need to be implemented in the form of convolutional network layers. Convolutions can be interpreted as the sliding of another tensor, or filter, typically of much smaller size than the input, over the input tensor. The filter slides from left to right of the input tensor, then moves down and repeats the process from left to right again. The mathematical operations can be considered as an expansion of the weighted sum and activation function approaches described earlier in this section. As the filter slides over the input, a weighted sum is calculated incorporating every cell overlap between the two tensors and generates a new output tensor that is then passed to an activation function. Like the layer nodes, convolution filters have weights that are trainable by gradient descent. Each hidden layer will perform similar operations on outputs from the previous layer but may have varying filter sizes or activation functions. Since the non-linearities introduced by the various layers sequentially add complexity, the earlier layers are believed to encode simpler features like edges, while the latter layers capture even more abstract features.

Analysis of images requires consideration of relationships between adjacent regions to capture spatial information. Though the weighted sum calculation in convolutions does account for neighboring pixels, the receptive fields are still quite small. Pooling layers are also carried out by filters and further aggregate local information from previous layers. The two major types are max and average pooling. For a 4 × 4 pooling filter, this would mean selecting the maximum value within the 4 × 4 receptive field in max pooling or averaging the 16 values for average pooling, as opposed to performing the weighted sum calculations that would occur in convolutions. Importantly, pooling layers do not have any trainable weights and represent fixed operations.

Convolutional layers typically reduce tensor height or width while increasing number of channel dimensions. Pooling layers do not affect channel dimensions but reduce tensor height and width. Thus, a series of convolutional and pooling layers will serve to reduce tensor height and width and increase channel dimensions relative to the original input.

The outputs of convolutional and pooling layers are often 3-dimensional and need to be flattened into a 1-dimensional vector towards the end of the network. The flattened 1-dimensional vectors then feed into a full-connected layer, a feed forward layer where nodes calculate weighted sums from the flattened vector input and pass values to an activation function. Finally, these outputs are utilized for the final classi-fication layer, which typically uses a softmax activation function in classification tasks. The softmax layer will have the same number of nodes as the number of possible classes to predict. The outputs of this layer will sum up to 1 and can be interpreted as the relative probabilities for each class prediction with each class corresponding to one softmax node.

Common landmark neural network architecturesAlthough the focus of this review is not to delve deeply into the different types of neural network architectures, those that appear will be covered briefly here to provide background. An overview of the network structures is shown in Figure 5.

The visual geometry group (VGG)-16 and VGG-19 networks published by Simonyan and Zisserman[10] consist of sequences of convolutions and pooling operations followed by fully connected layers for a total of 16 or 19 layers, respectively. The authors incorporated very small 3 × 3 convolutional filters and demonstrated the capacity to create a network that had a lot of layers relative to other networks at time of publication.

The Inception network was initially published in 2015 by Szegedy et al[11], though several improved versions, such as the Inception-v3 used by some studies in this review, have since been developed. The major contribution of these networks is the introduction of the inception module that performs 1 × 1 convolutions, 3 × 3 convolutions, 5 × 5 convolutions, and max pooling at the same layer. An n × n convolution refers to a convolutional layer with an n × n dimension filter. The general concept is that predicting the optimal convolution filter size may depend on the image at hand. Instead of selecting a single filter size, more may be learned by incorporating information from convolutions with different receptive fields along with max pooling.

He et al[12] introduced ResNets which contain a the residual block with a skip connection. Deep neural networks with many layers often experience the issue of vanishing or exploding gradients. With the high amount of mathematical complexity introduced by many layers, backprogating these gradients can approach local minima and maxima and impede training. Since calculated gradient values are used to update layer node weights, a value of zero means the weight barely shifts, while infinity causes too significant of a change. Without getting into much detail, these residual blocks allow for skipping of portions of the network where this occurs to allow for





Figure 5 Common landmark network architectures. Overviews of landmark network architectures utilized in this paper are presented. A: The visual geometry group network incorporates sequential convolutional and pooling layers into fully connected layers for classification; B: The inception block utilized in the inception networks incorporates convolutions with multiple filter sizes and max pooling onto inputs entering the same layer and concatenates to generate an output; C: The residual block used in ResNet networks incorporates a skip connection; D: Recurrent neural networks (RNNs) have repeating, sequential blocks that take previous block outputs as input. Predictions at each block are dependent on earlier block predictions; E: Long short-term memory network that also has a sequential format similar to RNN. The horizontal arrow at the top of the cell represents the memory component of these networks; F: Fully convolutional networks perform a series of convolution and pooling operations but have no fully-connected layer at the end. Instead, convolutional layers are added and deconvolution operations are performed to upsample and generate a segmentation map output of same dimensions as the input. Nuclear segmentation images are included for illustration purposes; G: U-Net exhibits a U-shape from the contraction path that does convolutions and pooling and from the decoder path that performs deconvolutions to upsample dimensions. Horizontal arrows show concatenation of feature maps from convolutional layers to corresponding deconvolution outputs. VGG: Visual geometry group.

continued training. This has allowed for networks like the ResNet-34 and ResNet-50, which have 34 and 50 layers, respectively, and for extraction of even higher dimensional features.

Though typically for sequential or temporal data, RNNs and LSTM networks are utilized by a few studies covered in this review. RNNs were developed earlier and process sequences of data. Each layer performs the same task; however, the decisions made at each layer is dependent on previous outputs. This capacity has been important for speech data as words typically have a relationship with the previous word in a sentence. LSTMs have similar use cases but with a superior ability to identify longer term dependencies and relationships than RNNs. In the context of this review, RNNs and LSTMs are useful in being able to take in a variable length sequences of inputs to provide one output. As WSIs are composed of varying numbers of patches, these networks are implemented to aggregate patch information into a WSI-level output. In these studies, patch sequences are typically shuffled to ensure input patch ordering has no effect on the output and focus on leveraging the capacity to take in variable length inputs as opposed to the temporal component. While not utilized in studies covered in this review, bidirectional encoder representations from transformers is a more modern technique that, instead of only reading sequences left-to-right or right-to-left, considers bidirectional contexts when making predictions[13].

Segmentation tasks, which generate a segmentation map output of the same dimensions as the input, require specific types of networks. Long et al[14] introduced the fully convolutional network (FCN), which replaces the fully connected layers described earlier in the CNN description, with additional convolutional layers. Forgoing the flattening operation and fully connected layers maintains the spatial relationships in the 3-dimensional tensors. The FCN then performs deconvolutions, also known as transpose convolutions, which in practice perform the opposite function of convolutions. Deconvolutions increase the height and width dimensions of inputs, allowing for eventual generation of an output with the same dimensions as input. FCNs are able to generate probability heatmaps of possible segmentation classes.

Ronneberg et al[15] built upon the FCN by developing the U-Net, named due to its U-shaped network structure. U-Net has 4 convolutional layers to generate a bottleneck tensor, then 4 deconvolutional layers to up-sample the bottleneck tensor back to the



original input dimensions. Additionally, each deconvolutional layer receives input feature maps from the corresponding convolutional layer. As the convolutional layers occur earlier and encode spatial relationships more, concatenating these feature maps to the deconvolutional layer outputs helps the network with localization, which is important for segmentation tasks.

Quantitative performance metricsLastly, it is important to understand the quantitative metrics used to assess model performance. The most popular are accuracy, precision, recall, and F1 score. These metrics are all calculated based off the total number of true positives, true negatives, false positives, and false negatives, and their formulas are shown in Figure 6. Area under the curves (AUCs) are also a common metric and involve calculating this metric from a plot of sensitivity vs (1-specificity).

RISE OF DIGITAL PATHOLOGYWhile ML and DL approaches have been applied to many input data types, the fields of computer vision and image analysis owe much of its popularity to the emergence of digital pathology. An early milestone for digital pathology was the development of software to view histology images, such as the Virtual Microscope developed from 1996 to 1998 that had to take advantage of existing methods for handling satellite and earth science data[16]. The Virtual Microscope was further refined to allow for capabilities like data caching, support for simultaneous queries from multiple users, and precomputed image pyramids. Modern viewers have continued to grow capabilities and allow for collaborative, multi-user work on the same images, annotations, the ability to zoom and inspect WSIs, and construction of imaging datasets and cohorts.

At the time the Virtual Microscope was being developed, WSI scanners were not yet available, so histology sections had to be digitally tiled up before uploading into viewing systems. Today, many commercially available WSI scanners that can scan entire slides exist, and this issue can be avoided. Furthermore, there are now two FDA-approved digital pathology platforms: the Phillips IntelliSite Pathology Solution[17] and the Sectra Digital Pathology Module[18].

Digital pathology comes with some clear benefits, including the ease of sample storage and access through software and the capacity to perform image analysis directly on digitized WSIs. However, the utilization of WSIs comes with its own set of quality concerns, which are covered nicely in the review by Kothari et al[19]. In brief, these methods can introduce image artifacts and batch effects. Image artifacts can occur both from scanning or preparation of samples. Some examples include blurring of tissue regions due to microscope autofocusing mechanisms, shadows in the image, pen marks from pathologists, or folding of tissue. In these cases, care needs to be taken to remove the artifacts, such as in the case of pen marks, or to filter out image areas with issues like blurring. Batch effects can occur due to the individual preparing the sample, the specific reagents used, the site of acquisition, or the microscope type. To address some of these concerns, studies frequently apply methods to normalize the color or pixel intensity values across their images. However, certain factors into batch effect exist that cannot be addressed computationally, such as varying patient population demographics based on location. As such, there is a strong need to incorporate multicenter data sources to develop more generalizable models, or a realization that certain models may only be used within specific demographics.

So long as these quality control concerns are recognized, however, digitized WSIs have the potential to be further adopted within practices. They are high-resolution gigapixel scale images that can be stored digitally and distributed for research. Furthermore, Al-Janabi et al[20] examined the feasibility of utilizing WSIs instead of classic light microscopy for diagnosis of gastrointestinal tract pathologies. For 100 cases of biopsies and resections along the entire gastrointestinal tract that had been diagnosed by light microscopy a year earlier, the authors recruited the same pathologists to re-diagnose their own cases using digitized WSIs. The study showed a 95% concordance between light microscopy- and WSI-based diagnoses with the discordant 5% of cases showing no clinical implications, highlighting the potential for the adoption of WSIs into the diagnostic workflow.

Finally, the growing popularity of DL imaging approaches owes itself to the development of computational hardware and software[21]. Graphical Processing Units (GPUs) were primarily used in the setting of video games, but their high capacities for



Figure 6 Quantitative performance metric equations. Quantitative metrics are based off of true positive, true negative, false positive, and false negative counts from results. Equations for precision, recall, F1 score, accuracy, and specificity. Precision is also known as the true positive rate, and recall is also known as sensitivity. TP: True positive; TN: True negative; FP: False positive; FN: False negative.

parallel computation were found to be ideal for DL methodologies. GPUs significantly increased DL model training speeds relative to Central Processing Units and played a strong part in the popularity growth of these methodologies. In addition, the DL community has been aided by the presence of open-source libraries for efficient DL GPU implementation. Some examples include PyTorch, Caffe, and Tensorflow. These libraries can easily load up and train major, landmark network architectures, simplify design of new networks relative to manual coding, and encourage cohesion amongst researchers by having standardized coding styles and pre-defined functions for common operations within each library. The emergence of digital pathology viewers, WSI scanners, GPUs, and open-source DL libraries has led to Pathomics, defined as the generation of quantitative imaging features to describe the diverse phenotypes found in tissue sample WSIs[22]. The foundations for DL-based imaging fields have thus been set and have welcomed a new influx of applications and studies within the biomedical disciplines.

EMULATING AND AUTOMATING THE PATHOLOGISTOne clear application of these ML- and DL-based methodologies is to replicate the tasks of pathologists. A well-trained model has the benefits of eliminating interob-server variability amongst pathologists and of achieving a level of throughput impossible to humans. This section will cover research in classification for both cancer and non-cancer pathologies and in segmentation tasks aimed at the identification of structures or cell types within images.

Cancer classificationThe most popular histopathological application of these methods in gastroenterology and hepatology occurs in the classification of cancers. The general concept is that if these classifications can be made with the human eye, then the models should be able to learn to make such distinctions themselves.

Colorectal: Thakur et al[23] recently published a comprehensive review of artificial intelligence applications in colorectal cancer pathology image analysis, but several papers will still be highlighted in this review.

In an earlier study, Yoon et al[24] trained a customized VGG-based network architecture on 28 normal and 29 colorectal cancer HE-stained slides that were tiled up into 256 × 256 pixel patches. After testing several, custom VGG-based networks, the best model had an accuracy of 93.5% with a sensitivity and specificity of about 95% and 93%, respectively, in determining if an image patch was cancer vs healthy. This study showed promise in the relatively simpler binary classification task of tumor vs normal.

In a study published since, Sena et al[25] took the classification task another step further to train a model to classify between normal mucosa, early neoplastic lesion, adenoma, and cancer in HE-stained samples. The authors used a custom network architecture similar to VGG with four sequential convolutional and pooling layers followed by dense layers. Even with this relatively simple network architecture, the model achieved about a 95% accuracy in predicting the exact label for its larger 864x648 pixel patches across the four classes.



While the above patch-level performances are encouraging, clinical diagnoses are typically at the slide level. To address this, researchers often train additional classifiers, in addition to the patch-level ones, that can make a prediction at the WSI-level by aggregating patch-level information. An example of this is the study by Iizuka et al[26], which initially trained the Inception-v3 network to classify between non-neoplastic, adenoma, and adenocarcinoma for 512 × 512 pixel patches from HE-stained colorectal and gastric biopsy WSIs. The authors utilized the trained Inception-v3 classifier as a feature extractor to generate 715-length feature vectors, the represent-ations, for each patch. The sequence of feature vectors of every patch in a WSI are used as the input and the WSI-label as the output in training a subsequent RNN. Though RNNs are typically used in temporal data, they have the advantage of being able to take in variable length, sequential inputs in generating final output labels. This is an important feature considering that every WSI has a varying number of total patches sampled. The trained RNN can thus predict the WSI diagnosis by aggregating extracted feature vectors of all patches in that sample. The study achieved WSI-level prediction AUCs of up to 0.97 and 0.99 for gastric adenocarcinoma and adenoma, respectively, and 0.96 and 0.99 for colonic adenocarcinoma and adenoma, respectively. Of note, the gastric classifier model outperformed pathologists in classification accuracy when pathologists were given a 30 s time limit, which is the average amount of time the model takes per WSI. The gastric model achieved an accuracy of 95.6% compared to the 85.89% ± 1.401% (n = 23) for the pathologists.

Russakovsky et al[27] utilized the AlexNet architecture pretrained on ImageNet, a large collection of non-biomedical, natural images with 1000 classes, as a feature extractor for classification and patch-based segmentation tasks on brain and colorectal HE datasets[28]. To address the common lack of annotated training datasets for these DL methodologies, the authors took this approach to demonstrate the potential of CNNs pretrained on non-biomedical images as feature extractors in biomedical applic-ations. For the classification task, CNN-extracted patch representations for each WSI were pooled and condensed by feature selection methods before input into a SVM classifier to generate a WSI prediction. In colorectal cancer classification, the network was trained for a binary classification task to recognize tumor vs normal, and a multiclass classification task to recognize between adenocarcinoma, mucinous carcinoma, serrated carcinoma, papillary carcinoma, cribriform comedo-type adenocarcinoma, and normal. SVM with these CNN features as inputs outperformed SVM with a set of manually extracted feature as inputs in both classification tasks, achieving a 98.0% accuracy in binary and 87.2% in multiclass classification compared to 90.1% and 75.75%, respectively. The segmentation task involved no feature pooling as a SVM was trained to utilize patch-level CNN features to generate a patch classi-fication prediction. By utilizing overlapping patches, pixel-level class predictions can be designated based off an ensemble method aggregating overlapping patch predictions. Again, the SVM with CNN prediction inputs outperformed the SVM with manually extracted feature inputs, showing an overall accuracy of 93.2% compared to 77.0%.

Although the above studies demonstrate the value of patch-level classifications in determining a WSI-level prediction, the annotations required for such a training dataset are highly time-consuming. Additionally, clinically archived tissue specimens are typically accompanied only by the WSI or patient-level diagnosis. MIL encompasses approaches to obtain insight into patches or patch-level features most critical for designation of the WSI-level label. MIL thus represents a possible way to generate effective patch classifier models utilizing only WSI-level annotations.

In MIL, each WSI is considered a bag in which multiple instances, or patches, are contained. If any one of these patches are positive for cancer presence, then the WSI can be determined to be cancer positive. While the instance-level patches have their own classes, these are unprovided or unknown. As such, the goal of MIL is to train an instance-level classifier based on the WSI-level labels to determine these unknown patch labels.

Xu et al[7] applied the MIL-Boost algorithm for HE colorectal slides for binary cancer vs non-cancer classification. In brief, the MIL-Boost algorithm trains the instance-level classifier by “boosting”. “Boosting” refers to the successive training of weak classifiers, where each classifier improves by adding weights to incorrect predictions made by the previous classifier. Here, weak classifier weights are iteratively updated by gradient descent on the bag-level classifier loss function. Backpropagation occurs along patch instances that most negatively affected the predicted bag-level cancer positivity relative to the true WSI-level label and adjusts algorithmic weights to reduce these errors on the next iteration of the weak, instance level classifiers. This process with weak classifiers is repeated until the loss function is



minimized and an effective instance-level classifier is developed. The authors demonstrated superior performance of this approach (96.30% accuracy) as opposed to a fully supervised, patch-annotated approach (95.40%) in the binary cancer vs normal classification task.

As patch instances make up a WSI bag, a MIL-type bag representation can be considered to be a collection of patch instance representations where positive instances are provided a higher weight. To this end, Ilse et al[6] utilized a CNN to extract patch feature representations, then incorporated an attention mechanism to output a weighted average of all instances in a bag. Notably, the attention mechanism weights are determined by a two-layer neural network, meaning they are trainable unlike conventional MIL pooling operators that calculate maxes and means. These weighted bag representations can also be used to identify the most important instances for the bag prediction. The authors utilized a published HE colorectal cancer dataset[29] with annotated nuclear patches for epithelial, inflammatory, fibroblast, and miscellaneous and formed the MIL problem so that a bag is considered positive if at least one epithelium-positive patch exists in a WSI. This MIL approach trained an epithelial patch classifier with an accuracy and F1 score of approximately 90% and AUC of 96.8%. Furthermore, the authors could threshold for only instances with high weights, leading to the visualization of epithelial regions in the original HE WSI. Although the focus of this paper was on superior performance of the neural network tunable attention-mechanism relative to fixed alternatives, the final performance metrics lend support to the capacity of MIL approaches in training patch-level classifiers from WSI-level annotations.

Another major part of the colorectal cancer field is in the histopathological evaluation of HE-stained polyps to determine cancerous potential. In 2017, Korbar et al[30,31] trained a ResNet-based network to detect between hyperplastic polyps, sessile serrated polyps, traditional serrated adenoma, tubular adenoma, and tubulovillous/villous adenoma. The authors trained a patch-based classifier, then designated WSI-level predictions according to the patch-level class prediction that was most prevalent in the sample, given that at least 5 patches outputted that prediction[31]. This model achieved a 93.0% overall accuracy [95% confidence interval (CI): 89.0-95.9]. In another study, the authors utilized the same network architecture to identify the 5 classes but focused on implementing Gradient-weighted Class Activation Mapping (Grad-CAM) approaches to address model interpretability[30]. Grad-CAM can backpropagate from a patch’s predicted class label to identify the regions in the input image that contributed most to the prediction. Though this was an early approach, the study showed promising potential for these Grad-CAM approaches to help identify regions of interests (ROIs) that were most influential in the patch-level polyp classification.

Esophageal: While different from Grad-CAM, Tomita et al[32] utilized a related concept in implementing attention-based mechanisms for weakly supervised training to detect 4 classes—normal, Barrett’s esophagus without dysplasia, Barrett’s esophagus with dysplasia, and esophageal adenocarcinoma—from HE-stained esophageal and gastroesophageal junction biopsies. The approach involved breaking up a WSI into patches, from which a CNN would extract features. Each WSI could then be represented as a feature map that is an aggregated patch grid of extracted feature vectors. These feature maps serve as inputs to the attention-based model, the goal of which is to identify the regions of the input feature maps most important for the output label classifications. Therefore, a concept is shared with Grad-CAM in identifying input image regions most influential to the class predictions. Unlike Grad-CAM, the attention-based model will learn to add weights to influential areas in the feature map to aid in final model classification performance. Of note, this process is considered weakly supervised because image output labels are only provided at the WSI-level, as opposed to the patch-level, yet the most influential patch types can be distinguished. The model manages to learn on its own the most salient image features and regions that were most important for the WSI label. The approach here achieved an overall accuracy of 83.0% (95%CI: 80-86) in identifying the 4 classes, outperforming the supervised baseline with an overall accuracy of 76% (95%CI: 73-80) that depends upon extraction of patches from ROI tediously annotated by pathologists. It should be noted, however, that the model achieves an F1 score of 0.59 (95%CI: 0.52-0.66) and the supervised baseline an F1 score of 0.50 (95%CI: 0.43-0.56) possibly indicating a high rate of false positives and negatives.

Moving even further away from supervised learning, Sali et al[33] demonstrated superior performance of unsupervised approaches in classifying HE-stained WSIs to be dysplastic Barrett’s esophagus, non-dysplastic Barrett’s esophagus, and squamous



tissue relative to supervised methods. The supervised approach was analogous to Iizuka et al[26]. Training patches labeled by pathologists were used to train the model, then an SVM or RF classifier aggregated the patch-level information for the WSI-level prediction.

The unsupervised feature extraction approach involved a deep convolutional autoencoder (CAE). Deep CAEs are broken up into an encoder and decoder branch. The encoder branch typically applies a series of convolution and pooling operations to act as a feature extractor that outputs a bottleneck feature vector. The decoder branch upsamples back from the bottleneck feature vector and reproduces the original image. Here, the loss function minimizes the differences between the input image and reproduced version, thereby enforcing that the bottleneck feature vector is an effective representation of the input. A helpful analogy is when one zips files on the computer. The process compresses the original file to a smaller memory size (encoding), but then still allows one to re-generate the full-size, original file (decoding). As one knows the zipping mechanism works, he or she can confidently share zipped file versions to others, instead of the larger, original file.

Once the deep CAE is trained, it can be utilized as a feature extractor for all patches in one’s training dataset. Then, by performing clustering approaches, such as k-nearest neighbors (k-NN) or Gaussian mixture models (GMM), on all feature vector-transformed patches, patch types or classes across the dataset can be defined. A SVM or RF classifier can be trained to predict the WSI class by using the relative proportions of the different patch class types in the sample. For WSI-level inference, the deep CAE extracts feature vectors from all patches in the WSI, bins and counts the number of patches per clustering-defined patch type, then utilizes the trained SVM or RF classifier to generate the WSI prediction. This process is called unsupervised because the different types of patches in the WSI are determined by the algorithm independent of any labelling. This is in contrast to the supervised approach, where a CNN was trained to classify between human-defined Barrett’s esophagus, non-dysplastic Barrett’s esophagus, and squamous tissue patch types. The unsupervised GMM method showed good performance with weighted averages for accuracy, AUC, F1, precision, and recall all above 90%. In contrast, the metrics for the supervised approaches ranged from 50%-80%.

Gastric: Though gastric pathologies and cancers will be covered further in other sections of this review, not a tremendous amount of literature exists regarding just classification of gastric cancers. Leon et al[34] demonstrated that, in gastric cancer classification, inputting image patches as a whole into a custom, Keras sequential model shows superior performance than utilizing nuclei extracted from these image patches as input. This may be explained by the fact that the whole image patch contains morphological features that might be important for classification, while the cell input approach sacrifices those portions of the image. The other major study to note is the one by Iizuka et al[26] mentioned earlier, which showed impressive performance in classifying gastric and colorectal adenomas and adenocarcinomas.

Liver: As in the mentioned studies by Iizuka et al[26] and Ilse et al[6], CNNs can be used to extract patch feature representations. These representations are 1-dimensional vectors comprised of numerical, float values, and higher values can be interpreted as features most important, or highly activated nodes, for the prediction at hand, while lower values may be interpreted as important for the other non-predicted class.

Since these feature values can be reflective of their relative importance in the predicted class, Sun et al[8] used a CNN to extract patch representations from HE-stained WSIs, performed a pooling operation to aggregate patch features at the image level, then sorted the representations to organize activation values from high to low importance in terms of liver cancer prediction. The authors selected a range of top-k and bottom-k features from this sorted list to use in patch representations, driven by the idea that high activations should indicate features important for cancer classific-ations, while the lower activations should correspond to normal. The variable length representations dependent on k were tested to generate condensed patch represent-ations in training a binary cancer vs normal classifier. A value of 100 for k was deemed optimal, and the authors used the patch classifications to predict WSI cancer vs normal status. The approach achieved an accuracy of 98%, a recall of 1.0, and an F1 score of 0.99.

In addition to the design of effective image classification algorithms, the incorporation of these methodologies into clinical workflow is important to consider. Kiani et al[35] trained a DenseNet CNN to classify between hepatocellular carcinoma and cholangiocarcinoma from HE image patches and developed a diagnostic support



tool that outputs predicted classes with probabilities and class activation maps (CAMs) to highlight areas of the input patch important for the prediction. The effects of the diagnostic support tool were analyzed and revealed that, while correct classifier predictions significantly improved accuracy, incorrect classifier predictions significantly decreased accuracy of diagnosing pathologists. Thus, this study highlights the important notion that the damaging effects of incorrect and misleading classifiers need to be strongly considered before clinical implementations.

Pancreatic neuroendocrine: IHC stains are another common technique applied to histopathological samples. The ability to detect specific antigens can be important for the characterization of certain cancer types. In pancreatic neoplasms, for example, the Ki67 stain is used to define proliferative rate and assign grades to pancreatic neuroen-docrine tumors (NETs). However, this process is complicated by Ki67 stain positivity in both tumor and non-tumor regions. To address this issue, Niazi et al[36] trained an Inception-v3 network pretrained on ImageNet in a transfer learning setting to detect tumor and non-tumor regions on Ki67-stained pancreatic NET WSIs. As with Xu et al[28], the concept is that learned features from training on ImageNet should be beneficial within the biomedical setting. By freezing weights on all layers except for the final classification layer, the authors ensured that the feature extraction portion of the network remains unchanged. Training thus affects only the manner in which the classification layer utilizes patch representations instead of affecting the feature extraction itself. The trained model was used to create probability maps for tumor and non-tumor predictions for every pixel in the WSI, then thresholded by 0.5 to generate masks for each class. As each pixel in the image was then assigned to its most probable class, the output generated a segmentation map-type output that is shaded by predicted classes. In identifying tumor and non-tumor regions on a Ki67-stained IHC slide, the model showed about 96%-99% overall accuracy with 97.8% sensitivity and 88.8% specificity.

Cancer lymphocyte interactions: In addition to the cancer itself, other cell types exist within the microenvironment. To address this, Saltz et al[37] trained a VGG-16 network to identify tumor-infiltrating lymphocyte (TIL) containing patches across 13 The Cancer Genome Atlas (TCGA) HE-stained tumor types. The study identified four types of TIL infiltration patterns: Brisk Diffuse, Brisk Band-like, Non-Brisk Multifocal, and Non-Brisk Focal. The study also found associations between TIL infiltration patterns, cancer type, inflammatory response subtype, and molecular cancer subtypes and supports the notion that spatial phenotypes have the exciting potential to correlate with molecular findings.

Cancer nuclei classification: Another avenue of classification tasks in cancer applic-ations has been in the study of nuclei. Pathologists are able to utilize visual, nuclear information, such as aberrant chromatin structures, to identify cancerous cells. Thus, groups have worked on replication this task of nuclei classification.

Chang et al[38] extracted HE-stained nuclei, used immunofluorescence (IF) pan-cytokeratin (panCK) stains aligned to the HE slide by image registration methods to label the HE-extracted nuclei as cancerous or non-cancerous, then trained a CNN to make these distinctions from just an HE input. The panCK-defined cancer positivity approach eliminated the need for tedious, pathologist annotations on the HE images and achieved a 91.3% accuracy with 89.9% sensitivity, 92.8% specificity, and 92.6% precision in classifying cancerous vs non-cancerous nuclei on the independent test set.

Sirinukunwattana et al[29] implemented a spatially constrained CNN to identify pixels most likely to represent the center of nuclei, then trained a subsequent CNN classifier to predict whether the nuclei came from an epithelial, inflammatory, fibroblast, or miscellaneous cell in colon cancer. The authors also implement a Neighboring Ensemble Predictor in the nuclei classifications, which, when predicting the class of a nuclei, incorporated the predictions from all neighboring patches. This approach achieved a weighted average F1 score of 0.784 and AUC of 0.917 in the nucleus classification tasks and a weighted average F1 score of 0.692 in the combined nucleus detection and classification tasks. In a follow up study since, Shapcott et al[39] utilized this nuclei classification algorithm to quantify the four cell types to correlate cellular proportions with different clinical variables in TCGA colorectal cancer patients. This led to findings such as samples with metastasis having more fibroblasts with fewer epithelial and inflammatory cells, samples with residual tumor having more fibroblasts and fewer epithelial and inflammatory cells, and that both venous and vascular invasion were associated with more fibroblasts.



Non-cancer classificationThough much focus in image classification has been in cancers, other image classi-fication applications exist and are highlighted here.

Celiac disease, environmental enteropathy, and nonspecific duodenitis: Wei et al[40] trained a ResNet-based model to classify between celiac disease, normal tissue, and nonspecific duodenitis on HE-stained WSIs with accuracies of 95.3%, 91.0%, and 89.2%, respectively. This was a supervised, patch-based approach for training, and WSIs were predicted to be nonspecific duodenitis if more than 5 patches were classified as such or predicted to be the dominant patch class otherwise.

In a similar supervised fashion, Srivastava et al[41] trained a ResNet model on duodenal HE biopsies to classify between celiac disease, environmental enteropathy, and normal tissue. Patch classifications were aggregated for the WSI prediction and returned an overall 97.6% accuracy.

Sali et al[42] also trained a ResNet model, but for the task of Marsh Score-based grading of celiac disease severity using HE-stained duodenal biopsies. The authors utilized a CAE to generate patch representations, then performed a 2-class k-NN clustering to filter out useless, non-tissue containing patches. The tissue-containing patches were then used for supervised training of the ResNet model to recognize between Marsh scores of I, IIIa, IIIb, and IIIC. Again, patch predictions were aggregated for a WSI-level prediction. The model showed an accuracy and F1 score of around 80-90% for all classes and also implemented CAM approaches to localize certain cell subsets contributing to some of these Marsh Score categories.

In another study, Sali et al[43] took a novel, hierarchical approach towards training a VGG classifier to detect 7 classes: Duodenum-celiac disease, Duodenum-Environ-mental enteropathy, Duodenum-normal, Ileum-Crohn’s, Ileum-normal, Esophagus-eosinophilic esophagitis, and Esophagus-normal. In addition to having the classifier predict the disease type with the final classification layer, the approach incorporated another output branch in the VGG network to predict anatomic location. The loss function combined outputs of the two branches and enforced the network to learn both anatomic origin and specific disease type. Additionally, the anatomic origin branch occurs before the final classification layer, meaning that the network needs to correctly determine the anatomic origin first, before homing in on the specific diagnosis. Across all 7 classes, the model exhibited F1 scores ranging from 0.714 for Duodenum-normal to 0.950 for Duodenum- Environmental enteropathy.

Helicobacter pylori gastritis and reactive gastropathy: Similar to the other examples, these represent diagnoses that can be made from HE-stained specimens. Martin et al[44] trained the commercially available HALO-AI CNN to classify between Helico-bacter pylori, reactive gastropathy, and normal in gastric biopsies. The model achieved sensitivity/specificity pairings of 73.7%/79.6%, 95.7%/100%, 100%/62.5% for normal, Helicobacter pylori, and reactive gastropathy, respectively.

Klein et al[45] developed a model that combines image processing techniques with DL. The authors utilized image processing techniques on both Giemsa- and HE-stained slides to identify potential Helicobacter pylori regions, then had experts review these as being positive or negative for Helicobacter pylori presence. These could then be utilized as input-output pairs to train a VGG-style network. The main goal of this paper, however, was to create a clinical decision support system that utilized the trained model and directs pathologists to Helicobacter pylori hotspots using Grad-CAM-style methodologies. Although this clinical decision support approach showed higher sensitivity than just microscopic diagnosis (100% vs 68.4%), specificity was lower than with just microscopic diagnosis (66.2% vs 92.6%).

SegmentationSegmentation generally refers to operations that localize and detect cells and structures within a WSI. As pathologists can detect these objects within a sample, the goal is to train models to replicate these tasks.

The gland segmentation in colon histology images challenge contest challenge: A key contributor to the progression of computer vision disciplines has been the presence of challenges that provide a dataset and rank submitted models based off of performance-related quantitative metrics such as F1 scores or AUC values. One example of this is the gland segmentation in colon histology images challenge contest (GlaS) that was held in 2015[46]. These challenges help to stimulate computational disciplines. For one, the announcement of the challenge itself encourages researchers worldwide to address and tackle the problem. Compared to standalone papers, these



challenges also have the advantage of pitting the best models against each other to generate a clear benchmark for state-of-the-art performance.

Furthermore, even after completion of the challenge, groups will continue to optimize their algorithms and will have the ability to compare performance to previous high rankers in the challenge. Even since the GlaS challenge, numerous groups have continued to work on gland segmentation models by incorporating novel mechanisms. In 2016, Xu et al[47] added multichannel feature extractions for region and edge probability maps that were then fed into the final CNN for instance segmentation. Also in 2016, BenTaieb et al[48] applied topological and geometric loss functions into their FCN-based model. In 2019, Graham et al[49] introduced a new network component, the minimal information loss unit, that re-introduces resized versions of the original input image to combat the loss of information that accompanies downsampling from the successive convolution and max-pooling operations that occur in neural networks. Most recently in 2020, Zhao et al[50] incorporated spatial attention to weight important spatial locations and channel attention to weight important features to improve gland segmentation performance.

Non-colon gland segmentations: In general, segmentation methodologies require an additional step of development compared to classification tasks. For example, identifying glands in colonic mucosa is an important task but needs additional interpretation to be useful in the clinic. Some possibilities include quantifying the total number of glands or extracting shape-based glandular information to feed into a colorectal cancer classifier. Classification tasks like “Tumor” vs “Healthy”, on the other hand, often already have a clear path towards clinical integration within the pathologist diagnostic workflow.

Reflective of this, many histopathological segmentation studies in gastroenterology and hepatology tend to be focused on optimizing segmentation results themselves, as opposed to continuing onto the translational application. However, high performance segmentations are critical in developing the downstream, clinically impactful algorithms. While some studies have continued onto the next step, the next few years will likely see some more of these segmentation studies bridging into more transla-tional studies.

To highlight some examples, Xiao et al[51] segment out liver portal area components for eventual hepatitis grading. Extraction of features from these segmented structures to train a classifier to grade hepatitis will likely be the next step of this process. Xu et al[52] used a patch-based segmentation approach to identify epithelial and stromal regions in HE-stained breast and epithelial growth factor receptor-stained colon cancer slides as tumor-stroma ratios are recognized to have prognostic value. Here, the next step would be to assess the impact of algorithm-derived epithelium and stroma ratios in patient prognosis or cancer classification. Similarly, to address the eventual use case of segmenting tumors to assess pre-surgical tumor burden, Wang et al[53] used multitask and ensemble learning techniques for pixel-wise HE hepatocellular carcinoma segmentation. For eventual use in computer-assisted diagnosis systems, Qaiser et al[54] develop a fast HE colorectal segmentation algorithm that defines persistent homology profiles to capture morphological differences between normal and cancer nuclei. The emergence of more directly translational follow up studies and validations should be exciting and will be important to monitor.

Moving downstream with segmentation outputs: Some studies have entered this second phase and will be highlighted in this section. Awan et al[55] utilized a modified version of U-Net to perform colon gland segmentation on HE-stained colorectal adenocarcinoma patches, then extracted quantitative measures of glandular aberrance to train a SVM classifier for normal vs tumor classification and for normal vs low grade vs high grade classification. Glandular aberrance correlated with tumor grade, and this method achieved an accuracy of 97% and 91% for the two-class and three-class classi-fications, respectively. Thus, application of segmentation outputs in this manner can allow for the definition and extraction of novel quantitative features to aid in classi-fication tasks and may provide a look into how these segmentation algorithms will be clinically implemented in the future.

Multiplex IHC (mIHC) involves concurrent histological staining of 6 cell markers or more, and Abousamra et al[56] developed an autoencoder-based color deconvolution algorithm to segment these different stains within a WSI. In a follow-up study, Fassler et al[57] utilized this algorithm on mIHC-stained pancreatic ductal adenocarcinoma (PDAC) WSIs to detect and perform spatial analyses on the cell types. Results indicated that CD16+ myeloid cells dominated the immune microenvironment and on average were of closer distance to tumor cells than CD3+, CD4+, CD8+, or CD20+



lymphocyte populations. In contrast to the study by Awan et al[55] that used segmentation outputs to inform a clinical task, Fassler et al[57] targeted a research application. A pipeline to detect all cell types from mIHC-stained WSIs, quantify, and perform special statistics would serve a wide audience of basic and translational researchers, and, in elevating analytical capacities, may stimulate research output.

A popular translational application of segmentation outputs has been in the field of hepatic steatosis quantification, which is important in the assessment of patients with fatty liver disease or to assess donor liver-quality for transplantation. In an earlier study, Lee et al[58] demonstrated correlation of steatosis quantification by image processing methods on WSIs with MRI measurements, pathologist visual scoring, and several clinical parameters, serving to validate the potential of image feature extraction from WSIs for these applications.

Forlano et al[59] took a ML-based approach to quantify the four histological features used in the Nonalcoholic Steatohepatitis Clinical Research Network (NASH CRN) Scoring System, in an effort to automate the process and assess how their computa-tionally extracted, quantitative histological metrics correlate with the semi-quantitative, categorial metrics of the NASH CRN Scoring System. The authors used image processing techniques to segment out and calculate percentages of fat, inflam-mation, ballooning, and collagen proportionate area, then fed the values into a binary logistic regression classifier to predict the presence of NASH. The authors argued that the traditional, semiquantitative approaches are outdated, due to their categorical nature and unavoidable interobserver variability, and demonstrated an AUC of 0.802 for their pipeline’s capacity to predict NASH.

Sun et al[60] took a modified VGG-16 patch-based segmentation approach to quantify macrovesicular steatosis in HE-stained frozen, donor liver biopsies. The network was trained on patches extracted from WSIs with steatosis regions annotated by pathologists. As such, the final portion of their network could be trained against the pathologist-annotated steatosis maps to output pixel-wise steatosis prediction maps from HE patch inputs. Steatosis percent could then be calculated by summing steatosis probabilities from the predictions maps and dividing by total tissue area. Overall, the model had a sensitivity of 71.4% and specificity of 97.3% in predicting samples with over 30% steatosis, which is the threshold used by some centers for donor rejection.

Roy et al[61] trained a network to segment foreground steatosis droplet pixels from background, a network to recognize steatosis droplet boundaries, and a third neural network that took both of those outputs as input to generate the final segmentation map. Their segmentation results allowed for the calculation of steatosis pixel percentage (DSP%) and steatosis droplet count percentage (DSC%). DSC% most strongly correlated with histologically determined macrovesicular steatosis percentage (rho = 0.90, P < 0.001) and total steatosis percentage (rho = 0.90, P < 0.001). DSP% showed the best correlation with MRI fat quantification (rho = 0.85, P < 0.001).

Lastly, Salvi et al[62] gained the capacity to quantify both micro- and macrosteatosis on HE-stained liver WSIs. The algorithm achieved an overall accuracy of 97.27% on the test set for steatosis segmentation and showed the lowest average error of 1.07% when comparing automated steatosis quantification with manual quantification methods.

BEYOND THE PATHOLOGIST—FEATURES INVISIBLE TO THE HUMAN EYE ?While the emulation and automation of pathologist tasks is a clear and exciting application of these methodologies, recent studies have shown that extraction of information that typically requires other sources of data or that are not obvious to the human eye are possible. This section will cover emerging research that utilizes histopathological specimens to extract such information.

Cancer survival and prognosisAlthough pathologists and physicians can estimate cancer patient prognosis, these determinations often require more than microscopically examining a histological slide. For example, the tumor-node-metastasis (TNM) staging system, though informative, can require information like tumor size, typically gathered from CT scans, or nodal and distal metastases status, which is not evident from a single histopathological slide. In other cases, pathologists may perform genetic testing or IHC-staining for further molecular characterization and subtyping of cancers. Recent work shows that these ML- and DL-based methodologies may be able to learn to predict such information from just histopathological samples.



Bychkov et al[63] developed an approach to predict 5-year survival from HE-stained tumor microarrays from colorectal cancer patients. As with Xu et al[28] and Niazi et al[36], the authors took the VGG-16 network pretrained on the ImageNet dataset[27] as a feature extractor. Each WSI’s collection of extracted patch features were then used to train a three-layer 1D LSTM network, since, similar to the RNN approach used by Iizuka et al[26], LSTM networks can take in a sequence of patch feature inputs. The LSTM model in this study was trained to generate a WSI-level 5-year prognosis probability. While the model’s capacity to predict disease-specific survival was not extremely high (AUC = 0.69), it outperformed histological grade (AUC = 0.57) and Visual Risk Score (AUC = 0.58).

Yue et al[64] incorporated an unsupervised patch clustering method to define patch types, trained a VGG-16 network to recognize the patch types, then implemented an SVM classifier to predict 5-year disease-specific survival. For the unsupervised patch clustering, patch features were extracted by a CNN and pooled, dimensionality reduction was performed by principal component analysis, and the k-means clustering was performed in this lower-dimensional feature space to define patch clusters. While the best performing model generated an accuracy and F1 score of 100%, the approach needs to be validated given the small dataset of 75 WSIs. However, this study is another example of how these unsupervised patch clustering methodologies might be effective in determining patch classes.

To generate a more interpretable model for colorectal cancer survival, Kather et al[65] developed a prognostically predictive “deep stromal score” that utilizes outputs from a CNN trained to recognize adipose, background (glass slide), colorectal adenocarcinoma epithelium, debris, lymphocyte, mucus, smooth muscle, normal colon mucosa, and cancer-associated stroma. The authors used their NCT-CRC-HE-100K dataset that contains 100000 image patches covering these nine tissue classes to train and compare several models in classification performance. The top performing VGG-19 model was then applied to a held-out portion of their dataset. When fitting univariate Cox proportional hazard models to each of these 9 classes across the held-out dataset, the authors found that higher activation of five of the nine classes correlated with poor survival, though three were not significant (NS): Adipose [hazard ratio (HR) = 1.150 (NS)]; debris [HR = 5.967 (P = 0.004)]; lymphocytes [HR = 1.226 (NS)]; muscle [HR = 3.761 (P= 0.025)]; stroma [R = 1.154 (NS)] These five class activations were combined to form the prognostic deep stromal score and validated independently on colorectal adenocarcinoma cases from TCGA program. Multivariate analysis should that the deep stromal score was significant as a prognostic metric for overall survival [HR = 1.63 (P = 0.008)], disease-specific survival [HR = 2.29 (P = 0.0004)], and relapse-free survival [HR = 1.92 (P = 0.0004)].

Focusing on Stage III colon cancer patients, Jiang et al[66] used the NCT-CRC-HE-100K dataset generated by Kather et al[65] to train a classifier to determine the proportion of these tissue types in WSIs, then predict prognosis. Like Kather et al[65], the authors tested several networks on the classification task. They identified InceptionResNetV2 as their top performing model, which was utilized to extract proportions of the nine different tissue types from their own colorectal Stage III cancer dataset. The tissue proportions were fed into several ML classifiers, and the Gradient Boosting Decision Tree was identified as the top performer for prognostic predictions. On Stage III colorectal adenocarcinoma cases from TCGA, this top-performing approach correctly allocated patients into high- and low-risk recurrence groups to predict disease-free survival risk by both univariate and multivariate Cox regression analysis [univariate: HR = 4.324 (P = 0.004); multivariate: HR = 10.273 (P = 0.003)]. In addition, this approach also showed the capacity to predict overall survival risk on the TCGA dataset [univariate: HR = 5.766 (P = 0.000); multivariate: HR = 5.033 (P = 0.002)]. These results highlight a potential avenue for more interpretable ML- and DL-based algorithms and also are evidence of the importance of groups like Kather et al[65] making datasets publicly available to help advance the field as a whole.

For prediction of survival after hepatocellular carcinoma resection, Saillard et al[67] compared an weakly approach with and without an additional, supervised attention mechanism. In both approaches, a pre-trained CNN first extracts features from all patches in the WSI. For the weakly supervised approach (CHOWDER), these patch features are fed into the network along with WSI-level survival data to eventually determine the patches most influential to the survival outcome in an iterative learning process. For the approach with additional supervision (SCHOWDER), the weakly supervised mechanism in CHOWDER is further coupled by an attention mechanism that localizes to tumoral slide regions annotated by pathologists to identify the most influential patches. The SCHOWDER and CHOWDER surprisingly generated highly similar c-indices for survival prediction on the discovery set with 0.78 and 0.75,



respectively, further supporting the potential of these weakly supervised methodo-logies to rival more supervised ones. As CHOWDER assigns risk scores to tiles, the authors could also re-extract then visually inspect types of tiles indicated to be most high and low risk. This involved tumor presence, macrotrabecular architectural tumoral pattern, and vascular spaces in the tumor as high-risk, and tumoral and non-tumoral fibrosis and non-tumoral immune cells as low-risk.

Also in a weakly supervised fashion, Wulczyn et al[68] developed a DL system (DLS) to predict disease specific survival across 10 cancer types, including colon and stomach adenocarcinoma, from TCGA using only WSI-level survival data. Assuming the frequency of informative patches for cancer diagnosis on a slide is P, the probability of a randomly sampled patch being uninformative is 1-P. As such, as one samples n patches, the probability of not sampling any informative patches exponen-tially approaches zero as (1-P)n. The authors leverage this property by randomly sampling multiple patches per slide, extracting features from each with CNNs that share weights, and performing average pooling of the extracted features. The outputs can then be fed into a fully-connected layer before the final logistic regression layer. Here, logistic regression is used as the authors found that discretizing time-to-event periods into specific intervals improved performance. In a combined cohort of all 10 cancers, the DLS was significantly associated with disease specific survival [HR = 1.58 (P < 0.0001)] after adjusting for cancer type, stage, age, and sex in multivariable Cox regression analysis. The DLS also outputs a risk score, allowing for stratification of stage II (P = 0025) and stage III (P < 0.001) patients. Finally, as with Saillard et al[67], high- and low-risk patches can be extracted and visualized for qualitative evaluation.

Although more of an intermediate endpoint, cancer metastasis is a specific event that correlates with reduced survival. Takamatsu et al[69] utilized image processing techniques to extract features from cytokeratin IHC-stained endoscopic resection samples and trained a RF ML classifier to predict lymph node metastasis. Their method demonstrated comparable performance to the predictive capacity of conven-tional histological features extracted from HE-stained slides. On the cross-validation approach, the ML achieved an average AUC of 0.822, compared to 0.855 for the conventional method. Although ML performance was not superior, comparable results by these algorithms are accompanied with the additional benefit of reduced interob-server variability. Furthermore, given the fact that the conventionally extracted HE features showed decent predictive power for lymph node metastasis, it will be interesting to see if a predictive classifier can be built off the HE images directly.

Circumventing staining methodologiesIF and IHC methods are often applied for further characterization of samples. However, these methodologies are time-consuming, can be costly, and require an unstained portion of the tissue. This section will thus focus on recent literature that trains using HE-stained images as input with output information that typically requires these additional staining techniques.

With the introduction of targeted molecular therapies for human epidermal growth factor receptor 2 (HER2) in gastric cancer, determining patient HER2 status has become increasingly important[70]. As this process requires IHC staining for HER2, automated extraction of such information from routinely collected HE samples has advantages in time, cost, and consistency. Sharma et al[71] utilized HER2 IHC-stained sections to define HER2 positive and negative regions on HE-stained sections from the same patient. HER2-stained IHC WSIs and HE-stained WSIs were aligned via semi-atuomatic image-registration approaches to define HER2+ and HER2- tumor regions on the HE based off of corresponding IHC positivity. Training on just the HE patches, the authors trained a custom CNN with three sequential convolution and pooling layers to classify between HER2+ tumor (69.6% accuracy), HER2- tumor (58.1% accuracy), and non-tumor patches (82.0%). Although the performance was modest in predicting IHC-defined output labels from HE, this was one of the earliest studies exploring this type of multimodal approach and used a relatively simple network architecture.

IHC can also be applied to metastatic NET samples to identify primary sites of origin. Redemann et al[72] trained the commercially available HALO-AI CNN on HE-stained samples of metastatic NET with known sites of origin to compare the algorithm’s capacity to make such predictions against IHC-based approaches. While the algorithm achieved a worse overall accuracy of 72% compared to 82% for IHC-based diagnosis, the results are promising given the relatively comparable performance to the gold standard IHC approach and the author’s training of an off-the-shelf, commercial algorithm. A comprehensive comparison of classification performance across multiple models may identify one with superior performance.



Govind et al[73] developed a DL-based pipeline to automate gastrointestinal NET grading, which classically involves IHC detection of a Ki-67-positive tumor hotspot region, then manual counting to obtain the percentage of Ki-67-positive tumor cells. The authors trained one model to detect Ki-67 hotpots and calculate a Ki-67 index from those hotspots similar to pathologists’ workflow, and another model that generates Ki-67 index-based heat maps to classify hot-spot-sized tiles in the WSI as background, non-tumor, G1 tumor, or G2 tumor. Importantly, both models used synaptophysin- and Ki-67-double-stained (DS) WSIs as input. As DS WSIs are not common in clinical practice, the authors computationally merged synaptophysin- and Ki-67-single-stained IHC WSIs to generate DS WSIs to be used for this study.

The first model, SKIE, detects synaptophysin-positive, Ki-67-dense hotspots from these DS WSIs and emulates current pathologist workflow by calculating Ki-67 indices that capture proportional Ki-67 tumor positivity within these hotspots. The second model, Deep-SKIE, was trained by extracting hot-spot sized patches, then assigning correct, output labels according to SKIE outputs on those patches. Specifically, the four classes were background (class 0: if the tile has > 70% background pixels), non-tumor (class 1: < 20% synaptophysin stain), tumor grade 1 (class 2: Ki-67 index < 3%) and tumor grade 2 (class 3: 3% < Ki-67 index < 20%). In predicting these hot-spot patch labels, Deep-SKIE trained on DS WSIs exhibited an overall accuracy of 90.98% compared to 84.84% when trained on SS WSIs, indicating that the additional information from multiple markers aids in classification performance.

The most interesting contribution of this paper with respect to this section, however, is the authors’ decision to train a cycle GAN to generate DS WSIs from SS WSI inputs. In brief, GANs attempt to generate synthetic imaging data that is indistinguishable from the real samples. There is typically a ‘discriminator’ module that attempts to correctly distinguish between the GAN-generated synthetic data and the real-world data. The worse your discriminator performs, the better your GAN is at generating synthetic data.

Here, the authors used a cycle GAN to generate synthetic DS WSIs from SS WSIs that are highly similar to the real DS WSIs already in the dataset. This approach has the ability to generate highly informative DS WSIs in the clinic without the required time and costs associated with additional stains. The authors showed that Deep-SKIE when trained on the GAN-generated DS WSIs showed an accuracy of 87.08% in predicting the four patch classes that was still significantly higher than 84.84% when trained on the SS WSIs.

IF staining is the other major staining technique. Unlike IHC, IF comes with an additional disadvantage regarding sample stability. As signal is carried by fluoro-phores, signals are often lost within a week, and samples are thus typically imaged immediately after staining. As such, DL approaches outputting IF-related data is not only informative and cost-efficient but may simplify acquisition of such data.

Burlingame et al[74] developed an experimental protocol allowing for HE and panCK IF staining in the same section of tissue, then trained a conditional GAN to output virtual panCK IF WSIs from HE PDAC WSI inputs. Similar to the cycle GAN used by Govind et al[73], the conditional GAN here depends upon a discriminator attempting to distinguish between real and virtual IF WSIs. As the protocol allows for HE and panCK IF staining on the same tissue, the authors have HE-IF WSI input-output pairs to train the conditional GAN. The virtual IF WSIs showed high similarity to the real IF WSIs in terms of structural similarity metrics, and the authors also present preliminary data on virtual IF generation for alpha-smooth muscle actin, a stromal marker.

This section has covered two categories of methodologies that circumvent the need for staining. The first category involves, directly from HE-stained inputs, the extraction of information that typically necessitates additional staining. These approaches may one day assist pathologists in quicker, cheaper molecular characteriz-ations of patients. The second category involves the generation of synthetic staining outputs directly from HE inputs. These have the potential to complement the first category of these methods in outputting a virtual staining for pathologists to reference and may improve model interpretability. The other exciting avenue for this second category is research. Currently, many HE-stained imaging databases exist. A reliable methodology to generate high-quality, synthetic IF or IHC WSIs can augment these datasets for researchers worldwide. The addition of these new types of WSI data to existing datasets allows for the application of methods not previously utilized on these HE-only datasets. For example, synthetic IHC WSIs allow for segmentation approaches to identify cell types by marker positivity and allow for types of additional characterizations previously impossible with just the original HE WSIs.



Prediction of expression and genomic dataA string of recent studies has begun to explore the capacity of these approaches in extracting genomic or expression data from histopathological samples. As molecular subtypes affect underlying biology, the concept is that morphological shifts occur in the image and can be detected by algorithms.

Mutational panels are commonly used to further subtype cancers in the clinic. A method to detect these mutations directly from HE will allow for rapid subtyping without the need for additional genetic testing. Chen et al[75] developed an Inception-v3 model to detect hepatocellular carcinoma and predict mutational status for CTNNB1, FMN2, TP53, and ZFX4 from frozen sections stained by HE. The mutational status prediction model was trained on the ten most significantly mutated genes in liver cancers. To validate the model, patients were split into mutated and wild-type groups for all of the ten genes, then the mutation prediction probabilities were examined across these cohorts. The four mentioned genes showed significant differences between the mutated and wild-type cohorts, indicating the model’s ability to predict these mutations. As with the cancer classification tasks, mutational subtyping from HE has a clear path to clinical integration, and these results are encouraging in supporting the value of imaging-based molecular phenotypes.

The consensus molecular subtypes (CMS) transcriptionally distinguish four groups of colorectal cancer with different clinical behaviors and biology, so Sirinukunwattana et al[76] trained a model to designate image-based CMS (imCMS) classes to HE-stained slides. The authors used two resection cohorts (TCGA, FOCUS) and one biopsy cohort (GRAMPIAN), and patches were extracted from WSI regions annotated by pathologists to be tumor. The FOCUS dataset was used for training, as their associated transcriptional data could be utilized to provide CMS labels to the extracted tumoral patches. ImCMS predictions were thus CMS predictions made on a patch-level from histology, and WSI-level subtypes were assigned according to the most prevalent patch subtype prediction. On external validation, the trained model achieved a macro-average classification AUC ranging from 0.80 to 0.83 across the TCGA and GRAMPIAN datasets.

Interestingly, to improve model generalizability, the authors implemented domain-adversarial training. Similar to the GANs mentioned earlier, the goal is to reduce the discriminative ability of this domain-adversarial module in determining whether the input data came from the TCGA, FOCUS, or GRAMPIAN datasets. In practice, the network learns to identify input features important to determine imCMS, while lessening the importance of features that simply vary based off of dataset origin. This improved macro average AUCs in TCGA to 0.84 and in GRAMPIAN to 0.85.

This study by Sirinukunwattana et al[76] offered two additional novelties in CMS classification. First, patches with high prediction confidence for imCMS subtypes could be extracted to examine histological patterns. imCMS1 was associated with mucinous differentiation and lymphocytic infiltration, imCMS2 with cribriform growth patterns and comedo-like necrosis, imCMS3 with ectatic, mucin-filled glandular structures, and imCMS4 with prominent desmoplasia. Secondly, samples with tumoral heterogeneity are currently considered unclassifiable by CMS. The authors here compared agreement between the second most prominent CMS, determined transcriptionally, and the second most prominent imCMS, predicted through this pipeline. The authors noted a high degree of significant, cosine similarity between all four CMS-imCMS pairs. This ability to identify imCMS tumor hetero-geneity may improve colorectal cancer classifications and is a nice example of how current molecular subtyping approaches may be augmented by improved spatial granularity.

Microsatellite instability (MSI) is another prognostic indicator in colorectal cancers that can be diagnosed via genetic analyses. Kather et al[77] thus trained a classifier to recognize MSI and microsatellite stability (MSS) from HE-stained TCGA slides. The approach involved training an initial ResNet-18 model to recognize tumor vs normal to eventually extract only tumor-containing patches from the WSIs. Patches were then assigned MSI or MSS labels based on the patient’s TCGA-recorded MSI status or as MSI-positive if patients have an unknown status but a mutation count over 1000. This labeled data was then used to train another ResNet-18 model to predict MSI and MSS. In external validation on colorectal cases, the model exhibited a patient-level AUC of 0.84 (95%CI: 0.72-0.92). Interestingly, the gastric MSI-detection model achieved a lower AUC of 0.69 (95%CI: 0.52-0.82) on a Japanese cohort, likely reflective of the TCGA stomach adenocarcinoma training cohort being composed of 80% non-Asians and indicates the necessity of multi-center training data for more generalizable models. Finally, patient MSI-levels could be correlated with transcriptomic data. Higher MSI-levels correlated with lymphocyte gene expression in gastric cancer and with PD-L1



expression and interferon- γ signal in colorectal cancer.This model was further refined by Kather et al[78] in a follow-up study. Specifically,

the authors trained the ShuffleNet network, a more lightweight architecture that performed comparably to the more complex ones. After validating on the capacity to classify MSI in colorectal cancer, the model was trained on other tasks and found to be able detect the mutation of at least one clinically actionable mutation in 13 of 14 cancers from TCGA.

To take the next step of validation, Echle et al[79] applied this refined model to a multicenter dataset across Europe. This required the authors to form the MSIDETECT consortium and led to the generation of an 8000-patient dataset with molecular alterations. The algorithm was trained using data from multiple sources, including the TCGA, a German, a United Kingdom, and a Netherlands cohort. The study achieved an impressive AUC of 0.96 for detecting MSI in a large, international validation cohort and exemplifies the importance of multi-source training data. Compared to the study from[77], where authors showed an inability of the model trained on an 80% non-Asian TCGA dataset to perform well on a Japanese dataset, this study improved model generalizability by incorporating a more global patient distribution during training.

While the above studies targeted a specific molecular phenotype in MSI, Schmauch et al[80] set to explore whether general RNA-sequencing expression could be inferred from HE-stained tumor WSIs. The authors extracted up to 8000 tissue-containing patches per WSI from all TCGA cancer types and utilized an ImageNet-pre-trained ResNet-50 to extract 2048-length feature vectors from each. By utilizing k-means clustering, the authors generated 100 supertiles per WSI from these patches on the basis of location. The supertile representation consisted of averaged values for all contained patches over the 2048 ResNet-50 features. For each slide, the feature-extracted supertiles could be fed into another network where the output classification layer contains nodes corresponding to every gene. Thus, the inputs represent supertile feature vectors for every supertile in a WSI, and the network deconvolutes TCGA patient-level transcriptional levels across the WSI supertiles. The network is thus able to detect relationships between supertile features and WSI expression levels to identify the supertile-features most important for certain gene expressions. This is then translated into gene expression outputs at the supertile-level, which could be aggregated to generate gene expression heatmaps at the WSI-level.

Results were validated by comparing predicted expression of CD3 and CD20 with actual IHC-stained sections. For both markers, expression predicted by the model highly correlated with the percentage of cells positively stained in the IHC sections (P < 0.0001 for both). Furthermore, this predictive ability for expression data allowed the authors to implement lists of genes involved in major cancer pathways, including angiogenesis, hypoxia, deregulation of DNA repair, cell-cycling, B-cell responses, and T-cell responses for Gene Set Enrichment Analysis. The authors were able to assess the activation of these pathways across a wide range of cancers.

Finally, as MSI should be defined in part by some changes in expression levels, the authors tested whether the transcriptomic representations learned in their approach can improve the MSI detection performance relative to Kather et al[77]. When applied to HE-stained TCGA colorectal cancer slides, the model achieved a superior AUC of 0.81 compared to 0.68 for the method by Kather et al[77] in predicting MSI. The takeaway message is that, since their model can generate transcriptomic represent-ations to the level of predicting supertile-level, gene-specific expression data, using this model as a feature extractor will generate sets of feature inputs superior in the subsequent MSI vs MSS classification compared to simply inputting the patches themselves. A method to detect expression levels with a patch-level resolution can impact not only these sorts of expression-based molecular characterizations, but also serves to provide the research community with a powerful tool to further leverage existing HE WSI datasets.

The above studies all occurred within the last few years and represent a growing application of these approaches. Historically in public datasets like TCGA, WSIs have been underutilized relative to the genomic data. These recent advances demonstrate promise in proving direct connection of imaging features with underlying genomic and expression data. Furthermore, even with the gradual decrease in sequencing costs, finances still inhibit widespread adoption and availability of these technologies. Thus, these sorts of algorithms may eventually alter clinical landscapes by providing access to genomic and expression characterizations worldwide at a fraction of the cost.



CHALLENGES MOVING FORWARDAlthough the field is in an exciting, fast-moving period, several challenges exist moving forward. In terms of clinical integration, the highly discussed “black box” problem still persists. Deep neural networks and the representations they learn in making their predictions lack in interpretability. This issue is magnified in healthcare where discussion of clinical decisions between patient and physician is critical and is therefore a field of highly active research. For imaging-based studies, CAM visualiz-ations like the one utilized by Kiani et al[35] can direct pathologists’ attention the image locations utilized for generating final predictions. In addition, the decision of Kiani et al[35] to output prediction probabilities and confidences for every possible class provided pathologists with more transparency. This allowed for a clinical decision support system that operated as a tool to help inform pathologists in their decisions and may be more easily integrable into the clinic than algorithms that simply generate a prediction. The study importantly identified, however, the detrimental effects on pathologist performance when the classifier outputs an incorrect prediction. Future care must be taken to fully consider the ramifications of incorrect predictions on patient care and to manage liability of pathologists in these situations.

Another general feature of ML- and DL-based algorithms as a whole is their highly specialized nature. As might be evident from this review, models are typically focused on specific pathologies within a specific organ site. For example, there is no current study that attempts to tackle classification of gastric carcinoma, nonspecific duodenitis, and Helicobacter pylori gastritis at once. As algorithms for specific pathologies get adopted into clinical care, the feasibility of this approach will likely be tested. Special care will likely need to be taken for cataloguing a wide range of models per healthcare system, keeping up with library updates and decisions to keep or update code with deprecated support of certain functions, and maintaining constant quality control mechanisms to ensure high model performance. Each of these will be amplified by the addition of more models into a healthcare system.

As many of these studies rely upon sample annotation for certain use cases, models often become highly specialized. However, more generalizable approaches may become necessary. Hosseini et al[81] addressed this issue by establishing an “Atlas of Digital Pathology” that contains around 18000 annotated images of different tissue types across the human body. The dataset contains images across three levels, with the top level addressing the general tissue types and subsequent levels addressing subtypes. An example from top to lowest level would be Epithelium - Simple Epithelium - Simple Squamous Epithelium. Using this Atlas of Digital Pathology, Chan et al[82] then trained a model that can segment out 31 of the tissue types in the database across over more than 10 organ types. The generalizability of the model may be attributed to the non-organ-specific nature of the tissue types in the Atlas of Digital Pathology.

Binder et al[83] developed a gland segmentation algorithm for colon adenocar-cinoma and breast invasive cancer by utilizing stromal masks. Here, this may be due to stroma appearing more similar across breast and colon than the glands themselves. Analogous approaches leveraging shared features across organ sites may thus help for future multi-organ models.

These research fields are highly interdisciplinary, requiring collaboration between the more quantitative computer scientists and the more biological physicians and academics. While the strength of these fields derives from the complementary nature of the two sides’ highly specialized skillsets, efforts should be made to further increase their cohesion. To illustrate a difference between the two sides, physicians and academics may be surprised to find that a large number of high impact computational studies are in the forms of conference papers or open-access online publications. This is in contrast to biomedical conferences typically being restricted to abstracts, posters, and oral presentations, and the emphasis on peer-reviewed journals.

The computer vision challenges, such as the GlaS challenge[46], may represent an avenue to introduce more cohesion. As mentioned earlier, these challenges serve an important role of generating excitement for an application and leads to submissions of high performing models from a multitude of groups. At present, however, endpoints tend to be metrics like F1 score that focus on the computational performance for the task at hand. Introduction of challenges with more biomedical endpoints may invite design and competition of pipelines that incorporate these methodologies for more directly translational tasks. An example of this would be hepatic steatosis quanti-fication. As opposed to having a challenge to only improve lipid droplet segmentation, a challenge that provides an HE dataset and evaluates submissions by ability to translate segmentation outputs into highly accurate steatosis quantification might fast



track the development of methods for specific clinical tasks. Furthermore, these sorts of challenges would require collaboration of computer scientists and physicians during submission, standardizing the concept of interdisciplinary workflows at a more preclinical point.

In a similar vein, Louis et al[84] argue the field of computational pathology faces an important need to “create a culture that considers the computer and computation as being as central to pathology as the microscope”. Integral to this, the authors posit, is the early exposure to computational concepts ideally during medical school. A certain level of computational awareness and literacy on the physicians’ side is integral to perpetuate excitement of these methodologies and for clinical integration. A similar need, however, exists on the computational side. Emerging computational scientists should be provided the opportunity and made aware of the various biomedical applic-ations of their methodologies. This exposure benefits both sides by instilling experience of collaboration at an early stage, recognition of constraints on the other side, and a cultural adoption of the notion that the two sides should be integrated.

Overall, computational pathology is in an exciting time with rapid advancements. The imaging applications have advanced to the point of defining imaging phenotypes and correlating with some clinical variables. As covered in the NCI workshop report by Colen et al[85], the integration of imaging approaches with -omics information will be a powerful strategy to further characterize and direct clinical care but necessitates the definition of imaging standards. Though the workshop focused on radiological phenotypes, the ideas translate similarly to histopathology. Standardization of methods for data analysis, feature extraction, data integration, and data acquisition will likely be important for robust comparison of methodologies for clinical evaluation. These steps will allow for decreased uncertainties when comparing across different clinical sites and provide a confidence in the imaging phenotypes that will be necessary when beginning to correlate with other -omic phenotypes. While some studies in this review excitingly extracted -omic information directly from histopatho-logical slides, future clinical decision support systems will likely still need to aggregate histopathological information with -omic data to a degree to inform users.

Lastly, these ML- and DL-based technologies are unique in their capacity to continuously learn in response to new data. The FDA has recognized this and published a discussion paper soliciting feedback for a potential premarket review approach for these technologies[86]. Specifically, the FDA has proposed a “Total Product Lifecycle Regulatory Approach” that covers not only premarket review and methods for transparent real-world monitoring upon rollout, but also required proposals for any anticipated changes and steps to be taken for model alterations. These changes include retraining based on new data, incorporation of new target demographics based on new data, increasing capability to different input types but with same intended use (being able to take in MRI in addition to CT for a particular diagnosis), or changing intended use (now able to diagnose an additional type of cancer). As indicated by the steps the FDA has already taken, these algorithms will need to be regulated in a unique way that maximizes the capacity for continued improvement.

In summary, these ML- and DL-based imaging methodologies are rapidly expanding and being increasingly applied in the biomedical domains. Even at this point in time, we are seeing studies that are focused on optimizing the computational tasks, on bridging into translational applications, and on integrating these technologies into clinical decision support tools. In addition to the exciting performance and potential over a wide range of topics, this field also represents an opportunity to further bring together computational scientists with their physician and academic counterparts. Future adoption of these technologies into the clinic will likely be accompanied by increased dependence of healthcare systems on computer scientists who can understand and manage the software and will hopefully encourage cultural standardization of these interdisciplinary workflows.

CONCLUSIONThe application ML- and DL-based methodologies on histopathological slides in the context of gastroenterology and hepatology continues to rise. Though still largely preclinical, recent studies have exhibited the exciting performance of these models in classification and segmentation tasks and in the extraction of features unseen to the human eye, such as prognosis, information that typically necessitates additional stains, or genomic and expression data. The field is in a time rife with studies demonstrating



these potential applications, and the FDA has already taken steps to begin considering the adoption of these technologies into healthcare systems. As such, it will be of importance and interest to monitor not only the methodologies themselves, but the considerations necessary in developing clinical tools.

ACKNOWLEDGEMENTSWe would like to thank Dr. Williams J for allowing us to use colorectal cancer and adjacent normal hematoxylin and eosin-stained slides, for which she helped organized IRB-approved collection, in our illustrative Figure 2A. IRB information is as follows: IRB Net #: 245765, CORIHS#: 2014-2821, Title: The Mechanisms of Intestinal Tumori-genesis, PI Name: Yang VW. We would like to thank Dr. Orzechowska E as well for providing the mouse organoid figures used in our illustrative Figure 2C.

REFERENCESRavi D, Wong C, Deligianni F, Berthelot M, Andreu-Perez J, Lo B, Yang GZ. Deep Learning for Health Informatics. IEEE J Biomed Health Inform 2017; 21: 4-21 [PMID: 28055930 DOI: 10.1109/JBHI.2016.2636665]

1

Cision PR Newswire. AI Medical Service Inc. announces FDA Breakthrough Device Designation for endoscopic AI system. [cited 22 January 2021]. In: Cision PR Newswire [Internet]. Available from: https://www.prnewswire.com/news-releases/ai-medical-service-inc-announces-fda-breakthrough-device-designation-for-endoscopic-ai-system-300953301.html

2

U.S. Food and Drug Administration. Artificial Intelligence and Machine Learning in Software as a Medical Device. [cited 20 December 2020]. In: U.S. Food and Drug Administration [Internet]. Available from: https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-intelligence-and-machine-learning-software-medical-device

3

Kotsiantis SB, Zaharakis ID, Pintelas PE. Machine learning: a review of classification and combining techniques. Artif Intell Rev 2006; 26: 159-190 [DOI: 10.1007/s10462-007-9052-3]

4

Hou L, Samaras D, Kurc TM, Gao Y, Davis JE, Saltz JH. Patch-based Convolutional Neural Network for Whole Slide Tissue Image Classification. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit 2016; 2016: 2424-2433 [PMID: 27795661 DOI: 10.1109/CVPR.2016.266]

5

Ilse M, Tomczak J, Welling M. Attention-based deep multiple instance learning. International conference on machine learning. Proceedings of Machine Learning Research; 2018

6

Xu Y, Mo T, Feng Q, Zhong P, Lai M, Eric I, Chang C. Deep learning of feature representation with multiple instance learning for medical image analysis. IEEE 2014; 1626-1630 [DOI: 10.1109/ICASSP.2014.6853873]

7

Sun C, Xu A, Liu D, Xiong Z, Zhao F, Ding W. Deep Learning-Based Classification of Liver Cancer Histopathology Images Using Only Global Labels. IEEE J Biomed Health Inform 2020; 24: 1643-1651 [PMID: 31670686 DOI: 10.1109/JBHI.2019.2949837]

8

Shrestha A, Mahmood A. Review of deep learning algorithms and architectures. IEEE Access 2019; 7: 53040-53065 [DOI: 10.1109/ACCESS.2019.2912200]

9

Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. 2014 Preprint. Available from: arXiv:14091556

10

Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A. Going deeper with convolutions. IEEE CVPR 2015; 1-9 [DOI: 10.1109/CVPR.2015.7298594]

11

He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. 2015 Preprint. Available from: arXiv:1512.03385

12

Devlin J, Chang MW, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding. 2018 Preprint. Available from: arXiv:1810.04805

13

Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. 2015 Preprint. Available from: arXiv:1411.4038

14

Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. In: Navab N, Hornegger J, Wells W, Frangi A. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015. Lecture Notes in Computer Science: Springer, 2015: 234-241

15

Pantanowitz L, Sharma A, Carter AB, Kurc T, Sussman A, Saltz J. Twenty Years of Digital Pathology: An Overview of the Road Travelled, What is on the Horizon, and the Emergence of Vendor-Neutral Archives. J Pathol Inform 2018; 9: 40 [PMID: 30607307 DOI: 10.4103/jpi.jpi_69_18]

16

U.S. Food and Drug Administration. 510(k) Substantial Equivalence Determination Summary for Phillips Intellisite Pathology Solution (PIPS). [cited 22 January 2021]. In: U.S. Food and Drug Administration [Internet]. Available from: https://www.accessdata.fda.gov/cdrh_docs/reviews/K172174.pdf.

17

U.S. Food and Drug Administration. 510(k) Substantial Equivalence Determination Summary for 18



https://www.prnewswire.com/news-releases/ai-medical-service-inc-announces-fda-breakthrough-device-designation-for-endoscopic-ai-system-300953301.html

https://www.prnewswire.com/news-releases/ai-medical-service-inc-announces-fda-breakthrough-device-designation-for-endoscopic-ai-system-300953301.html

https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-intelligence-and-machine-learning-software-medical-device

https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-intelligence-and-machine-learning-software-medical-device

https://dx.doi.org/10.1007/s10462-007-9052-3


https://dx.doi.org/10.1109/CVPR.2016.266

https://dx.doi.org/10.1109/ICASSP.2014.6853873



https://dx.doi.org/10.1109/ACCESS.2019.2912200

https://dx.doi.org/10.1109/CVPR.2015.7298594


https://dx.doi.org/10.4103/jpi.jpi_69_18

https://www.accessdata.fda.gov/cdrh_docs/reviews/K172174.pdf



Sectra Digital Pathology Module. [cited 22 January 2021]. In: U.S. Food and Drug Administration [Internet]. Available from: https://www.accessdata.fda.gov/cdrh_docs/reviews/K193054.pdf. Kothari S, Phan JH, Stokes TH, Wang MD. Pathology imaging informatics for quantitative analysis of whole-slide images. J Am Med Inform Assoc 2013; 20: 1099-1108 [PMID: 23959844 DOI: 10.1136/amiajnl-2012-001540]

19

Al-Janabi S, Huisman A, Vink A, Leguit RJ, Offerhaus GJ, ten Kate FJ, van Diest PJ. Whole slide images for primary diagnostics of gastrointestinal tract pathology: a feasibility study. Hum Pathol 2012; 43: 702-707 [PMID: 21937077 DOI: 10.1016/j.humpath.2011.06.017]

20

Litjens G, Kooi T, Bejnordi BE, Setio AAA, Ciompi F, Ghafoorian M, van der Laak JAWM, van Ginneken B, Sánchez CI. A survey on deep learning in medical image analysis. Med Image Anal 2017; 42: 60-88 [PMID: 28778026 DOI: 10.1016/j.media.2017.07.005]

21

Gupta R, Kurc T, Sharma A, Almeida JS, Saltz J. The emergence of pathomics. Curr Pathobiol Rep 2019; 7: 73-84

22

Thakur N, Yoon H, Chong Y. Current Trends of Artificial Intelligence for Colorectal Cancer Pathology Image Analysis: A Systematic Review. Cancers (Basel) 2020; 12 [PMID: 32668721 DOI: 10.3390/cancers12071884]

23

Yoon H, Lee J, Oh JE, Kim HR, Lee S, Chang HJ, Sohn DK. Tumor Identification in Colorectal Histology Images Using a Convolutional Neural Network. J Digit Imaging 2019; 32: 131-140 [PMID: 30066123 DOI: 10.1007/s10278-018-0112-9]

24

Sena P, Fioresi R, Faglioni F, Losi L, Faglioni G, Roncucci L. Deep learning techniques for detecting preneoplastic and neoplastic lesions in human colorectal histological images. Oncol Lett 2019; 18: 6101-6107 [PMID: 31788084 DOI: 10.3892/ol.2019.10928]

25

Iizuka O, Kanavati F, Kato K, Rambeau M, Arihiro K, Tsuneki M. Deep Learning Models for Histopathological Classification of Gastric and Colonic Epithelial Tumours. Sci Rep 2020; 10: 1504 [PMID: 32001752 DOI: 10.1038/s41598-020-58467-9]

26

Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M. Imagenet large scale visual recognition challenge. Int J Comput Vis 2015; 115: 211-252

27

Xu Y, Jia Z, Wang LB, Ai Y, Zhang F, Lai M, Chang EI. Large scale tissue histopathology image classification, segmentation, and visualization via deep convolutional activation features. BMC Bioinformatics 2017; 18: 281 [PMID: 28549410 DOI: 10.1186/s12859-017-1685-x]

28

Sirinukunwattana K, Ahmed Raza SE, Tsang YW, Snead DR, Cree IA, Rajpoot NM. Locality Sensitive Deep Learning for Detection and Classification of Nuclei in Routine Colon Cancer Histology Images. IEEE Trans Med Imaging 2016; 35: 1196-1206 [PMID: 26863654 DOI: 10.1109/TMI.2016.2525803]

29

Korbar B, Olofson AM, Miraflor AP, Nicka CM, Suriawinata MA, Torresani L, Suriawinata AA, Hassanpour S. Looking under the hood: Deep neural network visualization to interpret whole-slide image analysis outcomes for colorectal polyps. IEEE CVPRW 2017; 821-827 [DOI: 10.1109/CVPRW.2017.114]

30

Korbar B, Olofson AM, Miraflor AP, Nicka CM, Suriawinata MA, Torresani L, Suriawinata AA, Hassanpour S. Deep Learning for Classification of Colorectal Polyps on Whole-slide Images. J Pathol Inform 2017; 8: 30 [PMID: 28828201 DOI: 10.4103/jpi.jpi_34_17]

31

Tomita N, Abdollahi B, Wei J, Ren B, Suriawinata A, Hassanpour S. Attention-Based Deep Neural Networks for Detection of Cancerous and Precancerous Esophagus Tissue on Histopathological Slides. JAMA Netw Open 2019; 2: e1914645 [PMID: 31693124 DOI: 10.1001/jamanetworkopen.2019.14645]

32

Sali R, Moradinasab N, Guleria S, Ehsan L, Fernandes P, Shah TU, Syed S, Brown DE. Deep Learning for Whole-Slide Tissue Histopathology Classification: A Comparative Study in the Identification of Dysplastic and Non-Dysplastic Barrett's Esophagus. J Pers Med 2020; 10 [PMID: 32977465 DOI: 10.3390/jpm10040141]

33

Leon F, Gelvez M, Jaimes Z, Gelvez T, Arguello H. Supervised classification of histopathological images using convolutional neuronal networks for gastric cancer detection. IEEE STSIVA 2019; 1-5 [DOI: 10.1109/STSIVA.2019.8730284]

34

Kiani A, Uyumazturk B, Rajpurkar P, Wang A, Gao R, Jones E, Yu Y, Langlotz CP, Ball RL, Montine TJ, Martin BA, Berry GJ, Ozawa MG, Hazard FK, Brown RA, Chen SB, Wood M, Allard LS, Ylagan L, Ng AY, Shen J. Impact of a deep learning assistant on the histopathologic classification of liver cancer. NPJ Digit Med 2020; 3: 23 [PMID: 32140566 DOI: 10.1038/s41746-020-0232-8]

35

Niazi MKK, Tavolara TE, Arole V, Hartman DJ, Pantanowitz L, Gurcan MN. Identifying tumor in pancreatic neuroendocrine neoplasms from Ki67 images using transfer learning. PLoS One 2018; 13: e0195621 [PMID: 29649302 DOI: 10.1371/journal.pone.0195621]

36

Saltz J, Gupta R, Hou L, Kurc T, Singh P, Nguyen V, Samaras D, Shroyer KR, Zhao T, Batiste R, Van Arnam J; Cancer Genome Atlas Research Network, Shmulevich I, Rao AUK, Lazar AJ, Sharma A, Thorsson V. Spatial Organization and Molecular Correlation of Tumor-Infiltrating Lymphocytes Using Deep Learning on Pathology Images. Cell Rep 2018; 23: 181-193. e7 [PMID: 29617659 DOI: 10.1016/j.celrep.2018.03.086]

37

Chang YH, Thibault G, Madin O, Azimi V, Meyers C, Johnson B, Link J, Margolin A, Gray JW. Deep learning based Nucleus Classification in pancreas histological images. Annu Int Conf IEEE Eng Med Biol Soc 2017; 2017: 672-675 [PMID: 29059962 DOI: 10.1109/EMBC.2017.8036914]

38

Shapcott M, Hewitt KJ, Rajpoot N. Deep Learning With Sampling in Colon Cancer Histology. Front Bioeng Biotechnol 2019; 7: 52 [PMID: 30972333 DOI: 10.3389/fbioe.2019.00052]

39

https://www.accessdata.fda.gov/cdrh_docs/reviews/K193054.pdf


https://dx.doi.org/10.1136/amiajnl-2012-001540


https://dx.doi.org/10.1016/j.humpath.2011.06.017


https://dx.doi.org/10.1016/j.media.2017.07.005


https://dx.doi.org/10.3390/cancers12071884


https://dx.doi.org/10.1007/s10278-018-0112-9




https://dx.doi.org/10.1038/s41598-020-58467-9




https://dx.doi.org/10.1109/TMI.2016.2525803

https://dx.doi.org/10.1109/CVPRW.2017.114




https://dx.doi.org/10.1001/jamanetworkopen.2019.14645


https://dx.doi.org/10.3390/jpm10040141

https://dx.doi.org/10.1109/STSIVA.2019.8730284


https://dx.doi.org/10.1038/s41746-020-0232-8




https://dx.doi.org/10.1016/j.celrep.2018.03.086


https://dx.doi.org/10.1109/EMBC.2017.8036914


https://dx.doi.org/10.3389/fbioe.2019.00052



Wei JW, Wei JW, Jackson CR, Ren B, Suriawinata AA, Hassanpour S. Automated Detection of Celiac Disease on Duodenal Biopsy Slides: A Deep Learning Approach. J Pathol Inform 2019; 10: 7 [PMID: 30984467 DOI: 10.4103/jpi.jpi_87_18]

40

Srivastava A, Sengupta S, Kang SJ, Kant K, Khan M, Ali SA, Moore SR, Amadi BC, Kelly P, Syed S. Deep learning for detecting diseases in gastrointestinal biopsy images. IEEE SIEDS 2019; 1-4 [DOI: 10.1109/SIEDS.2019.8735619]

41

Sali R, Ehsan L, Kowsari K, Khan M, Moskaluk CA, Syed S, Brown DE. Celiacnet: Celiac disease severity diagnosis on duodenal histopathological images using deep residual networks. IEEE BIBM 2019; 962-967 [DOI: 10.1109/BIBM47256.2019.8983270]

42

Sali R, Adewole S, Ehsan L, Denson LA, Kelly P, Amadi BC, Holtz L, Ali SA, Moore SR, Syed S. Hierarchical Deep Convolutional Neural Networks for Multi-category Diagnosis of Gastrointestinal Disorders on Histopathological Images. 2020 Preprint. Available from: arXiv:2005.03868

43

Martin DR, Hanson JA, Gullapalli RR, Schultz FA, Sethi A, Clark DP. A Deep Learning Convolutional Neural Network Can Recognize Common Patterns of Injury in Gastric Pathology. Arch Pathol Lab Med 2020; 144: 370-378 [PMID: 31246112 DOI: 10.5858/arpa.2019-0004-OA]

44

Klein S, Gildenblat J, Ihle MA, Merkelbach-Bruse S, Noh KW, Peifer M, Quaas A, Büttner R. Deep learning for sensitive detection of Helicobacter Pylori in gastric biopsies. BMC Gastroenterol 2020; 20: 417 [PMID: 33308189 DOI: 10.1186/s12876-020-01494-7]

45

Sirinukunwattana K, Pluim JPW, Chen H, Qi X, Heng PA, Guo YB, Wang LY, Matuszewski BJ, Bruni E, Sanchez U, Böhm A, Ronneberger O, Cheikh BB, Racoceanu D, Kainz P, Pfeiffer M, Urschler M, Snead DRJ, Rajpoot NM. Gland segmentation in colon histology images: The glas challenge contest. Med Image Anal 2017; 35: 489-502 [PMID: 27614792 DOI: 10.1016/j.media.2016.08.008]

46

Xu Y, Li Y, Liu M, Wang Y, Lai M, Eric I, Chang C. Gland instance segmentation by deep multichannel side supervision. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2016. Lecture Notes in Computer Science: Springer, 2016: 496-504

47

BenTaieb A, Hamarneh G. Topology aware fully convolutional networks for histology gland segmentation. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2016. Lecture Notes in Computer Science: Springer, 2016: 460-468

48

Graham S, Chen H, Gamper J, Dou Q, Heng PA, Snead D, Tsang YW, Rajpoot N. MILD-Net: Minimal information loss dilated network for gland instance segmentation in colon histology images. Med Image Anal 2019; 52: 199-211 [PMID: 30594772 DOI: 10.1016/j.media.2018.12.001]

49

Zhao P, Zhang J, Fang W, Deng S. SCAU-Net: Spatial-Channel Attention U-Net for Gland Segmentation. Front Bioeng Biotechnol 2020; 8: 670 [PMID: 32719781 DOI: 10.3389/fbioe.2020.00670]

50

Xiao QE, Chung PC, Tsai HW, Cheng KS, Chow NH, Juang YZ, Tsai HH, Wang CH, Hsieh TA. Hematoxylin and Eosin (H&E) Stained Liver Portal Area Segmentation Using Multi-Scale Receptive Field Convolutional Neural Network. IEEE J Emerg Sel Top Circuits Syst 2019; 9: 623-634

51

Xu J, Luo X, Wang G, Gilmore H, Madabhushi A. A Deep Convolutional Neural Network for segmenting and classifying epithelial and stromal regions in histopathological images. Neurocomputing 2016; 191: 214-223 [PMID: 28154470 DOI: 10.1016/j.neucom.2016.01.034]

52

Wang X, Fang Y, Yang S, Zhu D, Wang M, Zhang J, Tong KY, Han X. A hybrid network for automatic hepatocellular carcinoma segmentation in H&E-stained whole slide images. Med Image Anal 2021; 68: 101914 [PMID: 33285479 DOI: 10.1016/j.media.2020.101914]

53

Qaiser T, Tsang YW, Taniyama D, Sakamoto N, Nakane K, Epstein D, Rajpoot N. Fast and accurate tumor segmentation of histology images using persistent homology and deep convolutional features. Med Image Anal 2019; 55: 1-14 [PMID: 30991188 DOI: 10.1016/j.media.2019.03.014]

54

Awan R, Sirinukunwattana K, Epstein D, Jefferyes S, Qidwai U, Aftab Z, Mujeeb I, Snead D, Rajpoot N. Glandular Morphometrics for Objective Grading of Colorectal Adenocarcinoma Histology Images. Sci Rep 2017; 7: 16852 [PMID: 29203775 DOI: 10.1038/s41598-017-16516-w]

55

Abousamra S, Fassler D, Hou L, Zhang Y, Gupta R, Kurc T, Escobar-Hoyos LF, Samaras D, Knudson B, Shroyer K. Weakly-Supervised Deep Stain Decomposition for Multiplex IHC Images. IEEE ISBI 2020; 481-485 [DOI: 10.1109/ISBI45749.2020.9098652]

56

Fassler DJ, Abousamra S, Gupta R, Chen C, Zhao M, Paredes D, Batool SA, Knudsen BS, Escobar-Hoyos L, Shroyer KR, Samaras D, Kurc T, Saltz J. Deep learning-based image analysis methods for brightfield-acquired multiplex immunohistochemistry images. Diagn Pathol 2020; 15: 100 [PMID: 32723384 DOI: 10.1186/s13000-020-01003-0]

57

Lee MJ, Bagci P, Kong J, Vos MB, Sharma P, Kalb B, Saltz JH, Martin DR, Adsay NV, Farris AB. Liver steatosis assessment: correlations among pathology, radiology, clinical data and automated image analysis software. Pathol Res Pract 2013; 209: 371-379 [PMID: 23707550 DOI: 10.1016/j.prp.2013.04.001]

58

Forlano R, Mullish BH, Giannakeas N, Maurice JB, Angkathunyakul N, Lloyd J, Tzallas AT, Tsipouras M, Yee M, Thursz MR, Goldin RD, Manousou P. High-Throughput, Machine Learning-Based Quantification of Steatosis, Inflammation, Ballooning, and Fibrosis in Biopsies From Patients With Nonalcoholic Fatty Liver Disease. Clin Gastroenterol Hepatol 2020; 18: 2081-2090. e9 [PMID: 31887451 DOI: 10.1016/j.cgh.2019.12.025]

59

Sun L, Marsh JN, Matlock MK, Chen L, Gaut JP, Brunt EM, Swamidass SJ, Liu TC. Deep learning quantification of percent steatosis in donor liver biopsy frozen sections. EBioMedicine 2020; 60: 103029 [PMID: 32980688 DOI: 10.1016/j.ebiom.2020.103029]

60



https://dx.doi.org/10.1109/SIEDS.2019.8735619

https://dx.doi.org/10.1109/BIBM47256.2019.8983270


https://dx.doi.org/10.5858/arpa.2019-0004-OA


https://dx.doi.org/10.1186/s12876-020-01494-7






https://dx.doi.org/10.3389/fbioe.2020.00670


https://dx.doi.org/10.1016/j.neucom.2016.01.034






https://dx.doi.org/10.1038/s41598-017-16516-w

https://dx.doi.org/10.1109/ISBI45749.2020.9098652


https://dx.doi.org/10.1186/s13000-020-01003-0


https://dx.doi.org/10.1016/j.prp.2013.04.001




https://dx.doi.org/10.1016/j.ebiom.2020.103029



Roy M, Wang F, Vo H, Teng D, Teodoro G, Farris AB, Castillo-Leon E, Vos MB, Kong J. Deep-learning-based accurate hepatic steatosis quantification for histological assessment of liver biopsies. Lab Invest 2020; 100: 1367-1383 [PMID: 32661341 DOI: 10.1038/s41374-020-0463-y]

61

Salvi M, Molinaro L, Metovic J, Patrono D, Romagnoli R, Papotti M, Molinari F. Fully automated quantitative assessment of hepatic steatosis in liver transplants. Comput Biol Med 2020; 123: 103836 [PMID: 32658781 DOI: 10.1016/j.compbiomed.2020.103836]

62

Bychkov D, Linder N, Turkki R, Nordling S, Kovanen PE, Verrill C, Walliander M, Lundin M, Haglund C, Lundin J. Deep learning based tissue analysis predicts outcome in colorectal cancer. Sci Rep 2018; 8: 3395 [PMID: 29467373 DOI: 10.1038/s41598-018-21758-3]

63

Yue X, Dimitriou N, Arandjelovic O. Colorectal cancer outcome prediction from H&E whole slide images using machine learning and automatically inferred phenotype profiles. 2019 Preprint. Available from: arXiv:1902.03582

64

Kather JN, Krisam J, Charoentong P, Luedde T, Herpel E, Weis CA, Gaiser T, Marx A, Valous NA, Ferber D, Jansen L, Reyes-Aldasoro CC, Zörnig I, Jäger D, Brenner H, Chang-Claude J, Hoffmeister M, Halama N. Predicting survival from colorectal cancer histology slides using deep learning: A retrospective multicenter study. PLoS Med 2019; 16: e1002730 [PMID: 30677016 DOI: 10.1371/journal.pmed.1002730]

65

Jiang D, Liao J, Duan H, Wu Q, Owen G, Shu C, Chen L, He Y, Wu Z, He D, Zhang W, Wang Z. A machine learning-based prognostic predictor for stage III colon cancer. Sci Rep 2020; 10: 10333 [PMID: 32587295 DOI: 10.1038/s41598-020-67178-0]

66

Saillard C, Schmauch B, Laifa O, Moarii M, Toldo S, Zaslavskiy M, Pronier E, Laurent A, Amaddeo G, Regnault H, Sommacale D, Ziol M, Pawlotsky JM, Mulé S, Luciani A, Wainrib G, Clozel T, Courtiol P, Calderaro J. Predicting Survival After Hepatocellular Carcinoma Resection Using Deep Learning on Histological Slides. Hepatology 2020; 72: 2000-2013 [PMID: 32108950 DOI: 10.1002/hep.31207]

67

Wulczyn E, Steiner DF, Xu Z, Sadhwani A, Wang H, Flament-Auvigne I, Mermel CH, Chen PC, Liu Y, Stumpe MC. Deep learning-based survival prediction for multiple cancer types using histopathology images. PLoS One 2020; 15: e0233678 [PMID: 32555646 DOI: 10.1371/journal.pone.0233678]

68

Takamatsu M, Yamamoto N, Kawachi H, Chino A, Saito S, Ueno M, Ishikawa Y, Takazawa Y, Takeuchi K. Prediction of early colorectal cancer metastasis by machine learning using digital slide images. Comput Methods Programs Biomed 2019; 178: 155-161 [PMID: 31416544 DOI: 10.1016/j.cmpb.2019.06.022]

69

Abrahao-Machado LF, Scapulatempo-Neto C. HER2 testing in gastric cancer: An update. World J Gastroenterol 2016; 22: 4619-4625 [PMID: 27217694 DOI: 10.3748/wjg.v22.i19.4619]

70

Sharma H, Zerbe N, Klempert I, Hellwich O, Hufnagl P. Deep convolutional neural networks for automatic classification of gastric carcinoma using whole slide images in digital histopathology. Comput Med Imaging Graph 2017; 61: 2-13 [PMID: 28676295 DOI: 10.1016/j.compmedimag.2017.06.001]

71

Redemann J, Schultz FA, Martinez C, Harrell M, Clark DP, Martin DR, Hanson JA. Comparing Deep Learning and Immunohistochemistry in Determining the Site of Origin for Well-Differentiated Neuroendocrine Tumors. J Pathol Inform 2020; 11: 32 [PMID: 33343993 DOI: 10.4103/jpi.jpi_37_20]

72

Govind D, Jen KY, Matsukuma K, Gao G, Olson KA, Gui D, Wilding GE, Border SP, Sarder P. Improving the accuracy of gastrointestinal neuroendocrine tumor grading with deep learning. Sci Rep 2020; 10: 11064 [PMID: 32632119 DOI: 10.1038/s41598-020-67880-z]

73

Burlingame EA, McDonnell M, Schau GF, Thibault G, Lanciault C, Morgan T, Johnson BE, Corless C, Gray JW, Chang YH. SHIFT: speedy histological-to-immunofluorescent translation of a tumor signature enabled by deep learning. Sci Rep 2020; 10: 17507 [PMID: 33060677 DOI: 10.1038/s41598-020-74500-3]

74

Chen M, Zhang B, Topatana W, Cao J, Zhu H, Juengpanich S, Mao Q, Yu H, Cai X. Classification and mutation prediction based on histopathology H&E images in liver cancer using deep learning. NPJ Precis Oncol 2020; 4: 14 [PMID: 32550270 DOI: 10.1038/s41698-020-0120-3]

75

Sirinukunwattana K, Domingo E, Richman SD, Redmond KL, Blake A, Verrill C, Leedham SJ, Chatzipli A, Hardy C, Whalley CM, Wu CH, Beggs AD, McDermott U, Dunne PD, Meade A, Walker SM, Murray GI, Samuel L, Seymour M, Tomlinson I, Quirke P, Maughan T, Rittscher J, Koelzer VH; S:CORT consortium. Image-based consensus molecular subtype (imCMS) classification of colorectal cancer using deep learning. Gut 2021; 70: 544-554 [PMID: 32690604 DOI: 10.1136/gutjnl-2019-319866]

76

Kather JN, Pearson AT, Halama N, Jäger D, Krause J, Loosen SH, Marx A, Boor P, Tacke F, Neumann UP, Grabsch HI, Yoshikawa T, Brenner H, Chang-Claude J, Hoffmeister M, Trautwein C, Luedde T. Deep learning can predict microsatellite instability directly from histology in gastrointestinal cancer. Nat Med 2019; 25: 1054-1056 [PMID: 31160815 DOI: 10.1038/s41591-019-0462-y]

77

Kather JN, Heij LR, Grabsch HI, Loeffler C, Echle A, Muti HS, Krause J, Niehues JM, Sommer KAJ, Bankhead P, Kooreman LFS, Schulte JJ, Cipriani NA, Buelow RD, Boor P, Ortiz-Brüchle NN, Hanby AM, Speirs V, Kochanny S, Patnaik A, Srisuwananukorn A, Brenner H, Hoffmeister M, van den Brandt PA, Jäger D, Trautwein C, Pearson AT, Luedde T. Pan-cancer image-based detection of clinically actionable genetic alterations. Nat Cancer 2020; 1: 789-799 [PMID: 33763651 DOI: 10.1038/s43018-020-0087-6]

78






https://dx.doi.org/10.1038/s41598-018-21758-3


https://dx.doi.org/10.1371/journal.pmed.1002730


https://dx.doi.org/10.1038/s41598-020-67178-0






https://dx.doi.org/10.1016/j.cmpb.2019.06.022




https://dx.doi.org/10.1016/j.compmedimag.2017.06.001






https://dx.doi.org/10.1038/s41598-020-74500-3


https://dx.doi.org/10.1038/s41698-020-0120-3






https://dx.doi.org/10.1038/s43018-020-0087-6



Echle A, Grabsch HI, Quirke P, van den Brandt PA, West NP, Hutchins GGA, Heij LR, Tan X, Richman SD, Krause J, Alwers E, Jenniskens J, Offermans K, Gray R, Brenner H, Chang-Claude J, Trautwein C, Pearson AT, Boor P, Luedde T, Gaisa NT, Hoffmeister M, Kather JN. Clinical-Grade Detection of Microsatellite Instability in Colorectal Tumors by Deep Learning. Gastroenterology 2020; 159: 1406-1416. e11 [PMID: 32562722 DOI: 10.1053/j.gastro.2020.06.021]

79

Schmauch B, Romagnoni A, Pronier E, Saillard C, Maillé P, Calderaro J, Kamoun A, Sefta M, Toldo S, Zaslavskiy M, Clozel T, Moarii M, Courtiol P, Wainrib G. A deep learning model to predict RNA-Seq expression of tumours from whole slide images. Nat Commun 2020; 11: 3877 [PMID: 32747659 DOI: 10.1038/s41467-020-17678-4]

80

Hosseini MS, Chan L, Tse G, Tang M, Deng J, Norouzi S, Rowsell C, Plataniotis KN, Damaskinos S. Atlas of digital pathology: A generalized hierarchical histological tissue type-annotated database for deep learning. IEEE CVPR 2019; 11747-11756

81

Chan L, Hosseini MS, Rowsell C, Plataniotis KN, Damaskinos S. Histosegnet: Semantic segmentation of histological tissue type in whole slide images. IEEE ICCV 2019; 10661-10670 [DOI: 10.1109/ICCV.2019.01076]

82

Binder T, Tantaoui EM, Pati P, Catena R, Set-Aghayan A, Gabrani M. Multi-Organ Gland Segmentation Using Deep Learning. Front Med (Lausanne) 2019; 6: 173 [PMID: 31428614 DOI: 10.3389/fmed.2019.00173]

83

Louis DN, Feldman M, Carter AB, Dighe AS, Pfeifer JD, Bry L, Almeida JS, Saltz J, Braun J, Tomaszewski JE, Gilbertson JR, Sinard JH, Gerber GK, Galli SJ, Golden JA, Becich MJ. Computational Pathology: A Path Ahead. Arch Pathol Lab Med 2016; 140: 41-50 [PMID: 26098131 DOI: 10.5858/arpa.2015-0093-SA]

84

Colen R, Foster I, Gatenby R, Giger ME, Gillies R, Gutman D, Heller M, Jain R, Madabhushi A, Madhavan S, Napel S, Rao A, Saltz J, Tatum J, Verhaak R, Whitman G. NCI Workshop Report: Clinical and Computational Requirements for Correlating Imaging Phenotypes with Genomics Signatures. Transl Oncol 2014; 7: 556-569 [PMID: 25389451 DOI: 10.1016/j.tranon.2014.07.007]

85

U.S. Food and Drug Administration. Proposed Regulatory Framework for Modifications to Artificial Intelligence/Machine Learning (AI/ML)-Based Software as a Medical Device (SaMD). [cited 22 January 2021]. In: U.S. Food and Drug Administration [Internet]. Available from: https://www.fda.gov/media/122535/download

86




https://dx.doi.org/10.1038/s41467-020-17678-4

https://dx.doi.org/10.1109/ICCV.2019.01076


https://dx.doi.org/10.3389/fmed.2019.00173


https://dx.doi.org/10.5858/arpa.2015-0093-SA


https://dx.doi.org/10.1016/j.tranon.2014.07.007

https://www.fda.gov/media/122535/download


World Journal of



MINIREVIEWS

COVID-19 in normal, diseased and transplanted liver

Alessandro Signorello, Ilaria Lenci, Martina Milana, Giuseppe Grassi, Leonardo Baiocchi

ORCID number: Alessandro Signorello 0000-0002-3831-7244; Ilaria Lenci 0000-0001-5704-9890; Martina Milana 0000-0003-2027-0481; Giuseppe Grassi 0000-0001-9182-8759; Leonardo Baiocchi 0000-0003-3672-4505.

Author contributions: Signorello A performed acquisition of data, analysis and interpretation, drafting of manuscript, critical revision; Lenci I, Milana M, and Grassi G performed acquisition of data, critical revision; Baiocchi L performed proposal of study, study conception, correction of manuscript, critical revision.

Conflict-of-interest statement: No conflict of interest to disclose.



Alessandro Signorello, Ilaria Lenci, Martina Milana, Giuseppe Grassi, Leonardo Baiocchi, Hepatology Unit, Department of Medicine, University of Tor Vergata, Rome 00133, Italy

Corresponding author: Leonardo Baiocchi, MD, PhD, Associate Professor, Hepatology Unit, Department of Medicine, University of Tor Vergata, Viale Oxford 81, Rome 00133, Italy. [email protected]

AbstractStarting from December 2019 the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has extended in the entire world giving origin to a pandemic. Although the respiratory system is the main apparatus involved by the infection, several other organs may suffer coronavirus disease 2019 (COVID-19)-related injuries. The human tissues expressing angiotensin-converting enzyme 2 (ACE2) are all possible targets of viral damage. In fact myocarditis, meningo-encephalitis, acute kidney injury and other complications have been described with regard to SARS-CoV-2 infection. The liver has a central role in the body homeostasis contributing to detoxification, catabolism and also synthesis of important factor such as plasma proteins. ACE2 is significantly expressed just by cholangiocytes within the liver, however transaminases are increased in more than one third of COVID-19 patients, at hospital admission. The reasons for liver impairment in the course of this infection are not completely clear at present and multiple factors such as: Direct viral effect, release of cytokines, ischemic damage, use of hepatotoxic drugs, sepsis, and others, may contribute to damage. While COVID-19 seems to elicit just a transient alteration of liver function tests in subjects with normal hepatic function, of concern, more severe sequelae are frequently observed in patients with a reduced hepatic reserve. In this review we report data regarding SARS-CoV-2 infection in subjects with normal or diseased liver. In addition the risks of COVID-19 in immunosuppressed patients (either transplanted or suffering for autoimmune liver diseases) are also described.

Key Words: COVID-19; Liver; Non-alcoholic fatty-liver-disease; Cirrhosis; Liver transplant; Angiotensin-converting enzyme 2


Core Tip: Severe acute respiratory syndrome coronavirus 2 infection, starting from December 2019 in China, has now extended in the whole world. While the respiratory



http://orcid.org/0000-0002-3831-7244

http://orcid.org/0000-0002-3831-7244

http://orcid.org/0000-0001-5704-9890

http://orcid.org/0000-0001-5704-9890

http://orcid.org/0000-0003-2027-0481

http://orcid.org/0000-0003-2027-0481

http://orcid.org/0000-0001-9182-8759

http://orcid.org/0000-0001-9182-8759

http://orcid.org/0000-0001-9182-8759

http://orcid.org/0000-0003-3672-4505

http://orcid.org/0000-0003-3672-4505

http://orcid.org/0000-0003-3672-4505





Signorello A et al. COVID-19 and liver



Country/Territory of origin: Italy

Peer-review report’s scientific quality classificationGrade A (Excellent): 0 Grade B (Very good): B, B Grade C (Good): C, C Grade D (Fair): 0 Grade E (Poor): 0

Received: February 18, 2021 Peer-review started: February 18, 2021 First decision: March 12, 2021 Revised: March 18, 2021 Accepted: May 7, 2021 Article in press: May 7, 2021 Published online: May 28, 2021

P-Reviewer: Bain V, Sempokuya T S-Editor: Fan JR L-Editor: A P-Editor: Liu JH

system is mainly involved in the infection other organs may be impaired by coronavirus disease 2019 (COVID-19). In this review we report the current finding regarding the liver during this infection. While mild liver effects occur in normal subjects with COVID-19, severe sequelae may be expected in individuals with a reduced hepatic reserve.

Citation: Signorello A, Lenci I, Milana M, Grassi G, Baiocchi L. COVID-19 in normal, diseased and transplanted liver. World J Gastroenterol 2021; 27(20): 2576-2585URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2576.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2576

INTRODUCTIONBetween December 2019 and January 2020, first China and then the rest of the world observed the emergence of a new pathogen, responsible for the recent pandemic still plaguing our planet. On January 5, 2020, the WHO issued a global alert after 44 patients, hospitalized for pneumonia of unknown origin, were reported by Chinese national authorities[1]. On March 11, 2020, the WHO declared the new coronavirus, identified as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and responsible for the so called coronavirus disease 2019 (COVID-19)[2], to be pandemic. To date (January 2021), the SARS-CoV-2 infection is approaching 100 million cases worldwide, while 2 million related deaths have been reported globally. SARS-CoV-2 is a virus belonging to the genus Betacoronavirus, family Coronaviridae, subgenus Sarbecovirus, discovered at the end of 2019. It is a single-stranded, positive-sense RNA virus with a 50-200 nm diameter. Since it has close genetic similarity with bat coronaviruses, a zoonotic origin has been suggested, based on the transmission from animal to humans. SARS-CoV-2 possesses four structural proteins called the S (spike), E (envelope), M (membrane), and N (nucleocapsid) proteins; the N protein binds the virus genome while the S, E, and M proteins contribute in the assembly of the viral envelope[3].

Impairment of respiratory function is the most frequent clinical concern during infection by this new virus; however, other physiological organ systems can be affected. Among these, SARS-CoV-2-induced hepatic changes are gaining interest for their possible relationship with liver and patient outcome.

The liver plays a fundamental role in metabolism and the synthesis of plasma proteins, as well as in the manipulation and detoxification of diet xenobiotics[4]. Several conditions may lead to chronic hepatitis and cirrhosis or hepato-carcinoma, including viral agents, alcohol abuse, metabolic derangements, and others[5]. The patient affected by liver disease is fragile, and his clinical stability is always in perennial equilibrium. Among the direct causes of hepatic decompensation, the detrimental effect of infections, both viral and bacterial, is well known, and in this particular setting, COVID-19 does not seem less important[6]. This review summarizes the effects of SARS-CoV-2 infection in patients with liver disease, as well as in healthy and liver transplanted subjects. Data available on clinical complications and outcomes are also reported.

INTERACTION OF SARS-COV-2 WITH TISSUESThe SARS-CoV-2 molecular pathways employed by the virus to favor host interaction and invasion are of major interest. Several studies have been conducted on this issue, and the majority of evidence converges to identify the angiotensin-converting enzyme 2 (ACE2), a surface peptidase responsible for the regulation of blood pressure, as a key factor. In fact, Zhou et al[7] demonstrated that ACE2 is the main entry route of SARS-CoV-2 within the cell, as also previously observed for the SARS-CoV and HCoV-NL6 viruses[7]. During the previous SARS pandemic in 2003, it was demonstrated that the use of antibodies against ACE2 effectively counteracted viral replication. On the other hand, ACE1 antibodies did not have any effect[8]. The interaction between this new virus and ACE2 has also been investigated in a study conducted by Xu et al[9], where the strong link between the virus and this receptor was confirmed. ACE2 is present in





different organs and tissues, such as alveolar type 2 cells of the lung; epithelial cells of the nasal, nasopharyngeal, and oral mucosa; in the smooth muscles of the gastric and colonic mucosa; in enterocytes of the duodenum, jejunum, ileum, and colon; in myocardiocytes; in the cells of the proximal tubule of the kidney; and in cholan-giocytes within the liver[10]. From all of the above, it is clear that any tissue expressing ACE2 might be a possible target for SARS-CoV-2, giving rise to different symptoms with a severity that may be a function of the baseline organ reserve. The main organs expressing ACE2 with the corresponding symptoms/complications COVID-19-related, are reported in Figure 1. Although the main clinical expression is represented by a flu-like mild syndrome, a severe bilateral interstitial pneumonia resulting in acute respiratory distress syndrome (ARDS) and requiring intensive care unit (ICU) management is not rare[11]. Other common clinical manifestations comprise: Nausea, vomiting, anorexia, diarrhea (2%-10% of patients), and altered liver function tests[12].

COVID-19 AND THE NORMAL LIVEROne large retrospective study, conducted in the city of New York, enrolled approx-imately 5700 COVID-19 patients. Liver transaminase levels were assessed at hospital admission. Increased aspartate transaminase (AST) or alanine transaminase (ALT) was found in 58% and 39% of patients, respectively[13]. Disease worsening during in-hospital stay was associated with a further increase of these enzymes. However, the pathogenetic mechanism leading to these alterations remained unclear. In fact, several factors may all have contributed to the liver changes in this heterogeneous group of patients, such as: (1) a direct viral cytopathic effect; (2) the activation of cytokine cascade; (3) the onset of multi-organ failure; (4) a state of disseminated intravascular coagulation; and (5) the use of potentially hepatotoxic drugs (such as remdesivir), among others[12]. As demonstrated by other research, in the course of COVID-19, the increase in liver enzymes is characterized by a preferential AST elevation, thus recalling the picture observed in alcoholic, metabolic, or ischemic liver con-ditions[14-16]. Abnormal levels of cholestasis markers are seldom observed[17], even if an increased gamma-glutamyltransferase (GGT) level was reported in 50% of patients in one study[18].

Even though SARS-CoV-2 viral particles were described in hepatocytes in two severe cases[19] a direct cytopathic effect does not seem to explain the liver changes since ACE2 is mainly expressed in the biliary tract, within the liver. On the other hand, despite: (1) the presence of the receptor on cholagiocytes; and (2) “in vitro” evidences demonstrating tight junction damage and bile acids transporter impairment in these cells[20]; cholestasis is seldom observed in COVID-19 clinical setting. With regard to liver damage, however, is to underscore that cardiomyopathy with the resulting myocardial dysfunction, is observed in nearly 33% of hospitalized COVID-19 patients. In this context, the onset of circulatory impairment could reasonably explain the biochemical pattern of the alteration of AST, ALT, and GGT, as in the more common picture of congestive liver disease[21-23]. Supporting this view, a postmortem analysis of liver tissue coming from COVID-19 patients denoted a necrotic injury associated with hypoperfusion and congestive changes[19]. Finally, even if liver involvement is not generally regarded as relevant in subjects without a pre-existing liver impairment, the evaluation of hepatic changes during COVID-19 remains difficult in the clinical setting, due to the lack of a clear understanding of this process[24]. In this perspective, bilirubin elevation, even if rarely observed, seems to expose the patient to an increased risk of mechanical ventilation or death in comparison with changes of other liver enzymes[25].

COVID-19 AND PATIENTS WITH NON-CIRRHOTIC, CHRONIC LIVER DISEASEAs stated in the previous paragraph, the alterations of liver function tests in the course of COVID-19 could be the result of different causal events that may act simultan-eously. When damage occurs in a subject with an impaired hepatic functional reserve, a more difficult resolution of the clinical picture may be expected. Of more concern is the large United States study including more than 60 million electronic medical records, which also demonstrated that subjects affected by chronic liver diseases (CLD) were more exposed to acquiring COVID-19[26]. In this western research, the



Figure 1 Organs of human body (different from lung and liver) in which angiotensin-converting enzyme 2 was detected, the main site of expression and the possible coronavirus disease 2019 related symptoms/complications are reported in the corresponding columns. COVID-19: Coronavirus disease 2019; ACE2: Angiotensin-converting enzyme 2.

prevalence of CLD in COVID-19 patients accounted for 5% of cases, while data coming from China demonstrated a baseline CLD in 2 to 11% of cases[27].

Several conditions may determine a CLD in humans, including those of a metabolic, toxic, viral, or autoimmune nature. Non-alcoholic- (or metabolic-associated-) fatty-liver-disease (NAFLD), a liver condition ranging from simple steatosis to non-alcoholic steatohepatitis and associated with metabolic syndrome, type II diabetes and obesity, has an estimated prevalence of 25% worldwide[28]. Because of its large burden, the role of NAFLD in determining the degree of liver injury in patients with COVID-19 has aroused much interest.

In China, 70 out of 324 COVID-19 patients (21.6%) were diagnosed with hepatic steatosis during radiological investigations by computed tomography. The severity of the clinical picture of COVID-19 was also increased in those with NAFLD[29]. Moreover, in a retrospective study on 76 patients with COVID-19, the presence of NAFLD was associated with the progression of lung impairment, with an odds ratio of 6.4[30]. In another study, NAFLD was again associated with a fourfold increased risk of a severe course of COVID-19[31]. The picture linking NAFLD to a severe outcome of this viral infection was later challenged by a study coming from Qatar[32]. This research, including 320 patients with NAFLD, showed that this metabolic condition was only associated with a worse outcome (increased ICU stay and requirement of mechanical ventilation) when univariate analysis was employed. On the other hand, the main predictor for mortality or a worst outcome at multivariate analysis were age > 50 years and diabetes, respectively. On the basis of the above studies, it is clear why the possible link between NAFLD and severity of COVID-19 remains controversial. Heterogeneity among studies, related to retrospective patient inclusion and classi-fication, do not contribute to a clear picture. A further analysis possibly shed some light on this issue, linking the grade of liver fibrosis in NAFLD, rather than NAFLD by itself, to a worse COVID-19 outcome[33]. In this perspective, it seems that the extent of liver damage might have a more relevant role in comparison with the metabolic derangement observed in these patients.

Alcoholic liver disease (ALD) represents an important cause of hepatic morbidity and mortality worldwide[34]. Despite the fact that clinical data are not available on COVID-19 ALD patients, concern exists for the increased frailty of these subjects during the pandemic. On the other hand, psychological stress and social distancing seem to have increased individual alcoholic beverage consumption in general.



Alcohol abuse disorders frequently facilitate several comorbidities such as viral infections, diabetes or renal failure, all factors that may contribute to an increased risk of a severe course once SARS-CoV-2 infection occurs[35]. Moreover, increased consumption of alcoholic beverages is considered to be a risk factor for the development of respiratory tract infections that may be complicated by the onset of ARDS, one of the main features of severe COVID-19[17]. Also, in the recovery phase after COVID-19, previous alcohol abuse is considered a predisposing factor for the development of pulmonary fibrosis[36]. However, evidence-based indications in the management of ALD COVID-19-infected patients are lacking at present, and further studies are needed to improve our understanding on this issue. For the time being, expert opinion suggests, in general, to avoid the use of steroids in alcoholic hepatitis while concomitant uncontrolled infections are ongoing, because of the lowering effect on patient's immune defenses[37]. In clinical practice, on the other hand, the use of these drugs might be beneficial in the treatment of COVID-19 despite ALD, challenging our before-SARS-CoV-2 beliefs.

Another important challenge concerns the management of patients with autoimmune hepatitis on immunosuppressive therapy. The lack of data from scientific evidence has encouraged an empirical approach based on dose reduction of immunomodulatory therapy in order to prevent the most deleterious effects of COVID-19[38]. However, preliminary data from the city of Bergamo, in Italy, did not support this view since an increased risk was not observed in immunosuppressed patients during the SARS-CoV-2 epidemic. On the other hand, it must be considered that a possible reactivation of autoimmune hepatitis, after immunosuppressive therapy tapering, would probably require high-dose corticosteroids, greatly increasing the risk of infectious complications in these patients[39]. Therefore, current guidelines do not recommend the reduction of immunomodulatory therapy in the absence of SARS-CoV-2 infection; in the case of overt infection, dose adjustment may be pursued in order to increase the white blood cell count[40,41].

Few data are available on other forms of CLD. With regard to HBV, in a Chinese series represented by 105 hepatitis B surface antigen positive patients hospitalized for COVID-19 (1.9% with cirrhosis, 12.4% on antiviral therapy), 14 developed significant liver injury[12]. In these subjects, liver damage was associated with a severe course of infection in nearly 80% of cases. Since the possible prevention or attenuation of COVID-19 should be postulated with the use of antiviral drugs employed for hepatitis B virus (HBV) or hepatitis C virus (HCV), a Spanish study focused on subjects with chronic viral hepatitis following their specific antiviral regimen[42], with 1 out of 341 HCV patients and 8 out 1764 HBV patients developing COVID-19. Although the majority of them (nearly 80%) were hospitalized, no cases of fatal outcome were observed in this group. Finally, is to underscore that extensive use of intravenous steroid therapy to reduce the inflammatory state can lead to significant HBV reactivation. Routine testing for this virus would be wise in COVID-19 severe patients in order to adopt a timely prophylactic treatment[12].

No data are available at present on the impact of COVID-19 in patients with cholestatic CLD, such as primary biliary cholangitis or primary sclerosing cholangitis.

COVID-19 IN PATIENTS WITH CIRRHOSIS OF THE LIVERSeveral factors may lead a cirrhotic patient from a compensated to a decompensated clinical condition. Among these, infections play a relevant role, also promoting the occurrence of acute-on-chronic-liver-failure (ACLF)[43,44]. In this perspective, SARS-CoV-2 infection, characterized by an important activation of the cytokine cascade, sepsis, and altered hepatic perfusion, may determine a significant impairment of hepatic reserve in cirrhosis[6]. Some studies evaluated the clinical outcome of COVID-19 in patients with liver cirrhosis. Preliminary data coming from two international registries and including 103 cirrhotic COVID-19 positive patients demonstrated an increased mortality, in comparison with the general population[45]. Moreover, increased fatality was strictly related to the degree of pre-existing liver impairment; in fact, more than 60% of Child-Pugh C patients died in this study. In the APCOLIS study, a worst COVID-19 outcome in cirrhotic patients was again confirmed[46]. In this research, 43 cirrhotic patients exhibited: (1) increased need of ICU care; (2) more liver–related complication; and (3) enhanced mortality in comparison with non-cirrhotic CLD patients. COVID-19 determined the onset of liver decompensation or ACLF in one out of five patients with cirrhosis. A Child Pugh score > 9 was a significant predictor of mortality.



A study conducted in Italy compared the 30-d in-hospital mortality in a group of patients admitted for COVID-19 infection with or without liver cirrhosis. The all-cause mortality rate was found to be significantly higher in cirrhotic patients with a model for end-stage liver disease score ≥ 15. In addition, this research also evidenced a deteri-oration of liver function in patients with a long-standing stable liver disease. In fact, a worsening of the Child-Pugh score was observed, from grade A at admission to status B/C during the hospital stay[47]. The mortality rate induced by COVID-19 in patients with decompensated cirrhosis was increased in comparison with other infective causes. In Figure 2, the main postulated mechanisms of liver injury, during COVID-19, are reported, together with the possible outcome as a function of previous liver condition. Finally, cirrhotic patients with hepatocarcinoma (HCC) deserve a separate consideration since neoplastic patients, in general, may have an additional immunologic impairment related to cancer and/or chemotherapy. A national study performed in China on 1590 patients in 575 hospitals showed that patients with an underlying malignancy had a higher risk of SARS-CoV-2 infection than the general population. Obviously, many patients with HCC had an underlying liver impairment that placed them in this higher risk group, with a decidedly more inauspicious clinical outcome[48]. Considering this increased risk, the America Association for Study on Liver Disease (AASLD) and the European Association for the Study of Liver (EASL) recommend a prolongation of the time of clinical and ultrasound surveillance of cirrhotic patients so as not to expose them to possible contagion in the places of treatment, deferring locoregional treatment for HCC where possible in patients with a relatively stable disease, and reserving interventions for patients in which the benefit outweighs the risk of acquiring SARS-CoV-2 infection[40,41]. However, the relationship between an impairment of the immune system (for cancer, immunological diseases, or transplant) and a worse COVID-19 outcome is not demonstrated as yet, as will be reported in details in the following paragraph.

COVID-19 IN LIVER TRANSPLANTED PATIENTSAs there is an assumption that the integrity of the immune-system is an important factor for preventing the most severe sequelae of COVID-19, since the pandemic began, concern has been raised about the possible outcome in liver transplant (LT) patients under continuous immunosuppressive regimen. However, current data do not confirm this worrisome view. An early pediatric analysis coming from the North of Italy (an area badly beaten by the pandemic) demonstrated just three children infected by COVID-19, and with an uneventful outcome, among those liver transplanted or under chemotherapy[39]. Data on adult LT recipients coming from the same geographic area gave a different picture[49]. Three patients having a LT more than ten years before, and with COVID-19, died from complications. All had metabolic impairment (diabetes, hypertension) and were > 65 years. Paradoxically, their immunosuppressive regimen was minimal in comparison with patients with a shorter transplant history, suggesting a possible protective rather detrimental effect of immunosuppression. A large series was published, gathering the data from two international registries (COVID-Hep and SECURE-Cirrhosis)[50]. In this study, data coming from 151 liver transplanted patients were compared with those of 627 controls. The main factors associated with death at multivariate analysis were: Age, creatinine levels, and non-liver malignancy. LT and immunosuppressive regimen were not associated to an increased risk of fatality. Finally, patients with LT required invasive ventilation more frequently (20% LT vs 5% control; P < 0.001), while the overall mortality was increased in the non LT control group (19% LT vs 27% control; P = 0.05). A Spanish prospective study reassessed this issue in 111 LT patients with COVID-19[51]. Although the fatality rate was slightly lower in LT patients, mycophenolate use (especially at > 1 g/daily dose) was associated with an increased risk of severe COVID-19 disease. From these findings, the authors suggested a possible benefit in mycophenolate tapering in COVID-19 LT patients, while a complete drug withdrawal was discouraged. Taken together, the available data do not suggest an increased fatality in LT patients with COVID-19; however, enhanced surveillance would be wise in these subjects because of the increased morbidity reported by some studies. While a possible detrimental effect has been suggested for mycophenolate, a clear picture of the relationship between immunosuppression and COVID-19 severity is still lacking.



Figure 2 The putative mechanisms inducing liver injury in the course of coronavirus disease 2019 are reported in the left side of the figure (? = controversial data on the specific mechanism). The right part describes the possible outcome according to patient baseline liver condition. NAFLD: Non-alcoholic fatty-liver-disease; COVID-19: Coronavirus disease 2019.

CONCLUSIONThe effects of COVID-19 on liver health are complex and largely dependent on the underling functional hepatic reserve. The reasons for the elevation of liver enzymes in a significant number of COVID-19 patients, is not clear at present. As a general rule, the effect of COVID-19 seems negligible on the normal liver, but concern exists for patients with impaired liver function. Expert societies, such as AASLD or EASL, are making a great effort to give indications on issues such as LT and the clinical management of liver disease patients during the SARS-CoV-2 pandemic, also in the lack of definitive evidence. While this scenario is quickly evolving, also for the possible emergence of some variant of concern[52] ,other issues are rising for patients with liver disease. Among the pandemic-associated and non-COVID-19-related adverse effects, the reduction of health resources for patients with liver disease may significantly impact their morbidity and mortality. A study in Italy evidenced that there was already a 25% reduction of transplant activity related to an increased ICUs saturation in the first 4 wk of the national pandemic[53]. Abrupt interruption of HCC surveillance in the majority of patients with liver cirrhosis will also have a cost in the future. Finally, a recent study modeling a one year delay in HCV cure since the pandemic estimated an increase of more than 100 thousand in liver related deaths and cancers in the population affected by this virus in the coming years[54]. In conclusion, while hepatologists are presently working in an emergency environment full of uncertainties and with limited resources, an “in search of lost time” attitude will probably be pursued by the same physicians in the future, when the pandemic ends.

REFERENCESCiotti M, Angeletti S, Minieri M, Giovannetti M, Benvenuto D, Pascarella S, Sagnelli C, Bianchi M, Bernardini S, Ciccozzi M. COVID-19 Outbreak: An Overview. Chemotherapy 2019; 64: 215-223 [PMID: 32259829 DOI: 10.1159/000507423]

1


https://dx.doi.org/10.1159/000507423



Acuti Martellucci C, Flacco ME, Cappadona R, Bravi F, Mantovani L, Manzoli L. SARS-CoV-2 pandemic: An overview. Adv Biol Regul 2020; 77: 100736 [PMID: 32773099 DOI: 10.1016/j.jbior.2020.100736]

2

Chen N, Zhou M, Dong X, Qu J, Gong F, Han Y, Qiu Y, Wang J, Liu Y, Wei Y, Xia J, Yu T, Zhang X, Zhang L. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study. Lancet 2020; 395: 507-513 [PMID: 32007143 DOI: 10.1016/S0140-6736(20)30211-7]

3

Glaser T, Baiocchi L, Zhou T, Francis H, Lenci I, Grassi G, Kennedy L, Liangpunsakul S, Glaser S, Alpini G, Meng F. Pro-inflammatory signalling and gut-liver axis in non-alcoholic and alcoholic steatohepatitis: Differences and similarities along the path. J Cell Mol Med 2020; 24: 5955-5965 [PMID: 32314869 DOI: 10.1111/jcmm.15182]

4

Wang FS, Fan JG, Zhang Z, Gao B, Wang HY. The global burden of liver disease: the major impact of China. Hepatology 2014; 60: 2099-2108 [PMID: 25164003 DOI: 10.1002/hep.27406]

5

Bonnel AR, Bunchorntavakul C, Reddy KR. Immune dysfunction and infections in patients with cirrhosis. Clin Gastroenterol Hepatol 2011; 9: 727-738 [PMID: 21397731 DOI: 10.1016/j.cgh.2011.02.031]

6

Zhou P, Yang XL, Wang XG, Hu B, Zhang L, Zhang W, Si HR, Zhu Y, Li B, Huang CL, Chen HD, Chen J, Luo Y, Guo H, Jiang RD, Liu MQ, Chen Y, Shen XR, Wang X, Zheng XS, Zhao K, Chen QJ, Deng F, Liu LL, Yan B, Zhan FX, Wang YY, Xiao GF, Shi ZL. A pneumonia outbreak associated with a new coronavirus of probable bat origin. Nature 2020; 579: 270-273 [PMID: 32015507 DOI: 10.1038/s41586-020-2012-7]

7

Li W, Moore MJ, Vasilieva N, Sui J, Wong SK, Berne MA, Somasundaran M, Sullivan JL, Luzuriaga K, Greenough TC, Choe H, Farzan M. Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus. Nature 2003; 426: 450-454 [PMID: 14647384 DOI: 10.1038/nature02145]

8

Xu X, Chen P, Wang J, Feng J, Zhou H, Li X, Zhong W, Hao P. Evolution of the novel coronavirus from the ongoing Wuhan outbreak and modeling of its spike protein for risk of human transmission. Sci China Life Sci 2020; 63: 457-460 [PMID: 32009228 DOI: 10.1007/s11427-020-1637-5]

9

Zou X, Chen K, Zou J, Han P, Hao J, Han Z. Single-cell RNA-seq data analysis on the receptor ACE2 expression reveals the potential risk of different human organs vulnerable to 2019-nCoV infection. Front Med 2020; 14: 185-192 [PMID: 32170560 DOI: 10.1007/s11684-020-0754-0]

10

Wu Z, McGoogan JM. Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention. JAMA 2020; 323: 1239-1242 [PMID: 32091533 DOI: 10.1001/jama.2020.2648]

11

Jothimani D, Venugopal R, Abedin MF, Kaliamoorthy I, Rela M. COVID-19 and the liver. J Hepatol 2020; 73: 1231-1240 [PMID: 32553666 DOI: 10.1016/j.jhep.2020.06.006]

12

Richardson S, Hirsch JS, Narasimhan M, Crawford JM, McGinn T, Davidson KW; the Northwell COVID-19 Research Consortium, Barnaby DP, Becker LB, Chelico JD, Cohen SL, Cookingham J, Coppa K, Diefenbach MA, Dominello AJ, Duer-Hefele J, Falzon L, Gitlin J, Hajizadeh N, Harvin TG, Hirschwerk DA, Kim EJ, Kozel ZM, Marrast LM, Mogavero JN, Osorio GA, Qiu M, Zanos TP. Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area. JAMA 2020; 323: 2052-2059 [PMID: 32320003 DOI: 10.1001/jama.2020.6775]

13

Guan WJ, Ni ZY, Hu Y, Liang WH, Ou CQ, He JX, Liu L, Shan H, Lei CL, Hui DSC, Du B, Li LJ, Zeng G, Yuen KY, Chen RC, Tang CL, Wang T, Chen PY, Xiang J, Li SY, Wang JL, Liang ZJ, Peng YX, Wei L, Liu Y, Hu YH, Peng P, Wang JM, Liu JY, Chen Z, Li G, Zheng ZJ, Qiu SQ, Luo J, Ye CJ, Zhu SY, Zhong NS; China Medical Treatment Expert Group for Covid-19. Clinical Characteristics of Coronavirus Disease 2019 in China. N Engl J Med 2020; 382: 1708-1720 [PMID: 32109013 DOI: 10.1056/NEJMoa2002032]

14

Xu L, Liu J, Lu M, Yang D, Zheng X. Liver injury during highly pathogenic human coronavirus infections. Liver Int 2020; 40: 998-1004 [PMID: 32170806 DOI: 10.1111/liv.14435]

15

Pratt DS, Kaplan MM. Evaluation of abnormal liver-enzyme results in asymptomatic patients. N Engl J Med 2000; 342: 1266-1271 [PMID: 10781624 DOI: 10.1056/NEJM200004273421707]

16

Schaefer EAK, Arvind A, Bloom PP, Chung RT. Interrelationship Between Coronavirus Infection and Liver Disease. Clin Liver Dis (Hoboken) 2020; 15: 175-180 [PMID: 32489653 DOI: 10.1002/cld.967]

17

Cai Q, Huang D, Yu H, Zhu Z, Xia Z, Su Y, Li Z, Zhou G, Gou J, Qu J, Sun Y, Liu Y, He Q, Chen J, Liu L, Xu L. COVID-19: Abnormal liver function tests. J Hepatol 2020; 73: 566-574 [PMID: 32298767 DOI: 10.1016/j.jhep.2020.04.006]

18

Bradley BT, Maioli H, Johnston R, Chaudhry I, Fink SL, Xu H, Najafian B, Deutsch G, Lacy JM, Williams T, Yarid N, Marshall DA. Histopathology and ultrastructural findings of fatal COVID-19 infections in Washington State: a case series. Lancet 2020; 396: 320-332 [PMID: 32682491 DOI: 10.1016/S0140-6736(20)31305-2]

19

Zhao B, Ni C, Gao R, Wang Y, Yang L, Wei J, Lv T, Liang J, Zhang Q, Xu W, Xie Y, Wang X, Yuan Z, Zhang R, Lin X. Recapitulation of SARS-CoV-2 infection and cholangiocyte damage with human liver ductal organoids. Protein Cell 2020; 11: 771-775 [PMID: 32303993 DOI: 10.1007/s13238-020-00718-6]

20

Arentz M, Yim E, Klaff L, Lokhandwala S, Riedo FX, Chong M, Lee M. Characteristics and Outcomes of 21 Critically Ill Patients With COVID-19 in Washington State. JAMA 2020; 323: 1612-

21


https://dx.doi.org/10.1016/j.jbior.2020.100736


https://dx.doi.org/10.1016/S0140-6736(20)30211-7


https://dx.doi.org/10.1111/jcmm.15182






https://dx.doi.org/10.1038/s41586-020-2012-7




https://dx.doi.org/10.1007/s11427-020-1637-5


https://dx.doi.org/10.1007/s11684-020-0754-0














https://dx.doi.org/10.1002/cld.967




https://dx.doi.org/10.1016/S0140-6736(20)31305-2


https://dx.doi.org/10.1007/s13238-020-00718-6



1614 [PMID: 32191259 DOI: 10.1001/jama.2020.4326]van Deursen VM, Damman K, Hillege HL, van Beek AP, van Veldhuisen DJ, Voors AA. Abnormal liver function in relation to hemodynamic profile in heart failure patients. J Card Fail 2010; 16: 84-90 [PMID: 20123323 DOI: 10.1016/j.cardfail.2009.08.002]

22

Weisberg IS, Jacobson IM. Cardiovascular diseases and the liver. Clin Liver Dis 2011; 15: 1-20 [PMID: 21111990 DOI: 10.1016/j.cld.2010.09.010]

23

Moon AM, Barritt AS 4th. Elevated Liver Enzymes in Patients with COVID-19: Look, but Not Too Hard. Dig Dis Sci 2020 [PMID: 32875529 DOI: 10.1007/s10620-020-06585-9]

24

Hundt MA, Deng Y, Ciarleglio MM, Nathanson MH, Lim JK. Abnormal Liver Tests in COVID-19: A Retrospective Observational Cohort Study of 1,827 Patients in a Major U.S. Hospital Network. Hepatology 2020; 72: 1169-1176 [PMID: 32725890 DOI: 10.1002/hep.31487]

25

Wang QQ, Kaelber DC, Xu R, Volkow ND. COVID-19 risk and outcomes in patients with substance use disorders: analyses from electronic health records in the United States. Mol Psychiatry 2021; 26: 30-39 [PMID: 32929211 DOI: 10.1038/s41380-020-00880-7]

26

Zhang C, Shi L, Wang FS. Liver injury in COVID-19: management and challenges. Lancet Gastroenterol Hepatol 2020; 5: 428-430 [PMID: 32145190 DOI: 10.1016/S2468-1253(20)30057-1]

27

Sheka AC, Adeyi O, Thompson J, Hameed B, Crawford PA, Ikramuddin S. Nonalcoholic Steatohepatitis: A Review. JAMA 2020; 323: 1175-1183 [PMID: 32207804 DOI: 10.1001/jama.2020.2298]

28

Qian ZP, Mei X, Zhang YY, Zou Y, Zhang ZG, Zhu H, Guo HY, Liu Y, Ling Y, Zhang XY, Wang JF, Lu HZ. [Analysis of baseline liver biochemical parameters in 324 cases with novel coronavirus pneumonia in Shanghai area]. Zhonghua Gan Zang Bing Za Zhi 2020; 28: 229-233 [PMID: 32270660 DOI: 10.3760/cma.j.cn501113-20200229-00076]

29

Ji D, Qin E, Xu J, Zhang D, Cheng G, Wang Y, Lau G. Non-alcoholic fatty liver diseases in patients with COVID-19: A retrospective study. J Hepatol 2020; 73: 451-453 [PMID: 32278005 DOI: 10.1016/j.jhep.2020.03.044]

30

Zhou YJ, Zheng KI, Wang XB, Sun QF, Pan KH, Wang TY, Ma HL, Chen YP, George J, Zheng MH. Metabolic-associated fatty liver disease is associated with severity of COVID-19. Liver Int 2020; 40: 2160-2163 [PMID: 32573883 DOI: 10.1111/liv.14575]

31

Mushtaq K, Khan MU, Iqbal F, Alsoub DH, Chaudhry HS, Ata F, Iqbal P, Elfert K, Balaraju G, Almaslamani M, Al-Ejji K, AlKaabi S, Kamel YM. NAFLD is a predictor of liver injury in COVID-19 hospitalized patients but not of mortality, disease severity on the presentation or progression - The debate continues. J Hepatol 2021; 74: 482-484 [PMID: 33223215 DOI: 10.1016/j.jhep.2020.09.006]

32

Targher G, Mantovani A, Byrne CD, Wang XB, Yan HD, Sun QF, Pan KH, Zheng KI, Chen YP, Eslam M, George J, Zheng MH. Risk of severe illness from COVID-19 in patients with metabolic dysfunction-associated fatty liver disease and increased fibrosis scores. Gut 2020; 69: 1545-1547 [PMID: 32414813 DOI: 10.1136/gutjnl-2020-321611]

33

Da BL, Im GY, Schiano TD. Coronavirus Disease 2019 Hangover: A Rising Tide of Alcohol Use Disorder and Alcohol-Associated Liver Disease. Hepatology 2020; 72: 1102-1108 [PMID: 32369624 DOI: 10.1002/hep.31307]

34

Ahmed MZ, Ahmed O, Aibao Z, Hanbin S, Siyu L, Ahmad A. Epidemic of COVID-19 in China and associated Psychological Problems. Asian J Psychiatr 2020; 51: 102092 [PMID: 32315963 DOI: 10.1016/j.ajp.2020.102092]

35

Ojo AS, Balogun SA, Williams OT, Ojo OS. Pulmonary Fibrosis in COVID-19 Survivors: Predictive Factors and Risk Reduction Strategies. Pulm Med 2020; 2020: 6175964 [PMID: 32850151 DOI: 10.1155/2020/6175964]

36

Crabb DW, Im GY, Szabo G, Mellinger JL, Lucey MR. Diagnosis and Treatment of Alcohol-Associated Liver Diseases: 2019 Practice Guidance From the American Association for the Study of Liver Diseases. Hepatology 2020; 71: 306-333 [PMID: 31314133 DOI: 10.1002/hep.30866]

37

Lleo A, Invernizzi P, Lohse AW, Aghemo A, Carbone M. Management of patients with autoimmune liver disease during COVID-19 pandemic. J Hepatol 2020; 73: 453-455 [PMID: 32283134 DOI: 10.1016/j.jhep.2020.04.002]

38

D'Antiga L. Coronaviruses and Immunosuppressed Patients: The Facts During the Third Epidemic. Liver Transpl 2020; 26: 832-834 [PMID: 32196933 DOI: 10.1002/lt.25756]

39

Fix OK, Hameed B, Fontana RJ, Kwok RM, McGuire BM, Mulligan DC, Pratt DS, Russo MW, Schilsky ML, Verna EC, Loomba R, Cohen DE, Bezerra JA, Reddy KR, Chung RT. Clinical Best Practice Advice for Hepatology and Liver Transplant Providers During the COVID-19 Pandemic: AASLD Expert Panel Consensus Statement. Hepatology 2020; 72: 287-304 [PMID: 32298473 DOI: 10.1002/hep.31281]

40

Boettler T, Newsome PN, Mondelli MU, Maticic M, Cordero E, Cornberg M, Berg T. Care of patients with liver disease during the COVID-19 pandemic: EASL-ESCMID position paper. JHEP Rep 2020; 2: 100113 [PMID: 32289115 DOI: 10.1016/j.jhepr.2020.100113]

41

Lens S, Miquel M, Mateos-Muñoz B, García-Samaniego J, Forns X. SARS-CoV-2 in patients on antiviral HBV and HCV therapy in Spain. J Hepatol 2020; 73: 1262-1263 [PMID: 32673740 DOI: 10.1016/j.jhep.2020.07.007]

42

Jalan R, Fernandez J, Wiest R, Schnabl B, Moreau R, Angeli P, Stadlbauer V, Gustot T, Bernardi M, Canton R, Albillos A, Lammert F, Wilmer A, Mookerjee R, Vila J, Garcia-Martinez R, Wendon J, Such J, Cordoba J, Sanyal A, Garcia-Tsao G, Arroyo V, Burroughs A, Ginès P. Bacterial infections in cirrhosis: a position statement based on the EASL Special Conference 2013. J Hepatol 2014; 60:

43




https://dx.doi.org/10.1016/j.cardfail.2009.08.002


https://dx.doi.org/10.1016/j.cld.2010.09.010


https://dx.doi.org/10.1007/s10620-020-06585-9




https://dx.doi.org/10.1038/s41380-020-00880-7


https://dx.doi.org/10.1016/S2468-1253(20)30057-1




https://dx.doi.org/10.3760/cma.j.cn501113-20200229-00076












https://dx.doi.org/10.1016/j.ajp.2020.102092


https://dx.doi.org/10.1155/2020/6175964










https://dx.doi.org/10.1016/j.jhepr.2020.100113





1310-1324 [PMID: 24530646 DOI: 10.1016/j.jhep.2014.01.024]Arroyo V, Moreau R, Jalan R. Acute-on-Chronic Liver Failure. N Engl J Med 2020; 382: 2137-2145 [PMID: 32459924 DOI: 10.1056/NEJMra1914900]

44

Moon AM, Webb GJ, Aloman C, Armstrong MJ, Cargill T, Dhanasekaran R, Genescà J, Gill US, James TW, Jones PD, Marshall A, Mells G, Perumalswami PV, Qi X, Su F, Ufere NN, Barnes E, Barritt AS, Marjot T. High mortality rates for SARS-CoV-2 infection in patients with pre-existing chronic liver disease and cirrhosis: Preliminary results from an international registry. J Hepatol 2020; 73: 705-708 [PMID: 32446714 DOI: 10.1016/j.jhep.2020.05.013]

45

Sarin SK, Choudhury A, Lau GK, Zheng MH, Ji D, Abd-Elsalam S, Hwang J, Qi X, Cua IH, Suh JI, Park JG, Putcharoen O, Kaewdech A, Piratvisuth T, Treeprasertsuk S, Park S, Wejnaruemarn S, Payawal DA, Baatarkhuu O, Ahn SH, Yeo CD, Alonzo UR, Chinbayar T, Loho IM, Yokosuka O, Jafri W, Tan S, Soo LI, Tanwandee T, Gani R, Anand L, Esmail ES, Khalaf M, Alam S, Lin CY, Chuang WL, Soin AS, Garg HK, Kalista K, Batsukh B, Purnomo HD, Dara VP, Rathi P, Al Mahtab M, Shukla A, Sharma MK, Omata M; APASL COVID Task Force, APASL COVID Liver Injury Spectrum Study (APCOLIS Study-NCT 04345640). Pre-existing liver disease is associated with poor outcome in patients with SARS CoV2 infection; The APCOLIS Study (APASL COVID-19 Liver Injury Spectrum Study). Hepatol Int 2020; 14: 690-700 [PMID: 32623632 DOI: 10.1007/s12072-020-10072-8]

46

Iavarone M, D'Ambrosio R, Soria A, Triolo M, Pugliese N, Del Poggio P, Perricone G, Massironi S, Spinetti A, Buscarini E, Viganò M, Carriero C, Fagiuoli S, Aghemo A, Belli LS, Lucà M, Pedaci M, Rimondi A, Rumi MG, Invernizzi P, Bonfanti P, Lampertico P. High rates of 30-day mortality in patients with cirrhosis and COVID-19. J Hepatol 2020; 73: 1063-1071 [PMID: 32526252 DOI: 10.1016/j.jhep.2020.06.001]

47

Liang W, Guan W, Chen R, Wang W, Li J, Xu K, Li C, Ai Q, Lu W, Liang H, Li S, He J. Cancer patients in SARS-CoV-2 infection: a nationwide analysis in China. Lancet Oncol 2020; 21: 335-337 [PMID: 32066541 DOI: 10.1016/S1470-2045(20)30096-6]

48

Bhoori S, Rossi RE, Citterio D, Mazzaferro V. COVID-19 in long-term liver transplant patients: preliminary experience from an Italian transplant centre in Lombardy. Lancet Gastroenterol Hepatol 2020; 5: 532-533 [PMID: 32278366 DOI: 10.1016/S2468-1253(20)30116-3]

49

Webb GJ, Marjot T, Cook JA, Aloman C, Armstrong MJ, Brenner EJ, Catana MA, Cargill T, Dhanasekaran R, García-Juárez I, Hagström H, Kennedy JM, Marshall A, Masson S, Mercer CJ, Perumalswami PV, Ruiz I, Thaker S, Ufere NN, Barnes E, Barritt AS 4th, Moon AM. Outcomes following SARS-CoV-2 infection in liver transplant recipients: an international registry study. Lancet Gastroenterol Hepatol 2020; 5: 1008-1016 [PMID: 32866433 DOI: 10.1016/S2468-1253(20)30271-5]

50

Colmenero J, Rodríguez-Perálvarez M, Salcedo M, Arias-Milla A, Muñoz-Serrano A, Graus J, Nuño J, Gastaca M, Bustamante-Schneider J, Cachero A, Lladó L, Caballero A, Fernández-Yunquera A, Loinaz C, Fernández I, Fondevila C, Navasa M, Iñarrairaegui M, Castells L, Pascual S, Ramírez P, Vinaixa C, González-Dieguez ML, González-Grande R, Hierro L, Nogueras F, Otero A, Álamo JM, Blanco-Fernández G, Fábrega E, García-Pajares F, Montero JL, Tomé S, De la Rosa G, Pons JA. Epidemiological pattern, incidence, and outcomes of COVID-19 in liver transplant patients. J Hepatol 2021; 74: 148-155 [PMID: 32750442 DOI: 10.1016/j.jhep.2020.07.040]

51

Galloway SE, Paul P, MacCannell DR, Johansson MA, Brooks JT, MacNeil A, Slayton RB, Tong S, Silk BJ, Armstrong GL, Biggerstaff M, Dugan VG. Emergence of SARS-CoV-2 B.1.1.7 Lineage - United States, December 29, 2020-January 12, 2021. MMWR Morb Mortal Wkly Rep 2021; 70: 95-99 [PMID: 33476315 DOI: 10.15585/mmwr.mm7003e2]

52

Angelico R, Trapani S, Manzia TM, Lombardini L, Tisone G, Cardillo M. The COVID-19 outbreak in Italy: Initial implications for organ transplantation programs. Am J Transplant 2020; 20: 1780-1784 [PMID: 32243677 DOI: 10.1111/ajt.15904]

53

Blach S, Kondili LA, Aghemo A, Cai Z, Dugan E, Estes C, Gamkrelidze I, Ma S, Pawlotsky JM, Razavi-Shearer D, Razavi H, Waked I, Zeuzem S, Craxi A. Impact of COVID-19 on global HCV elimination efforts. J Hepatol 2021; 74: 31-36 [PMID: 32777322 DOI: 10.1016/j.jhep.2020.07.042]

54








https://dx.doi.org/10.1007/s12072-020-10072-8




https://dx.doi.org/10.1016/S1470-2045(20)30096-6


https://dx.doi.org/10.1016/S2468-1253(20)30116-3


https://dx.doi.org/10.1016/S2468-1253(20)30271-5




https://dx.doi.org/10.15585/mmwr.mm7003e2






World Journal of



ORIGINAL ARTICLE

Basic Study

Upregulation of long noncoding RNA W42 promotes tumor development by binding with DBN1 in hepatocellular carcinoma

Guang-Lin Lei, Yan Niu, Si-Jie Cheng, Yuan-Yuan Li, Zhi-Fang Bai, Ling-Xiang Yu, Zhi-Xian Hong, Hu Liu, Hong-Hong Liu, Jin Yan, Yuan Gao, Shao-Geng Zhang, Zhu Chen, Rui-Sheng Li, Peng-Hui Yang

ORCID number: Guang-Lin Lei 0000-0002-2324-4337; Yan Niu 0000-0002-5217-2408; Si-Jie Cheng 0000-0003-4851-6017; Yuan-Yuan Li 0000-0002-9656-0541; Zhi-Fang Bai 0000-0001-7591-7276; Ling-Xiang Yu 0000-0001-6610-6287; Zhi-Xian Hong 0000-0003-2143-5073; Hu Liu 0000-0001-7473-2608; Hong-Hong Liu 0000-0002-9221-8300; Jin Yan 0000-0002-1521-0348; Yuan Gao 0000-0001-6281-4980; Shao-Geng Zhang 0000-0002-0476-6279; Zhu Chen 0000-0001-8717-5484; Rui-Sheng Li 0000-0002-9267-1009; Peng-Hui Yang 0000-0001-6519-5927.

Author contributions: Lei GL and Niu Y contributed equally to this work; Li RS and Yang PH are both corresponding authors; Lei GL and Niu Y performed the experiments and wrote the manuscript; Cheng SJ, Li YY, Bai ZF, Yu LX, Hong ZX and Liu H collected the data; Liu HH, Yan J, Gao Y, Zhang SG, Chen Z, Li RS and Yang PH analyzed the data; all authors approved the final version of the article.

Institutional review board statement: This study was approved by the Ethics Committee of the Fifth Medical Center of Chinese PLA General Hospital, and carried out according to the Declaration of Helsinki.

Guang-Lin Lei, Si-Jie Cheng, Yuan-Yuan Li, Zhi-Fang Bai, Ling-Xiang Yu, Zhi-Xian Hong, Hu Liu, Hong-Hong Liu, Jin Yan, Yuan Gao, Shao-Geng Zhang, Zhu Chen, Rui-Sheng Li, Peng-Hui Yang, Fifth Medical Center of Chinese PLA General Hospital, Beijing 100039, China

Yan Niu, Inner Mongolia Medical University, Hohhot 010110, Inner Mongolia Autonomous Region, China

Corresponding author: Peng-Hui Yang, MD, Professor, Fifth Medical Center of Chinese PLA General Hospital, No. 100 Xisihuan Zhong Road, Fengtai District, Beijing 100039, China. [email protected]

AbstractBACKGROUND Hepatocellular carcinoma (HCC) is a malignancy found globally. Accumulating studies have shown that long noncoding RNAs (lncRNAs) play critical roles in HCC. However, the function of lncRNA in HCC remains poorly understood.

AIM To understand the effect of lncRNA W42 on HCC and dissect the underlying molecular mechanisms.

METHODS We measured the expression of lncRNA W42 in HCC tissues and cells (Huh7 and SMMC-7721) by quantitative reverse transcriptase polymerase chain reaction. Receiver operating characteristic curves were used to assess the sensitivity and specificity of lncRNA W42 expression. HCC cells were transfected with pcDNA3.1-lncRNA W42 or shRNA-lncRNA W42. Cell functions were detected by cell counting Kit-8 (CCK-8), colony formation, flow cytometry and Transwell assays. The interaction of lncRNA W42 and DBN1 was confirmed by RNA immunoprecipitation and RNA pull down assays. An HCC xenograft model was used to assess the role of lncRNA W42 on tumor growth in vivo. The Kaplan-Meier curve was used to evaluate the overall survival and recurrence-free survival after surgery in patients with HCC.

RESULTS In this study, we identified a novel lncRNA (lncRNA W42), and investigated its biological functions and clinical significance in HCC. LncRNA W42 expression



http://orcid.org/0000-0002-2324-4337

http://orcid.org/0000-0002-2324-4337

http://orcid.org/0000-0002-2324-4337

http://orcid.org/0000-0002-5217-2408

http://orcid.org/0000-0002-5217-2408

http://orcid.org/0000-0002-5217-2408

http://orcid.org/0000-0003-4851-6017

http://orcid.org/0000-0003-4851-6017

http://orcid.org/0000-0003-4851-6017

http://orcid.org/0000-0002-9656-0541

http://orcid.org/0000-0002-9656-0541

http://orcid.org/0000-0002-9656-0541

http://orcid.org/0000-0001-7591-7276

http://orcid.org/0000-0001-7591-7276

http://orcid.org/0000-0001-7591-7276

http://orcid.org/0000-0001-6610-6287

http://orcid.org/0000-0001-6610-6287

http://orcid.org/0000-0001-6610-6287

http://orcid.org/0000-0003-2143-5073

http://orcid.org/0000-0003-2143-5073

http://orcid.org/0000-0001-7473-2608

http://orcid.org/0000-0001-7473-2608

http://orcid.org/0000-0001-7473-2608

http://orcid.org/0000-0002-9221-8300

http://orcid.org/0000-0002-9221-8300

http://orcid.org/0000-0002-1521-0348

http://orcid.org/0000-0002-1521-0348

http://orcid.org/0000-0002-1521-0348

http://orcid.org/0000-0001-6281-4980

http://orcid.org/0000-0001-6281-4980

http://orcid.org/0000-0001-6281-4980

http://orcid.org/0000-0002-0476-6279

http://orcid.org/0000-0002-0476-6279

http://orcid.org/0000-0001-8717-5484

http://orcid.org/0000-0001-8717-5484

http://orcid.org/0000-0002-9267-1009

http://orcid.org/0000-0002-9267-1009

http://orcid.org/0000-0001-6519-5927

http://orcid.org/0000-0001-6519-5927


Lei GL et al. LncRNA W42 and HCC


Institutional animal care and use committee statement: All procedures involving animals were reviewed and approved by the Institutional Animal Care and Use Committee of the Fifth Medical Center of Chinese PLA General Hospital.

Conflict-of-interest statement: All authors have nothing to disclose.

Data sharing statement: No additional data are available.

ARRIVE guidelines statement: The authors have read the ARRIVE guidelines, and the manuscript was prepared and revised according to the ARRIVE guidelines.


Manuscript source: Unsolicited manuscript



Peer-review report’s scientific quality classificationGrade A (Excellent): A Grade B (Very good): B Grade C (Good): C Grade D (Fair): 0 Grade E (Poor): 0

Received: November 18, 2020 Peer-review started: November 18, 2020 First decision: January 23, 2021 Revised: February 10, 2021 Accepted: April 2, 2021

was upregulated in HCC tissues and cells. Overexpression of lncRNA W42 notably promoted the proliferative and invasion of HCC, and inhibited cell apoptosis. LncRNA W42 directly bound to DBN1 and activated the downstream pathway. LncRNA W42 knockdown suppressed HCC xenograft tumor growth in vivo. The clinical investigation revealed that HCC patients with high lncRNA W42 expression exhibited shorter survival times.

CONCLUSION In vitro and in vivo results suggested that the novel lncRNA W42, which is upregulated in HCC, may serve as a potential candidate prognostic biomarker and therapeutic target in HCC patients.

Key Words: Hepatocellular carcinoma; Long noncoding RNA; Long noncoding RNA W42; Tumor; DBN1; Cancer


Core Tip: The results of this study demonstrated, for the first time, that long noncoding RNA (lncRNA) W42 expression was significantly higher in hepatocellular carcinoma (HCC) tissues than in normal tissues and that dysregulation of lncRNA W42 was positively correlated with tumor number, liver cirrhosis and tumor recurrence in patients with HCC. Moreover, increased lncRNA W42 expression was associated with a poor prognosis in patients with HCC. These findings suggest that lncRNA W42 is an important marker for indicting prognosis and may have potential as a diagnostic and therapeutic target for HCC.

Citation: Lei GL, Niu Y, Cheng SJ, Li YY, Bai ZF, Yu LX, Hong ZX, Liu H, Liu HH, Yan J, Gao Y, Zhang SG, Chen Z, Li RS, Yang PH. Upregulation of long noncoding RNA W42 promotes tumor development by binding with DBN1 in hepatocellular carcinoma. World J Gastroenterol 2021; 27(20): 2586-2602URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2586.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2586

INTRODUCTIONLiver cancer is a major health problem[1] and is currently the third leading cause of cancer-related mortality in the world[2]. Among all primary liver cancers, hepato-cellular carcinoma (HCC) accounts for approximately 90% of cases[3]. Several complex pathways with frequent mutations in HCC, including the telomere maintenance, cell cycle, WNT-β-catenin, epigenetic and chromatin remodeling pathways, have been well defined[4-8]. Although advances in the diagnosis and treatment of HCC have been made, its prognosis remains poorly understood. Even with radical treatment of HCC, the median 5-year survival rate remains lower than 50%[9-11].

Long noncoding RNAs (lncRNAs) are noncoding transcripts that exceed 200 nt in length; they have been defined as the largest subclass in the noncoding transcriptome in humans[12-14]. The majority of lncRNAs with limited protein coding potential are transcribed by RNA polymerase II and polyadenylated at their 5′ and 3′ ends, respectively[15]. Recent reports have shown that a growing number of cancer transcriptomes have indeed revealed that thousands of aberrant lncRNAs are closely related to a variety of cancer types[16-18], including HCC. In particular, previous studies have revealed that a number of lncRNAs are dysregulated and exert essential roles in HCC progression; these lncRNAs include BZRAP1-AS1[19], RAB5IF[20], SNHG15[21], MAGI2-AS3[22], ATB[23], etc. We also previously demonstrated that the decreased expression of long intergenic noncoding RNA P7 predicts unfavorable prognosis and promotes tumor proliferation via the modulation of the STAT1-MAPK pathway in HCC[24], and lncRNA W5 inhibits progression and predicts favorable prognosis in HCC[25].

In this report, for the first time, we identified a novel lncRNA, W42, and invest-igated its expression levels in HCC tumor tissues and its association with the clinical-








Article in press: April 2, 2021 Published online: May 28, 2021

P-Reviewer: Li J, Watanabe T S-Editor: Gao CC L-Editor: Webster JR P-Editor: Liu JH

pathological parameters of HCC patients. The expression, biological function and underlying molecular mechanisms of lncRNA W42 in HCC were also studied in vivo and in vitro.

MATERIALS AND METHODSClinical samplesData were obtained from 92 patients with HCC (72 males and 20 females) who underwent surgery for HCC at the Department of Hepatobiliary Surgery of the Fifth Medical Center, Chinese PLA General Hospital between October 2013 and July 2019. After obtaining written informed consent from all patients, tumor tissues and paired adjacent noncancerous tissue samples were collected from patients with HCC in accordance with the institutional guidelines of the hospital’s Ethics Committee. Resected tissues were immediately snap-frozen in liquid nitrogen within 30 min and preserved in an HCC Tissue Biobank until use. This study was approved by the Ethics Committee of the Fifth Medical Center of Chinese PLA General Hospital, and carried out according to the Declaration of Helsinki.

Cell linesThe human HCC cell lines HepG2 and Huh7 were purchased from the American Type Culture Collection (ATCC, Manassas, VA, United States). The normal liver cell line LO2 and human HCC cell line SMMC-7721 were obtained from Cell Bank, Shanghai Institutes for Biological Sciences, Chinese Academy of Science (Shanghai, China). SMMC-7721 cells were established from a 50-year-old Chinese male HCC patient with negative HBsAg and positive α-fetoprotein (AFP) by the scholars of Second Military Medical University (Shanghai, China) in 1977[26]. The use of the cell lines was approved by the Ethics Committee of the Fifth Medical Center of Chinese PLA General Hospital. All of the cell lines used in this study were maintained in DMEM (Thermo Fisher, Beijing, China) supplemented with 10% fetal bovine serum (Gibco, Beijing, China) in 5% CO2 at 37°C.

RNA extraction and real-time polymerase chain reactionTotal RNA from frozen fresh HCC tissues and paired adjacent noncancerous tissues was extracted using TRIzol reagent (Thermo Fisher) according to the manufacturer's instructions. RNA concentration and purity were detected using a NanoDrop ND-1000 (Thermo Fisher, Beijing, China). LncRNA W42 expression levels were examined by quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) performed with Maxima SYBR Green on an ABI 7500 instrument (Applied Biosystems, CA, United States). All reactions were run at least three times using lncRNA W42-specific primers. GAPDH (glyceraldehyde-3-phosphate dehydrogenase) expression was monitored as the endogenous control, and all samples were normalized to human GAPDH[24,27,28]. The primer sequences of lncRNA W42 were as follows: forward: 5’-TGTGAACCAGGTTTGCTGGA-3’, reverse: 5’-CTCAACCATGCCGACGAGAA-3’; GAPDH forward: 5’-CAGCCTCAAGATCATCAGCA-3’ reverse: 5’-TGTGGTCAT-GAGTCCTTCCA-3’. The amplification procedure was 95°C for 5 min, followed by 40 cycles of denaturation at 95°C for 15 s and annealing at 60°C for 30 s, extension at 72°C for 30 s. The median of triplicate reactions was used to calculate relative lncRNA expression (ΔCt = Ct median lncRNA - Ct median GAPDH). Expression changes were calculated using the 2-ΔΔCt method.

Vector construction and Lentiviral InfectionThe lncRNA W42 overexpression vector was constructed and cloned into the BamHI and EcorI sites of the pcDNA3.1 (+) vector, yielding pcDNA3.1-lncRNA W42. The primers used in this experiment were as follows: forward: 5’-AGCTGAGG-GAGCCGGCT-3’, reverse: 5’-ATTAATGAGAGAATTATGGTAT-3’. In addition, gene knockdown used RNA interference experiments. LncRNA W42 was knocked down by shRNA lentiviruses. shRNA-lncRNA W42-1 and shRNA-lncRNA W42-2 were designed and synthesized by RIBOBIO Co., Ltd. (Guangzhou, China). Their sequences were as follows: W42-1: 3’-dTdT GGCUCAGUCUUUGAUCAAA-5’; W42-2: 3’-dTdT CCUGUAGUAAACUACUCAA-5’; the shRNA negative control sequence number was siN05815122147 of RIBOBIO. SMMC7721 cells (1 × 104) were stably transfected with lentiviral vector containing lncRNA W42 with the help of Hanbio Biotechnology Co., Ltd. (Shanghai, China). After infection for 72 h, the cells were collected for RNA



quantitation of lncRNA W42 and targeted genes using qRT-PCR.

Nucleocytoplasmic separation experimentA nucleocytoplasmic separation experiment was conducted with the PARISTM Protein and RNA Isolation System (Ambion, Life Technologies Lithuania). The experiment followed the manufacturer’s instructions.

Cell proliferation assayCell proliferation was assessed by the Cell Counting Kit-8 (CCK-8, Dojindo Laboratories, Kumamoto, Japan) assay according to the manufacturer’s protocols. Huh7 or SMMC7721 cells were plated in 12-well plates (5 × 105 cells/well) and then transfected with lncRNA W42 overexpression vector, and the numbers of cells per well were examined and counted by the absorbance (450 nm) of reduced WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-isulfophenyl)-2H-tetrazolium, monosodium salt] at the indicated time.

Colony formation assayHuh7 or SMMC7721 cells were plated in 6-well plates and incubated in DMEM with 10% fetal bovine serum (FBS) with 5% CO2 at 37°C. Six days later, the cells were fixed and stained with 0.1% crystal violet. The number of colonies, defined as > 50 cells/colony, was recorded and photographed[24].

Transwell assayCell migration and invasion were determined using Transwell chambers (Sigma). The upper chamber of the Transwell was treated with Matrigel (without Matrigel coating for the migration assay). Briefly, SMMC7721 or Huh7 cells (50000 cells/well) were resuspended in serum-free DMEM (Gibco). Then, the medium containing the cells was seeded into the upper chamber. In addition, the lower chamber was supplemented with medium containing 10% FBS (Gibco) as a chemoattractant. After incubation for 48 h, the cells that did not migrate or invade the membrane were scraped away, while the migrated and invaded cells in the lower chamber of the Transwell were fixed with methanol and stained with 0.1% crystal violet (Sigma). The number of migrated and invaded cells was counted using a microscope.

Cell apoptosis assay and cell cycle analysisAn Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) apoptosis kit (BestBio, Shanghai, China) was used for the apoptosis assay and analyzed by flow cytometry. After treatment, SMMC7721 and Huh7 were washed three times with phosphate buffered saline (PBS) and stained with Annexin V-FITC/PI. After filtration, the suspension of each group was analyzed within 1 h using a BD FACSCalibur flow cytometer (BD Biosciences, Franklin Lakes, NJ, United States). The cells were transfected with lncRNA W42 overexpression vector or control and cultured for 48 h. Then the cells were washed twice with PBS and fixed in cold 70% ethanol for 1 h followed by incubation with 20 µg/mL PI (BestBio, Shanghai, China) and 200 µg/mL RNase A (Sigma) for 30 min. Finally, the cells were analyzed utilizing a flow cytometer (BD Biosciences, Franklin Lakes, NJ, United States).

RNA pull-down assaysRNA pull-down assays were performed as previously described[29,30]. Briefly, in vitro biotin-labeled RNAs (lncRNA W42 and its antisense RNA) were transcribed by biotin RNA labeling mix (Roche) and T7 RNA polymerase (Roche), treated with RNase-free DNase I (Promega) and purified using the RNeasy Mini Kit (QIAGEN). Biotinylated RNA was incubated with cell lysate, and precipitated proteins were separated via sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and subjected to mass spectrometric (MS) analysis. RNA pull-down of AVAN-associated proteins was performed using biotinylated lncRNA W42 or antisense probes (immunoglobulin G, IgG). Isolated proteins were resolved by SDS-PAGE followed by silver staining. Silver staining was performed with the SilverQuest™ Silver Staining Kit (Invitrogen) following the manufacturer’s recommendations.

RNA immunoprecipitationThe RNA immunoprecipitation (RIP) assay was performed as previously de-scribed[31,32]. With IgG antibodies as the isotype-matched control, nuclear extracts were immunoprecipitated with 2.5 mg hnRNP U (Abcam, clone 4D11, ab6106) overnight. RNA-protein-antibody complexes were captured using Dynabead Protein



A/G (Thermo Fisher Scientific). RNA was eluted by adding TRIzol directly to magnetic beads and isolated according to the manufacturer’s instructions. cDNA was synthesized using TransScript First-Strand cDNA Synthesis SuperMix (TransGen Biotech) and analyzed by qPCR. The results were normalized to input RNA and are shown as fold-enrichment over control IgG RIP.

In vivo experimentsA total of 16 female 4-6-wk-old nude BALB/C mice were purchased from Charles River Laboratories (Beijing, China) and were maintained in a SPF (specific pathogen-free) facility under pathogen-free conditions at 22°C and 40%-50% humidity with a 12/12 h light/dark cycle, and had ad libitum access to food and water. Each mouse was randomly assigned to one of the following two groups: LncRNA W42 knockdown and control group (n = 8 per group). SMMC7721 cells (2 × 106) stably lncRNA W42 knockdown or empty vector were subcutaneously injected into the flanks of nude mice. Tumor growth was monitored by tumor volume and weight. After 4 d injection, tumor size was recorded every 3 d, and tumor volumes were calculated with the formula: Tumor volume (mm3) = length × width2 × 0.5. After 36 d, the mice were euthanized by i.p. injection of sodium pentobarbital (150 mg/kg), and tumor tissues were removed, weighed and fixed in polyformaldehyde after imaging of the tumors for further examination. All animal procedures were performed in strict accordance with the recommendations of the National Institutes of Health Laboratory Animal Care and Use Guidelines. This study was approved by the Ethics Committee of the Fifth Medical Center of Chinese PLA General Hospital (2020-002).

Statistical analysisAll data were analyzed using the SPSS 20.0 software package (Chicago, IL, United States). All statistical results are expressed as the mean ± SD. Kaplan-Meier analyses were used to represent the correlations between lncRNA W42 expression level and overall survival (OS) rate or recurrence-free survival (RFS). The data from the two groups were assessed using the Student t-test, the differences between various groups were analyzed by one-way analysis of variance (ANOVA), followed by Tukey’s post hoc test. All experiments were carried out at least three times. P values < 0.05 were considered significant.

RESULTSLncRNA W42 was upregulated in human HCC tumor tissues and cell linesBased on our previous bioinformatics analysis, we chose another 10 candidate lncRNAs and re-validated their expression in the normal liver cell line LO2 and the human HCC cell line Huh7 (Supplementary Figure 1A). Subsequently, we re-validated the expression of lncRNA W42 in 25 pairs of HCC tissues and paired adjacent noncan-cerous tissues. The results showed that lncRNA W42 was the most differentially expressed among these lncRNAs. Hence, we chose lncRNA W42 as a candidate lncRNA for HCC in the following study. To define the role of lncRNA W42 in human HCC, we first measured lncRNA W42 expression in 92 pairs of HCC tissues and adjacent noncancerous tissues by qRT-PCR. As shown in Figure 1, we observed that lncRNA W42 was significantly increased in HCC tissues compared with paired adjacent noncancerous tissues (P < 0.001, Figure 1A). Next, qRT-PCR analysis also showed significantly higher expression of lncRNA W42 in three human HCC cell lines (SMMC-7721, HepG2 and Huh7) than in the normal liver cell line LO2 (P < 0.05, Figure 1B). Moreover, we used receiver operating characteristic (ROC) curves to assess the sensitivity and specificity of lncRNA W42 expression to discriminate between tumor and noncancerous samples, the sensitivity was 74.76% and the specificity was 72.82%. It was noticeable that lncRNA W42 had predictive value, with an area under the curve (AUC, denotes discrimination accuracy) of 0.808 (Figure 1C), revealing that lncRNA W42 has adequate sensitivity and specificity to discriminate between HCC tissues and paired adjacent noncancerous tissues. In addition, the nucleocytoplasmic separation experiment showed that lncRNA W42 was mainly located in the nucleus (Figure 1D). In addition, we sequenced full-length lncRNA W42 and found that its length was 1598 nt and that it was located on chromosome 7q: 94423057-94429430 and shares a transcript with the COL1A2 gene (Supplementary Figure 1B and C). Thus, these results suggest that lncRNA W42 may provide imperative clinical guidance in HCC diagnosis and serve a critical role in HCC progression.

https://f6publishing.blob.core.windows.net/af1afa1c-cc39-4751-9bd2-6f2b9a58a166/WJG-27-2586-supplementary-material.pdf




Figure 1 The levels of long noncoding RNA W42 expression in human hepatocellular carcinoma tissues and cell lines. A: Relative expression levels of long noncoding RNA (lncRNA) W42 in human hepatocellular carcinoma (HCC) tissue compared to paired adjacent noncancerous tissues (n = 92). LncRNA W42 expression was examined by quantitative reverse transcriptase polymerase chain reaction and normalized to GAPDH (glyceraldehyde-3-phosphate dehydrogenase) expression (cP < 0.001); B: The level of lncRNA W42 expression was higher in HCC cell lines than in the normal liver cell line LO2 (cP < 0.001); C: The area under the receiver operating characteristic curve was used for prediction of HCC based on lncRNA W42 expression, using paired adjacent noncancerous tissues as a control; D: Quantitative reverse transcriptase polymerase chain reaction analysis of lncRNA W42 in subcellular fractions of SMMC 7721 and Huh7 cells. β-Actin and U6 acted as cytoplasmic and nuclear markers, respectively (n = 3). ROC: Receiver operating characteristic.

LncRNA W42 promoted proliferation and invasion and suppressed early apoptosis of HCC cells in vitroTo further investigate the biological functions of lncRNA W42 in HCC, we constructed a lentiviral vector for stably overexpressing lncRNA W42 in HCC cell lines, i.e., SMMC7721 and Huh7, and designed two shRNAs for W42 (Figure 2A). CCK-8 assays revealed that overexpression of lncRNA W42 in these HCC cell lines significantly enhanced the cell proliferation ability (Figure 2B). Colony formation assays also indicated that the lncRNA W42-transfected HCC cells formed significantly more colonies than the control cells (Figure 2C). Conversely, we inhibited lncRNA W42 expression by transfecting lncRNA W42-specific shRNA into HCC cells. As indicated by CCK-8 assays, the repression of lncRNA W42 significantly decreased cell prolif-eration in HCC cells. Similarly, colony formation assays showed that compared to the control, knockdown of lncRNA W42 inhibited HCC cell colony formation. Moreover, overexpression of cellular lncRNA W42 not only increased HCC cell proliferation but







Figure 2 Long noncoding RNA W42 affects proliferation and invasion in hepatocellular carcinoma cells in vitro. A: Long noncoding RNA (lncRNA) W42 knockdown in SMMC 7721 and Huh7 cells. The expression level of lncRNA W42 was robustly enhanced or inhibited when cell lines were infected with lentivirus containing the lncRNA W42 coding sequence or two shRNAs targeting lncRNA W42, respectively. Empty vector and shRNA control were used as the negative controls; B: CCK-8 assays were used to determine the viability of SMMC7721 and Huh7 cells transfected with shRNA-lncRNA W42. The cell number was examined from 24 h to 96 h. The results are shown as the mean ± SE. aP < 0.05, compared with the control by two-sided t test; C: Colony forming assays were performed to determine the proliferation of SMMC7721 and Huh7 cells transfected with shRNA-lncRNA W42; D: Transwell assays to measure the impact of lncRNA W42 knockdown on HCC migration ability; E: Annexin V-fluorescein isothiocyanate/propidium iodide assay for the effect of overexpression or knockdown of lncRNA W42 on early apoptosis of HCC cell lines. aP < 0.05, bP < 0.01, cP < 0.001. LncRNA: Long noncoding RNA.

also promoted the cell migration and invasion activity (Figure 2D) of HCC cells compared with that in the control group. In addition, we also examined apoptosis when over-expressing lncRNA W42 on HCC cells, the results showed that there was a significant decrease in the early apoptosis of HCC cells (Figure 2E). In addition, no significant difference was observed in the cell cycle analysis between the lncRNA W42 and control group (Supplementary Figure 2). Collectively, these data imply that lncRNA W42 exerts an important role in regulating HCC cell proliferation and invasion in vitro.

LncRNA W42 directly bound to DBN1 and enhanced DBN1-mediated tumor-promoting effectsGiven that lncRNA W42 is mainly located in the nucleus, we further investigated the molecular mechanism of lncRNA W42. RNA pull-down assays and MS analysis were performed in SMMC7721 cells. The results showed that DBN1, which is a key player in the advancement of several cancers, bound lncRNA W42 in SMMC7721 cells more readily than in lncRNA W42 antisense control cells (Figure 3A). This result was further confirmed by a lncRNA W42 RNA pull-down western blot assay (Figure 3B). To validate the interaction between lncRNA W42 and DBN1, we immunoprecipitated DBN1 from SMMC7721 cells and quantified the protein-bound lncRNA W42. Significantly higher expression of lncRNA W42 was detected with endogenous (Figure 3C). Together, these results indicate that lncRNA W42 RNA physically interacts with DBN1.




Figure 3 Long noncoding RNA W42 directly bound to DBN1 and enhanced DBN1-mediated tumor-promoting effects. A: RNA pull-down of AVAN-associated proteins using biotinylated long noncoding RNA (lncRNA) W42 or antisense probes (immunoglobulin G, IgG). Isolated proteins were resolved by SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis) followed by silver staining; B: Pull-down western blot showing that lncRNA W42 can bind directly to DBN1; C: RNA immunoprecipitation of DBN1 in SMMC7721 cells using anti-DBN1 or anti-IgG antibodies. The relative fold enrichment of lncRNA W42 was calculated by quantitative reverse transcriptase polymerase chain reaction. IgG: Immunoglobulin G; LncRNA: Long noncoding RNA; NC: Negative control.

Downregulation of lncRNA W42 inhibited HCC xenograft tumor growth in vivoTo elucidate the in vivo roles of lncRNA W42 in tumorigenesis, we determined whether decreased expression of lncRNA W42 could suppress tumor growth in the HCC xenograft tumor model. SMMC7721 cells with stable knockdown of lncRNA W42 or empty vector, generated by lentiviral vector transduction, were inoculated into nude mice. Thirty-six days after injection, the tumors were dissected from the mice (Supplementary Figure 3). Representative images of dissected tumors are shown in Figure 4A. As illustrated in Figure 4B, we observed that the growth rate of tumors in the lncRNA W42 knockdown group was lower than that in the control group. Six weeks after injection, the weight of tumors dissected from mice in the lncRNA W42 knockdown group was significantly lower than that in their control groups (Figure 4C). Consistent with the in vitro results, the tumor volumes and weights were markedly lower in the lncRNA W42 knockdown group than in the empty vector group, indicating that lncRNA W42 promotes HCC xenograft tumor growth in vivo.

Relationship between lncRNA W42 expression, clinicopathological characteristics and patient prognosisWe further determined whether lncRNA W42 expression was correlated with the clinicopathological characteristics and outcome of HCC patients. HCC patients (n = 92) were divided into a high lncRNA W42 expression group (above the average expression, n = 46) and a low lncRNA W42 expression group (below the average expression, n = 46). As illustrated in Table 1, no obvious differences were observed in age, tumor size, or AFP levels. However, lncRNA W42 expression was closely associated with tumor number, liver cirrhosis and tumor recurrence (P < 0.05).

Subsequently, we determined whether lncRNA W42 expression was associated with the prognosis of HCC patients. All 92 HCC patients included in this experiment were classified into two cohorts with high or low lncRNA W42 expression according to median lncRNA W42 expression values. Kaplan-Meier analysis in the 92 patients with HCC showed that the 3-year OS rate after surgery was 64.13% (P = 0.022; Figure 5A). Specifically, a higher lncRNA W42 expression level in HCC tissues was significantly associated with a reduction in OS (P = 0.005; Figure 5B) and RFS (P = 0.008; Figure 5C) during the 3 year follow up period, supporting the critical roles of lncRNA W42 in the pathogenesis of HCC.

DISCUSSIONWith the advances of next-generation sequencing technology, lncRNAs have been reported to be aberrantly expressed in multiple types of cancers[33-35]. A number of lncRNAs have been validated to be critical regulators of biological processes in




Table 1 The relationship between long noncoding RNA W42 expression and clinicopathological features in hepatocellular carcinoma patients

W42 expressionParameters Group Total

Low HighP value

Gender Male 72 35 37 0.810

Female 20 11 9

Age (yr) < 60 27 15 12 0.681

≥ 60 65 31 34

Tumor size (cm) < 5 52 25 27 0.846

≥ 5 40 21 19

AFP < 20 27 11 16 0.411

≥ 20 65 35 30

Histological grade Well/moderately 14 8 6 0.776

Poorly 78 38 40

Clinical stage I and II 69 31 38 0.182

III 23 15 8

Tumor number Solitary 69 49 20 0.009b

Multiple 23 6 17

Drinking status Yes 34 19 23 0.999

No 58 31 27

Smoking status Yes 48 22 26 0.563

No 44 24 20

HBV Yes 66 30 36 0.419

No 26 14 12

Recurrence Yes 29 9 20 0.016a

No 79 49 30

PVTT Yes 42 22 20 0.847

No 50 24 26

Microvascular invasion Yes 73 37 35 0.999

No 19 9 10

Liver cirrhosis Absence 45 16 29 0.030a

Presence 47 30 17

aP < 0.05.bP < 0.01. AFP: α-Fetoprotein; HBV: Hepatitis B virus; PVTT: Portal vein tumor thrombosis.

cancer[36]. In particular, several HCC hot lncRNAs, such as HULC[37], P53[38], MCM3AP-AS1[39], ATB[40], HOTAIR[14] and lnc00624[41], have been reported, and increasingly evidence showed that further investigation on the relationship between lncRNA and HCC might provide potential targets for targeted therapy of HCC. Herein, we revealed differential expression of lncRNA W42 using qRT-PCR analysis of HCC and paired adjacent noncancerous tissues. Our results showed that lncRNA W42 was upregulated in HCC tissues compared to their normal counterparts. The ROC AUC of lncRNA W42 was 0.808, suggesting its specificity and sensitivity in the diagnosis of HCC. We also identified the function of lncRNA W42 in HCC cells by applying gain-of-function and loss-of-function approaches. Overexpression of lncRNA W42 promoted the proliferation and invasion and suppressed early apoptosis of HCC cells. However, this phenomenon was reversed when lncRNA W42 was downreg-ulated.



Figure 4 Effects of long noncoding RNA W42 expression on tumor growth in hepatocellular carcinoma nude mouse model. A: SMMC7721 cells (5 × 106) stably expressing long noncoding RNA (lncRNA) W42 were inoculated into hepatocellular carcinoma (HCC) nude mice, and the effect of lncRNA W42 on HCC tumor growth was examined every 3 d for 36 d (n = 8). Thirty-six days after injection, tumors were dissected from the mice in the lncRNA W42 knockdown and control groups. Representative images are shown; B: The growth rate of tumors in the lncRNA W42 knockdown group was lower than that in the control group. mean ± SE for eight mice in each group. cP < 0.001 compared with the control group; C: The weights of tumors dissected from mice in the lncRNA W42 knockdown group were significantly lower than those of their control counterparts. The results are shown as the mean ± SE for eight mice in each group. bP < 0.01 compared with control cells by two-sided t test. LncRNA: Long noncoding RNA.

DBN1 is an actin-binding protein that was initially found in embryonic chicken brains, and it was identified to be a neuron-specific protein[42]. However, several studies have shown the presence of DBN1 in various nonneuronal tissues, even in tumor tissues, such as recurrent non-small-cell lung cancer[43] and lymphoblastic leukemia[44] tissues. Herein, we showed that the overexpression of lncRNA W42 and DBN1 may be involved in the progression of HCC. Our results also showed that lncRNA W42 RNA physically interacts with DBN1 and that lncRNA W42 might directly bind to DBN1 and enhance the DBN1-mediated tumor-promoting effects. In addition, the mechanism by which lncRNA W42 regulates DBN1 warrants further investigation in subsequent studies.

Moreover, in vivo experiments revealed that lncRNA W42 knockdown inhibited tumor growth in nude mice, as indicated by tumor volume and weight. Subsequently, we will conduct in vivo experiments to analyze the effects of lncRNA W42 overex-pression on liver tumors in nude mice. A clinical investigation showed that lncRNA W42 upregulation was closely correlated with tumor number, liver cirrhosis and tumor recurrence. More importantly, we observed that upregulated lncRNA W42 expression was closely associated with decreased OS rates and RFS rates. Based on these data, we propose that lncRNA W42 may play an important role in the develop-ment and progression of HCC.

Several previous studies have shown that liver cirrhosis is a predictor of and an independent risk factor for HCC recurrence and that patients with liver cirrhosis tend to express different levels lncRNAs[45-47]. Patients with liver cirrhosis usually have a higher probability of tumor development once a virus infection occurs because they are resistant to direct-acting antiviral agents. HCC develops frequently in the advanced fibrosis stage, and eradicating HBV or HCV infection has been a promising prophylactic therapy for preventing the occurrence of liver fibrosis and HCC. To date, several lncRNAs have been validated to be involved in HCC progression, but the role of lncRNAs in cirrhosis has not been well explained. On the other hand, lncRNA HULC is upregulated in the plasma samples of patients with HBV-related cirrho-sis[48]. In line with these findings, we also noted that high expression of lncRNA W42 in patients was associated with cirrhosis, suggesting that the lncRNA profiles may be changed in HCC patients with cirrhosis. Furthermore, factors different from carcino-genesis or genetic and epigenetic factors are related to the flow from chronic hepatitis



Figure 5 Association between long noncoding RNA W42 expression and hepatocellular carcinoma patient survival rates as determined by Kaplan-Meier analyses. A: The three-year overall survival rate of 92 hepatocellular carcinoma (HCC) patients was 64.13%; B: Patients with HCC with high long noncoding RNA (lncRNA) W42 expression levels showed significantly shorter overall survival than those with low lncRNA W42 expression levels (P = 0.005, log-rank test); C: Patients with HCC with high lncRNA W42 expression levels showed significantly shorter recurrence-free survival than those with low lncRNA W42 expression levels (P = 0.008, log-rank test).

to liver cirrhosis or from focal HCC carcinogenesis in the liver to multi-organ metastasis. As shown in Table 1, we observed that lncRNA W42 expression was closely associated with liver cirrhosis and tumor recurrence, but no obvious differences were observed in age, gender, tumor size, HBV and AFP levels. Of course, the relationship between lncRNA W42 and the etiology of HCC (HCV, HBV, NAFLD, alcohol, etc.) needs to be validated in a much larger cohort. Moreover, further study will explore the levels of lncRNA W42 in serum and the AUC values of AFP in tumor tissue to ascertain if lncRNA W42 provides a higher diagnostic capacity compared to AFP in HCC patients.

Several studies have shown that microRNA (miRNA) and circulating tumor DNA detected in blood and body fluids such as urine, are useful for screening the presence of various cancers[49,50]. Similarly, the abundant mRNA, circular RNA, and lncRNA in human blood could be utilized as potential biomarkers for the diagnosis of various cancers, which were revealed by extracellular vesicles long RNA sequencing[51]. However, whether lncRNA W42 could be detected in peripheral blood needs to be investigated in subsequent studies.

These results further support that lncRNA W42 might participate in the prolif-eration and apoptosis regulation of HCC cells. Of course, the proliferation results need to be confirmed by flow cytometry (e.g., CFSE assay) in subsequent studies. In addition, the effect and underlying molecular mechanism of lncRNA W42 in HCC progression and potential lncRNA W42 targets, for example, the modulation of lncRNA W42 impacts on the response to chemo-or-targeted therapies, remain to be investigated in the future.

CONCLUSIONIn conclusion, our results demonstrated, for the first time, that lncRNA W42 expression was significantly higher in HCC tissues than in normal tissues and that dysregulation of lncRNA W42 was positively correlated with tumor number, liver cirrhosis and tumor recurrence in patients with HCC. Moreover, increased lncRNA



W42 expression was associated with a poor prognosis in patients with HCC. These findings suggest that lncRNA W42 is an important marker for indicting prognosis and may have potential as a diagnostic and therapeutic target for HCC.

ARTICLE HIGHLIGHTSResearch backgroundHepatocellular carcinoma (HCC) is a malignancy found globally.

Research motivationThe function of long noncoding RNAs (lncRNAs) in HCC requires further invest-igation.

Research objectivesWe aimed to understand the effect of lncRNA W42 on HCC and dissect the underlying molecular mechanisms.

Research methodsLncRNA W42 expression in HCC tissues and cells (Huh7 and SMMC-7721) was detected by quantitative reverse transcriptase polymerase chain reaction. After transfection with pcDNA3.1-lncRNA W42 or shRNA-lncRNA W42, cell functions were assessed by cell counting Kit-8 (CCK-8), colony formation, flow cytometry and Transwell assays. An HCC xenograft model was used to assess the role of lncRNA W42 on tumor growth in vivo. Receiver operating characteristic curves and Kaplan-Meier curves were used for clinical investigation.

Research resultsLncRNA W42 expression was upregulated in HCC tissues and cells. LncRNA W42 directly bound to DBN1 and activated the downstream pathway to promote cell prolif-eration, and invasion of HCC. LncRNA W42 knockdown suppressed HCC xenograft tumor growth in vivo. HCC patients with high lncRNA W42 expression exhibited shorter survival times.

Research conclusionsUpregulation of lncRNA W42 promotes tumor development by binding with DBN1 in HCC.

Research perspectivesLncRNA W42, which is upregulated in HCC, may serve as a potential candidate prognostic biomarker and therapeutic target in HCC.

REFERENCESTorre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin 2015; 65: 87-108 [PMID: 25651787 DOI: 10.3322/caac.21262]

1

Wang H, Huo X, Yang XR, He J, Cheng L, Wang N, Deng X, Jin H, Wang C, Zhao F, Fang J, Yao M, Fan J, Qin W. STAT3-mediated upregulation of lncRNA HOXD-AS1 as a ceRNA facilitates liver cancer metastasis by regulating SOX4. Mol Cancer 2017; 16: 136 [PMID: 28810927 DOI: 10.1186/s12943-017-0680-1]

2

Sartorius K, Sartorius B, Aldous C, Govender PS, Madiba TE. Global and country underestimation of hepatocellular carcinoma (HCC) in 2012 and its implications. Cancer Epidemiol 2015; 39: 284-290 [PMID: 25922178 DOI: 10.1016/j.canep.2015.04.006]

3

Nault JC, Datta S, Imbeaud S, Franconi A, Mallet M, Couchy G, Letouzé E, Pilati C, Verret B, Blanc JF, Balabaud C, Calderaro J, Laurent A, Letexier M, Bioulac-Sage P, Calvo F, Zucman-Rossi J. Recurrent AAV2-related insertional mutagenesis in human hepatocellular carcinomas. Nat Genet 2015; 47: 1187-1193 [PMID: 26301494 DOI: 10.1038/ng.3389]

4

Rogacki K, Kasprzak A, Stępiński A. Alterations of Wnt/β-catenin signaling pathway in hepatocellular carcinomas associated with hepatitis C virus. Pol J Pathol 2015; 66: 9-21 [PMID: 26017875 DOI: 10.5114/pjp.2015.51148]

5

Zucman-Rossi J, Villanueva A, Nault JC, Llovet JM. Genetic Landscape and Biomarkers of Hepatocellular Carcinoma. Gastroenterology 2015; 149: 1226-1239. e4 [PMID: 26099527 DOI: 10.1053/j.gastro.2015.05.061]

6




https://dx.doi.org/10.1186/s12943-017-0680-1


https://dx.doi.org/10.1016/j.canep.2015.04.006


https://dx.doi.org/10.1038/ng.3389


https://dx.doi.org/10.5114/pjp.2015.51148





Ding SL, Yang ZW, Wang J, Zhang XL, Chen XM, Lu FM. Integrative analysis of aberrant Wnt signaling in hepatitis B virus-related hepatocellular carcinoma. World J Gastroenterol 2015; 21: 6317-6328 [PMID: 26034368 DOI: 10.3748/wjg.v21.i20.6317]

7

Yao S, Johnson C, Hu Q, Yan L, Liu B, Ambrosone CB, Wang J, Liu S. Differences in somatic mutation landscape of hepatocellular carcinoma in Asian American and European American populations. Oncotarget 2016; 7: 40491-40499 [PMID: 27246981 DOI: 10.18632/oncotarget.9636]

8

Li X, Li C, Zhang L, Wu M, Cao K, Jiang F, Chen D, Li N, Li W. The significance of exosomes in the development and treatment of hepatocellular carcinoma. Mol Cancer 2020; 19: 1 [PMID: 31901224 DOI: 10.1186/s12943-019-1085-0]

9

Chen B, Garmire L, Calvisi DF, Chua MS, Kelley RK, Chen X. Harnessing big 'omics' data and AI for drug discovery in hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol 2020; 17: 238-251 [PMID: 31900465 DOI: 10.1038/s41575-019-0240-9]

10

Vogel A, Saborowski A. Current strategies for the treatment of intermediate and advanced hepatocellular carcinoma. Cancer Treat Rev 2020; 82: 101946 [PMID: 31830641 DOI: 10.1016/j.ctrv.2019.101946]

11

Wei L, Wang X, Lv L, Liu J, Xing H, Song Y, Xie M, Lei T, Zhang N, Yang M. The emerging role of microRNAs and long noncoding RNAs in drug resistance of hepatocellular carcinoma. Mol Cancer 2019; 18: 147 [PMID: 31651347 DOI: 10.1186/s12943-019-1086-z]

12

Lim LJ, Wong SYS, Huang F, Lim S, Chong SS, Ooi LL, Kon OL, Lee CG. Roles and Regulation of Long Noncoding RNAs in Hepatocellular Carcinoma. Cancer Res 2019; 79: 5131-5139 [PMID: 31337653 DOI: 10.1158/0008-5472.CAN-19-0255]

13

Wong CM, Tsang FH, Ng IO. Non-coding RNAs in hepatocellular carcinoma: molecular functions and pathological implications. Nat Rev Gastroenterol Hepatol 2018; 15: 137-151 [PMID: 29317776 DOI: 10.1038/nrgastro.2017.169]

14

Dhanoa JK, Sethi RS, Verma R, Arora JS, Mukhopadhyay CS. Long non-coding RNA: its evolutionary relics and biological implications in mammals: a review. J Anim Sci Technol 2018; 60: 25 [PMID: 30386629 DOI: 10.1186/s40781-018-0183-7]

15

Chen X, Wang S. LncRNA-ZFAS1 contributes to colon cancer progression through the miR-150-5p/VEGFA axis. Ann Oncol 2018; 29: v52-v53 [DOI: 10.1093/annonc/mdy151.187]

16

Kirchhoff T, Simpson D, Hekal T, Ferguson R, Kazlow E, Moran U, Lee Y, Izsak A, Wilson MA, Shapiro R, Pavlick A, Osman I. Discovery of novel germline genetic biomarkers of melanoma recurrence impacting exonic and long non-coding RNA (lncRNA) transcripts. Ann Oncol 2018; 29 (suppl_8): viii442-viii466 [DOI: 10.1093/annonc/mdy289]

17

Zhang J, Han X, Jiang L, Han Z, Wang Z. LncRNA CERNA2 is an independent predictor for clinical prognosis and is related to tumor development in gastric cancer. Int J Clin Exp Pathol 2018; 11: 5783-5791 [PMID: 31949664]

18

Wang W, Chen G, Wang B, Yuan Z, Liu G, Niu B, Chen Y, Zhou S, He J, Xue H. Long non-coding RNA BZRAP1-AS1 silencing suppresses tumor angiogenesis in hepatocellular carcinoma by mediating THBS1 methylation. J Transl Med 2019; 17: 421 [PMID: 31847842 DOI: 10.1186/s12967-019-02145-6]

19

Koo JI, Lee HJ, Jung JH, Im E, Kim JH, Shin N, Sim DY, Hwang J, Kim SH. The Pivotal Role of Long Noncoding RNA RAB5IF in the Proliferation of Hepatocellular Carcinoma Via LGR5 Mediated β-Catenin and c-Myc Signaling. Biomolecules 2019; 9 [PMID: 31717443 DOI: 10.3390/biom9110718]

20

Dai W, Dai JL, Tang MH, Ye MS, Fang S. lncRNA-SNHG15 accelerates the development of hepatocellular carcinoma by targeting miR-490-3p/ histone deacetylase 2 axis. World J Gastroenterol 2019; 25: 5789-5799 [PMID: 31636472 DOI: 10.3748/wjg.v25.i38.5789]

21

Pu J, Wang J, Wei H, Lu T, Wu X, Wu Y, Shao Z, Luo C, Lu Y. lncRNA MAGI2-AS3 Prevents the Development of HCC via Recruiting KDM1A and Promoting H3K4me2 Demethylation of the RACGAP1 Promoter. Mol Ther Nucleic Acids 2019; 18: 351-362 [PMID: 31629962 DOI: 10.1016/j.omtn.2019.08.020]

22

Wang CZ, Yan GX, Dong DS, Xin H, Liu ZY. LncRNA-ATB promotes autophagy by activating Yes-associated protein and inducing autophagy-related protein 5 expression in hepatocellular carcinoma. World J Gastroenterol 2019; 25: 5310-5322 [PMID: 31558875 DOI: 10.3748/wjg.v25.i35.5310]

23

Cheng S, Li T, Wang C, Wang K, Lai C, Yan J, Fan H, Sun F, Wang Z, Zhang P, Yu L, Hong Z, Lei G, Sun B, Gao Y, Xiao Z, Ji X, Wang R, Wu J, Wang X, Zhang S, Yang P. Decreased long intergenic noncoding RNA P7 predicts unfavorable prognosis and promotes tumor proliferation via the modulation of the STAT1-MAPK pathway in hepatocellular carcinoma. Oncotarget 2018; 9: 36057-36066 [PMID: 30546827 DOI: 10.18632/oncotarget.23282]

24

Lei GL, Fan HX, Wang C, Niu Y, Li TL, Yu LX, Hong ZX, Yan J, Wang XL, Zhang SG, Ren MJ, Yang PH. Long non-coding ribonucleic acid W5 inhibits progression and predicts favorable prognosis in hepatocellular carcinoma. World J Gastroenterol 2021; 27: 55-68 [PMID: 33505150 DOI: 10.3748/wjg.v27.i1.55]

25

Dong RC. Establishment of SMMC-7721 human liver cancer cell line and preliminary observation of its biological characteristics. J Sec Milit Med Univ 1980; 1: 5-9

26

Wang Y, Fu J, Jiang M, Zhang X, Cheng L, Xu X, Fan Z, Zhang J, Ye Q, Song H. MiR-410 is overexpressed in liver and colorectal tumors and enhances tumor cell growth by silencing FHL1 via a direct/indirect mechanism. PLoS One 2014; 9: e108708 [PMID: 25272045 DOI:

27






https://dx.doi.org/10.1186/s12943-019-1085-0


https://dx.doi.org/10.1038/s41575-019-0240-9


https://dx.doi.org/10.1016/j.ctrv.2019.101946






https://dx.doi.org/10.1038/nrgastro.2017.169


https://dx.doi.org/10.1186/s40781-018-0183-7

https://dx.doi.org/10.1093/annonc/mdy151.187

https://dx.doi.org/10.1093/annonc/mdy289



https://dx.doi.org/10.1186/s12967-019-02145-6


https://dx.doi.org/10.3390/biom9110718




https://dx.doi.org/10.1016/j.omtn.2019.08.020










10.1371/journal.pone.0108708]Fu J, Cheng L, Wang Y, Yuan P, Xu X, Ding L, Zhang H, Jiang K, Song H, Chen Z, Ye Q. The RNA-binding protein RBPMS1 represses AP-1 signaling and regulates breast cancer cell proliferation and migration. Biochim Biophys Acta 2015; 1853: 1-13 [PMID: 25281386 DOI: 10.1016/j.bbamcr.2014.09.022]

28

Xiu B, Chi Y, Liu L, Chi W, Zhang Q, Chen J, Guo R, Si J, Li L, Xue J, Shao ZM, Wu ZH, Huang S, Wu J. LINC02273 drives breast cancer metastasis by epigenetically increasing AGR2 transcription. Mol Cancer 2019; 18: 187 [PMID: 31856843 DOI: 10.1186/s12943-019-1115-y]

29

Li D, Liu X, Zhou J, Hu J, Zhang D, Liu J, Qiao Y, Zhan Q. Long noncoding RNA HULC modulates the phosphorylation of YB-1 through serving as a scaffold of extracellular signal-regulated kinase and YB-1 to enhance hepatocarcinogenesis. Hepatology 2017; 65: 1612-1627 [PMID: 28027578 DOI: 10.1002/hep.29010]

30

Yan X, Zhang D, Wu W, Wu S, Qian J, Hao Y, Yan F, Zhu P, Wu J, Huang G, Huang Y, Luo J, Liu X, Liu B, Chen X, Du Y, Chen R, Fan Z. Mesenchymal Stem Cells Promote Hepatocarcinogenesis via lncRNA-MUF Interaction with ANXA2 and miR-34a. Cancer Res 2017; 77: 6704-6716 [PMID: 28947421 DOI: 10.1158/0008-5472.CAN-17-1915]

31

Li O, Li Z, Tang Q, Li Y, Yuan S, Shen Y, Zhang Z, Li N, Chu K, Lei G. Long Stress Induced Non-Coding Transcripts 5 (LSINCT5) Promotes Hepatocellular Carcinoma Progression Through Interaction with High-Mobility Group AT-hook 2 and MiR-4516. Med Sci Monit 2018; 24: 8510-8523 [PMID: 30472720 DOI: 10.12659/MSM.911179]

32

An HX, Xu B, Wang Q, Li YS, Shen LF, Li SG. Up-regulation of long non-coding RNA SNHG6 predicts poor prognosis in renal cell carcinoma. Eur Rev Med Pharmacol Sci 2018; 22: 8624-8629 [PMID: 30575902 DOI: 10.26355/eurrev_201812_16626]

33

Wang D, Hu Y. Long Non-coding RNA PVT1 Competitively Binds MicroRNA-424-5p to Regulate CARM1 in Radiosensitivity of Non-Small-Cell Lung Cancer. Mol Ther Nucleic Acids 2019; 16: 130-140 [PMID: 30861415 DOI: 10.1016/j.omtn.2018.12.006]

34

Li L, Peng M, Xue W, Fan Z, Wang T, Lian J, Zhai Y, Lian W, Qin D, Zhao J. Integrated analysis of dysregulated long non-coding RNAs/microRNAs/mRNAs in metastasis of lung adenocarcinoma. J Transl Med 2018; 16: 372 [PMID: 30587197 DOI: 10.1186/s12967-018-1732-z]

35

Liu L, Zhou XY, Zhang JQ, Wang GG, He J, Chen YY, Huang C, Li L, Li SQ. LncRNA HULC promotes non-small cell lung cancer cell proliferation and inhibits the apoptosis by up-regulating sphingosine kinase 1 (SPHK1) and its downstream PI3K/Akt pathway. Eur Rev Med Pharmacol Sci 2018; 22: 8722-8730 [PMID: 30575912 DOI: 10.26355/eurrev_201812_16637]

36

Xin X, Wu M, Meng Q, Wang C, Lu Y, Yang Y, Li X, Zheng Q, Pu H, Gui X, Li T, Li J, Jia S, Lu D. Long noncoding RNA HULC accelerates liver cancer by inhibiting PTEN via autophagy cooperation to miR15a. Mol Cancer 2018; 17: 94 [PMID: 29895332 DOI: 10.1186/s12943-018-0843-8]

37

Qin G, Tu X, Li H, Cao P, Chen X, Song J, Han H, Li Y, Guo B, Yang L, Yan P, Li P, Gao C, Zhang J, Yang Y, Zheng J, Ju HQ, Lu L, Wang X, Yu C, Sun Y, Xing B, Ji H, Lin D, He F, Zhou G. Long Noncoding RNA p53-Stabilizing and Activating RNA Promotes p53 Signaling by Inhibiting Heterogeneous Nuclear Ribonucleoprotein K deSUMOylation and Suppresses Hepatocellular Carcinoma. Hepatology 2020; 71: 112-129 [PMID: 31148184 DOI: 10.1002/hep.30793]

38

Wang Y, Yang L, Chen T, Liu X, Guo Y, Zhu Q, Tong X, Yang W, Xu Q, Huang D, Tu K. A novel lncRNA MCM3AP-AS1 promotes the growth of hepatocellular carcinoma by targeting miR-194-5p/FOXA1 axis. Mol Cancer 2019; 18: 28 [PMID: 30782188 DOI: 10.1186/s12943-019-0957-7]

39

Li W, Kang Y. A new Lnc in metastasis: long noncoding RNA mediates the prometastatic functions of TGF-β. Cancer Cell 2014; 25: 557-559 [PMID: 24823634 DOI: 10.1016/j.ccr.2014.04.014]

40

Li Z, Lu X, Liu Y, Zhao J, Ma S, Yin H, Huang S, Zhao Y, He X. Gain of LINC00624 Enhances Liver Cancer Progression by Disrupting the Histone Deacetylase 6/Tripartite Motif Containing 28/Zinc Finger Protein 354C Corepressor Complex. Hepatology 2020 [PMID: 32869873 DOI: 10.1002/hep.31530]

41

Lin Q, Tan HT, Lim TK, Khoo A, Lim KH, Chung MC. iTRAQ analysis of colorectal cancer cell lines suggests Drebrin (DBN1) is overexpressed during liver metastasis. Proteomics 2014; 14: 1434-1443 [PMID: 24610677 DOI: 10.1002/pmic.201300462]

42

Mitra R, Lee J, Jo J, Milani M, McClintick JN, Edenberg HJ, Kesler KA, Rieger KM, Badve S, Cummings OW, Mohiuddin A, Thomas DG, Luo X, Juliar BE, Li L, Mesaros C, Blair IA, Srirangam A, Kratzke RA, McDonald CJ, Kim J, Potter DA. Prediction of postoperative recurrence-free survival in non-small cell lung cancer by using an internationally validated gene expression model. Clin Cancer Res 2011; 17: 2934-2946 [PMID: 21242119 DOI: 10.1158/1078-0432.CCR-10-1803]

43

Vaskova M, Kovac M, Volna P, Angelisova P, Mejstrikova E, Zuna J, Brdicka T, Hrusak O. High expression of cytoskeletal protein drebrin in TEL/AML1pos B-cell precursor acute lymphoblastic leukemia identified by a novel monoclonal antibody. Leuk Res 2011; 35: 1111-1113 [PMID: 21497902 DOI: 10.1016/j.leukres.2011.03.028]

44

Zhang K, Shi Z, Zhang M, Dong X, Zheng L, Li G, Han X, Yao Z, Han T, Hong W. Silencing lncRNA Lfar1 alleviates the classical activation and pyoptosis of macrophage in hepatic fibrosis. Cell Death Dis 2020; 11: 132 [PMID: 32071306 DOI: 10.1038/s41419-020-2323-5]

45

Li LJ, Wu XY, Tan SW, Xie ZJ, Pan XM, Pan SW, Bai WR, Li HJ, Liu HL, Jiang J, Wu B. Lnc-TCL6 is a potential biomarker for early diagnosis and grade in liver-cirrhosis patients. Gastroenterol Rep (Oxf) 2019; 7: 434-443 [PMID: 31857905 DOI: 10.1093/gastro/goz050]

46

Zhang K, Zhang M, Yao Q, Han X, Zhao Y, Zheng L, Li G, Liu Q, Chang Y, Zhang P, Cui H, Shi Z, 47



https://dx.doi.org/10.1016/j.bbamcr.2014.09.022








https://dx.doi.org/10.12659/MSM.911179


https://dx.doi.org/10.26355/eurrev_201812_16626


https://dx.doi.org/10.1016/j.omtn.2018.12.006




https://dx.doi.org/10.26355/eurrev_201812_16637


https://dx.doi.org/10.1186/s12943-018-0843-8




https://dx.doi.org/10.1186/s12943-019-0957-7


https://dx.doi.org/10.1016/j.ccr.2014.04.014




https://dx.doi.org/10.1002/pmic.201300462




https://dx.doi.org/10.1016/j.leukres.2011.03.028


https://dx.doi.org/10.1038/s41419-020-2323-5


https://dx.doi.org/10.1093/gastro/goz050



Chen T, Yao Z, Han T, Hong W. The hepatocyte-specifically expressed lnc-HSER alleviates hepatic fibrosis by inhibiting hepatocyte apoptosis and epithelial-mesenchymal transition. Theranostics 2019; 9: 7566-7582 [PMID: 31695787 DOI: 10.7150/thno.36942]Zhang Y, Li Z, Zhang Y, Zhong Q, Chen Q, Zhang L. Molecular mechanism of HEIH and HULC in the proliferation and invasion of hepatoma cells. Int J Clin Exp Med 2015; 8: 12956-12962 [PMID: 26550214]

48

He Q, Fang Y, Lu F, Pan J, Wang L, Gong W, Fei F, Cui J, Zhong J, Hu R, Liang M, Fang L, Wang H, Yu M, Zhang ZF. Analysis of differential expression profile of miRNA in peripheral blood of patients with lung cancer. J Clin Lab Anal 2019; 33: e23003 [PMID: 31541491 DOI: 10.1002/jcla.23003]

49

Patel H, Okamura R, Fanta P, Patel C, Lanman RB, Raymond VM, Kato S, Kurzrock R. Clinical correlates of blood-derived circulating tumor DNA in pancreatic cancer. J Hematol Oncol 2019; 12: 130 [PMID: 31801585 DOI: 10.1186/s13045-019-0824-4]

50

Li Y, Zhao J, Yu S, Wang Z, He X, Su Y, Guo T, Sheng H, Chen J, Zheng Q, Li Y, Guo W, Cai X, Shi G, Wu J, Wang L, Wang P, Huang S. Extracellular Vesicles Long RNA Sequencing Reveals Abundant mRNA, circRNA, and lncRNA in Human Blood as Potential Biomarkers for Cancer Diagnosis. Clin Chem 2019; 65: 798-808 [PMID: 30914410 DOI: 10.1373/clinchem.2018.301291]

51


https://dx.doi.org/10.7150/thno.36942



https://dx.doi.org/10.1002/jcla.23003


https://dx.doi.org/10.1186/s13045-019-0824-4


https://dx.doi.org/10.1373/clinchem.2018.301291


World Journal of



ORIGINAL ARTICLE


Understanding celiac disease monitoring patterns and outcomes after diagnosis: A multinational, retrospective chart review study

Knut EA Lundin, Ciaran P Kelly, David S Sanders, Kristina Chen, Sheena Kayaniyil, Sisi Wang, Rajvi J Wani, Caitlin Barrett, Shakira Yoosuf, Ellen S Pettersen, Robert Sambrook, Daniel A Leffler

ORCID number: Knut EA Lundin 0000-0003-1713-5545; Ciaran P Kelly 0000-0002-8216-1240; David S Sanders 0000-0002-3739-0124; Kristina Chen 0000-0003-3428-9570; Sheena Kayaniyil 0000-0003-0239-9580; Sisi Wang 0000-0002-7845-4817; Rajvi J Wani 0000-0002-2422-7758; Caitlin Barrett 0000-0001-7928-0890; Shakira Yoosuf 0000-0002-8779-7694; Ellen S Pettersen 0000-0002-4469-8557; Robert Sambrook 0000-0001-7317-6830; Daniel A Leffler 0000-0002-0750-3386.

Author contributions: Lundin KE, Kelly CP, and Sanders DS provided data acquisition and interpretation and contributed equally to the study; Chen K provided data interpretation; Kayaniyil S, Wang S, Sambrook R and Leffler DA were responsible for design of the study, and provided data acquisition, analysis, and interpretation; Wani RJ provided data analysis and interpretation; Barrett C, Yoosuf S and Pettersen ES provided data acquisition and interpretation; all authors contributed to the development of the manuscript.

Institutional review board statement: Ethics approval was obtained before data collection commenced at each site (Beth Israel Deaconess Medical Center

Knut EA Lundin, K.G. Jebsen Coeliac Disease Research Centre, University of Oslo, Oslo 0450, Norway

Knut EA Lundin, Ellen S Pettersen, Department of Gastroenterology, Oslo University Hospital Rikshospitalet, Oslo 0372, Norway

Ciaran P Kelly, Caitlin Barrett, Shakira Yoosuf, Celiac Center Beth Israel Deaconess Medical Center, Celiac Research Program Harvard Medical School, Boston, MA 02115, United States

David S Sanders, Royal Hallamshire Hospital, University of Sheffield, Sheffield S10 2TN, United Kingdom

Kristina Chen, Daniel A Leffler, Takeda Pharmaceuticals International Co., Cambridge, MA 02139, United States

Sheena Kayaniyil, Rajvi J Wani, Real World Evidence Strategy and Analytics, ICON plc., Toronto, ON L7N 3G2, Canada

Sisi Wang, Robert Sambrook, Real World Evidence Strategy and Analytics, ICON plc., Vancouver, BC V6B 1P1, Canada

Corresponding author: Daniel A Leffler, MD, Doctor, Takeda Pharmaceuticals International Co., 40 Landsdowne Street, Cambridge, MA 02139, United States. [email protected]

AbstractBACKGROUND Long-term outcomes and monitoring patterns in real-world practice are largely unknown among patients with celiac disease.

AIM To understand patterns of follow-up and management of patients with celiac disease, and to characterize symptoms and villous atrophy after diagnosis.

METHODS A retrospective chart review study was performed using medical chart data of patients diagnosed with celiac disease. Three gastroenterology referral centers, with substantial expertise in celiac disease, participated in the United Kingdom, United States, and Norway. Demographic and clinical data were collected from



http://orcid.org/0000-0003-1713-5545

http://orcid.org/0000-0003-1713-5545

http://orcid.org/0000-0002-8216-1240

http://orcid.org/0000-0002-8216-1240

http://orcid.org/0000-0002-3739-0124

http://orcid.org/0000-0002-3739-0124

http://orcid.org/0000-0003-3428-9570

http://orcid.org/0000-0003-3428-9570

http://orcid.org/0000-0003-0239-9580

http://orcid.org/0000-0003-0239-9580

http://orcid.org/0000-0002-7845-4817

http://orcid.org/0000-0002-7845-4817

http://orcid.org/0000-0002-2422-7758

http://orcid.org/0000-0002-2422-7758

http://orcid.org/0000-0001-7928-0890

http://orcid.org/0000-0001-7928-0890

http://orcid.org/0000-0002-8779-7694

http://orcid.org/0000-0002-8779-7694

http://orcid.org/0000-0002-8779-7694

http://orcid.org/0000-0002-4469-8557

http://orcid.org/0000-0002-4469-8557

http://orcid.org/0000-0002-4469-8557

http://orcid.org/0000-0001-7317-6830

http://orcid.org/0000-0001-7317-6830

http://orcid.org/0000-0002-0750-3386

http://orcid.org/0000-0002-0750-3386


Lundin KEA et al. Multinational chart review in celiac disease


09/11/2018; Health Research Authority 19/11/2018; and Regional Committees for Medical and Health Research Ethics 05/06/2018).

Informed consent statement: A waiver of consent was approved for the United States site; the other sites had collected patient consent for research with their patient database.

Conflict-of-interest statement: KEA Lundin has served as a speaker/consultant/advisory board member for Amyra Biotech AG, Bioniz Therapeutics, Chugai Pharmaceutical, Dr. Falk Pharma GMBH, Immusant Therapeutics, and Interexon Actobiotics. CP Kelly has served as a consultant/advisory board member for Aptalis, Cour Pharma, Glutenostics, ImmunogenX, Innovate, Janssen, Kanyos, Takeda Pharmaceuticals, Merck, and Theravance; CP Kelly has received a grant from Aptalis; S Kayaniyil, S Wang, RJ Wani, and R Sambrook are salaried employees of ICON, which received research funds from Takeda Pharmaceuticals for conducting the study and preparation of the manuscript for publication. K Chen was a salaried employee of Takeda Pharmaceuticals at the time of study. DA Leffler is a salaried employee of Takeda Pharmaceuticals; CP Kelly owns shares in Cour Pharma and Glutenostics; DS Sanders, C Barrett, S Yoosuf, and ES Pettersen have no conflicts of interest to declare.

Data sharing statement: Data are available upon request from the corresponding author.

STROBE statement: The authors have read the STROBE Statement – checklist of items, and the manuscript was prepared and revised according to the STROBE Statement – checklist of items.

Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external

medical charts. Descriptive analyses were conducted on patients with biopsy-confirmed celiac disease, diagnosed between 2008 and 2012, with at least one follow-up visit before December 31, 2017. Patient demographic and clinical characteristics, biopsy/serology tests and results, symptoms, and comorbidities were captured at diagnosis and for each clinic visit occurring within the study period (i.e., before the study end date of December 31, 2017).

RESULTS A total of 300 patients were included in this study [72% female; mean age at diagnosis: 38.9 years, standard deviation (SD) 17.2]. Patients were followed-up for a mean of 29.9 mo (SD 22.1) and there were, on average, three follow-up visits per patient during the study period. Over two-thirds (68.4%) of patients were recorded as having ongoing gastrointestinal symptoms and 11.0% had ongoing symptoms and enteropathy during follow-up. Approximately 80% of patients were referred to a dietician at least once during the follow-up period. Half (50.0%) of the patients underwent at least one follow-up duodenal biopsy and 36.6% had continued villous atrophy. Patterns of monitoring varied between sites. Biopsies were conducted more frequently in Norway and patients in the United States had a longer follow-up duration.

CONCLUSION This real-world study demonstrates variable follow-up of patients with celiac disease despite most patients continuing to have abnormal histology and symptoms after diagnosis.

Key Words: Celiac disease; Outcomes research; Endoscopy; Real-world; General practice; Villous atrophy


Core Tip: Inadequately managed celiac disease can lead to health complications. However, there are limited published data on real-world monitoring and outcomes for patients with celiac disease. In this real-world study, country/site-specific differences in the routine monitoring of patients after diagnosis were evident, including differences in the frequency of follow-up biopsy. A large proportion of patients continued to have villous atrophy and symptoms after diagnosis and, therefore, there is a continued need for more routine follow-up assessments, including mucosal assessments of celiac disease activity.

Citation: Lundin KE, Kelly CP, Sanders DS, Chen K, Kayaniyil S, Wang S, Wani RJ, Barrett C, Yoosuf S, Pettersen ES, Sambrook R, Leffler DA. Understanding celiac disease monitoring patterns and outcomes after diagnosis: A multinational, retrospective chart review study. World J Gastroenterol 2021; 27(20): 2603-2614URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2603.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2603

INTRODUCTIONCeliac disease is a chronic, immune-mediated disorder that affects genetically susceptible individuals. The only accepted current standard of care for celiac disease is a life-long gluten-free diet (GFD). Previous studies have reported that adherence rates to a GFD range between 42% and 91%[1,2]. Inadequately managed celiac disease can lead to health complications such as malnutrition, osteoporosis, neurologic complaints, and lymphoma[2]. It has been hypothesized that long-term management and regular follow-up of patients with celiac disease will improve adherence to a GFD, and improve disease outcomes including mucosal healing and symptom resolution. For this reason, long-term management and regular follow-up of patients with celiac disease are advocated by current practice guidelines[3,4], yet it is unclear how these are actually implemented in practice. It is understood, however, that practice patterns





reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/Licenses/by-nc/4.0/



Country/Territory of origin: Canada

Peer-review report’s scientific quality classificationGrade A (Excellent): 0 Grade B (Very good): B Grade C (Good): C Grade D (Fair): D Grade E (Poor): 0

Received: November 12, 2020 Peer-review started: November 12, 2020 First decision: January 23, 2021 Revised: March 9, 2021 Accepted: April 28, 2021 Article in press: April 28, 2021 Published online: May 28, 2021

P-Reviewer: Aufiero VR, Kroupa R, Vorobjova T S-Editor: Fan JR L-Editor: A P-Editor: Ma YJ

vary widely both between countries and between practices.Given that celiac disease is a chronic disorder, it is important to understand real-

world, long-term outcomes and routine monitoring practices; however, there are few published data in these areas. Therefore, the aims of this multinational study were twofold. First, to understand, in real-world clinical practice, patterns of patient follow-up and management and how these practices vary by country. The second aim was to characterize patient outcomes, specifically related to ongoing symptoms and ongoing villous atrophy after diagnosis. Together, these data may be helpful in informing clinical practice, studies, and interventions aimed at improving celiac disease outcomes, and for quality improvement initiatives.

MATERIALS AND METHODSA retrospective chart review study was conducted using medical chart data of patients diagnosed with celiac disease. Three large gastroenterology centers with substantial expertise in celiac disease participated, capturing patients in the United Kingdom, the United States, and Norway. Each site contributed 100 patients. Ethics approval was obtained before data collection commenced.

Patients were eligible if they had biopsy-confirmed celiac disease[3,5,6], were diagnosed between 2008 and 2012, and had at least one follow-up visit before 31 December 2017. This study period was selected to allow for at least five years of follow-up after diagnosis. Patients were excluded if they had initiated a GFD before receiving a diagnosis of celiac disease.

Using the database of patients at each site, the assigned staff at each center identified eligible patients by first looking at the date of diagnosis. The data abstractor reviewed and identified eligible patients who were diagnosed in December 2012, and then continued review of eligibility for patients consecutively backwards from that date (back to a diagnosis date in 2008). After examining the date of diagnosis, other inclusion/exclusion criteria were assessed to verify patient eligibility for the study. All three sites were explicitly asked to follow the same approach regarding selection of consecutive patients to avoid selection bias. The assigned staff at each site responsible for data abstraction then entered de-identified data for eligible patients into a custom electronic case report form. All data collected were based on the patient’s pre-existing medical record. No direct personal identifiers were attached to the abstracted data. Data describing patient demographic and clinical characteristics, biopsy/serology tests and results, symptoms, and comorbidities were captured at diagnosis and for each clinic visit occurring within the study period (i.e., before the study end date of December 31, 2017).

In terms of diagnostic testing, available serology results were collected, including tissue transglutaminase-immunoglobulin (Ig) A, IgA endomysial antibody, total serum IgA, deamidated gliadin peptide (DGP) IgA, DGP IgG, and DGP IgA-IgG. As not all pathology reports across sites utilized Marsh-Oberhuber classification, a descriptive assessment of biopsy results was recorded as follows: normal, increased intraepithelial lymphocytes only, mild/partial villous atrophy, subtotal villous atrophy, total/ complete atrophy, and other.

AnalysisData are summarized by descriptive statistics [mean, standard deviation (SD), median, and interquartile range for continuous variables, and number and percentage for categorical variables]. Gastrointestinal symptoms and extraintestinal comorbidities/ complications (termed extraintestinal manifestations) are described at diagnosis and during study follow-up.

The presence of symptoms during the follow-up period was characterized specifically for patients who had a symptom at diagnosis and a record of symptoms at least once during follow-up. For each patient, the duration of the follow-up period was calculated as the time from diagnosis to the last follow-up visit within the study period. The mean number of visits per patient and the number of follow-up visits per patient with biopsy data were summarized overall and by country.

Following the classification proposed by Kurien et al[2], subsets of study patients with available symptom (defined as diarrhea, abdominal pain, abdominal distention, poor appetite, weight loss, tiredness/lethargy, brain fog, malabsorption and/or bloating) and biopsy data were grouped into four main disease states at diagnosis and at each follow-up visit: class 1 (no symptoms and normal duodenal histology); class 2 (no symptoms and abnormal duodenal histology); class 3 (symptoms and normal






duodenal histology); class 4 (symptoms and abnormal duodenal histology). This classi-fication provides an intuitive framework for assessing celiac disease outcomes and may help to identify patients with the highest risk of complications. In addition, biopsy results reported as mild/partial/subtotal/total/complete villous atrophy were considered as abnormal histology; all other findings were considered normal for this classification. Those with ‘other’ biopsy findings were excluded in the classification.

Analyses were based on available data. Descriptive statistics were restricted to the subset of patients for whom data were available, with relevant denominator information provided in the results. All analyses were conducted in SAS 9.4.

RESULTSA total of 300 patients with celiac disease were included in this study, comprising 100 patients from each of the three participating gastroenterology referral centers in the United Kingdom, the United States, and Norway. Table 1 presents the demographic and clinical characteristics of included patients at diagnosis.

Patients were, on average, 39 years of age at diagnosis, with 24 patients (8%) less than 18 years of age; there were 216 females in the study (72.0%). The study populations across the three sites were quite similar with respect to age, gender, and ethnicity distributions (Table 1). Gastrointestinal symptoms were the most common reason leading to diagnosis. There was a significantly greater proportion of patients in the United Kingdom (57.0%, n = 57) who presented with extraintestinal manifestations at diagnosis compared with patients in the United States (17.0%) and Norway (17.0%) (P < 0.0001). Nutritional deficiency was the most commonly reported extraintestinal manifestation in the United States and Norway, whereas in the United Kingdom anemia was most frequently documented at diagnosis (Table 2). Almost all (n = 299, 99.7%) patients had an esophagogastroduodenoscopy (EGD) conducted at diagnosis, and two patients (0.7%) had an enteroscopy. Overall, 90.7% (n = 272) of patients had serologic testing concurrently with biopsy, and these findings were similar across patients at the three sites. Biopsy results are presented in Table 1. Serology results at diagnosis and during the follow-up period are presented in Supplemental Table 1.

The types of gastrointestinal and extraintestinal manifestations and associated conditions at diagnosis and during follow-up were similar across sites and are presented in Table 2. At diagnosis, 256 patients (85.3%) and 228 patients (76.0%) had at least one gastrointestinal or extraintestinal manifestation, respectively. The most common symptoms across all sites were diarrhea, abdominal pain and bloating and the most common laboratory findings included nutrient deficiencies, anemia and low bone mineral density. Interestingly, both weight loss and weight gain were more commonly reported in the United States compared to the United Kingdom and Norway. There were 147 patients (49.0%) who presented with diarrhea, 124 (41.3%) who presented with abdominal pain, and 90 (30.0%) who presented with bloating. In addition, 104 patients (34.7%) had documentation of a nutritional deficiency, and 34 patients (11.3%) presented with another autoimmune disease, in addition to celiac disease, at diagnosis. During follow-up, diarrhea [n = 100 (33.3%)], abdominal pain [n = 93 (31.0%)], and bloating [n = 76 (25.3%)] continued to be the most frequently reported gastrointestinal symptoms. Of the 256 patients who had gastrointestinal symptoms at diagnosis, 175 (68.4%) had at least one visit reporting gastrointestinal symptoms during the follow-up period.

The duration of follow-up and average number of follow-up visits for the overall study population and by country are presented in Table 3. Patients were followed up for a mean of 29.9 mo (SD: 22.1) and there were, on average, three follow-up visits per patient during the study period. Patients from the United States site had the longest follow-up duration during the study period (mean: 38.7 mo), compared with the United Kingdom and Norway sites (mean: 26.5 and 24.5 mo, respectively; P < 0.0001). Overall referral patterns to other specialists were captured, indicating that approx-imately 80% of patients were referred to a dietician at least once during the follow-up period. Details on the last-recorded follow-up with the patient indicated that almost half (48%) of all patients had a follow-up appointment scheduled. Some were discharged (approximately 10%) or were referred to another specialist (approximately 19%), otherwise, the last follow-up decision was recorded as ‘unknown’ or ‘other’.

After EGD, bone densitometry was the next most frequently reported procedure during follow-up, performed in 89 patients (29.7%) from the overall study population. Bone densitometry was performed at least once in 45 United States patients (45.0%) during the follow-up period, compared with patients who received this procedure in

https://f6publishing.blob.core.windows.net/2571df79-ffed-4a92-ab5b-da2eeade1b37/WJG-27-2603-supplementary-material.pdf



Table 1 Demographic and clinical characteristics of patients at diagnosis, by country

All patients (n = 300)

United Kingdom patients (n = 100)

United States patients (n = 100)

Norway patients (n = 100)

n (%) n (%) n (%) n (%)Age at diagnosis1

mean (SD) 38.9 (17.2) 42.4 (16.9) 39.1 (14.6) 33.1 (18.6)

Median (IQR) 36 (25.0-50.0) 41 (29.0-55.5) 38.5 (26.5-49.0) 31 (21.0-46.0)

Sex

Male 84 (28.0) 29 (29.0) 24 (24.0) 31 (31.0)

Race/Ethnicity

White/Caucasian 287 (95.7) 94 (94.0) 96 (96.0) 97 (97.0)

Black/African-American/Caribbean 2 (0.7) 1 (1.0) 0 (0.0) 1 (1.0)

Hispanic/Latino 2 (0.7) 0 (0.0) 2 (2.0) 0 (0.0)

Asian 6 (2.0) 4 (4.0) 0 (0.0) 2 (2.0)

Other 1 (0.3) 1 (1.0) 0 (0.0) 0 (0.0)

Unknown 2 (0.7) 0 (0.0) 2 (2.0) 0 (0.0)

Confirmation of diagnosis

Biopsy and serology 272 (90.7) 99 (99.0) 91 (91.0) 82 (82.0)

Biopsy only 28 (9.3) 1 (1.0) 9 (9.0) 18 (18.0)

Reason for diagnosis2

Gastrointestinal symptoms 231 (77.0) 70 (70.0) 80 (80.0) 81 (81.0)

Extraintestinal manifestations3 91 (30.3) 57 (57.0) 17 (17.0) 17 (17.0)

Family history of celiac disease 56 (18.7) 15 (15.0) 18 (18.0) 23 (23.0)

Screening for associated disorders 26 (8.7) 5 (5.0) 6 (6.0) 15 (15.0)

Family history of autoimmune disorders 13 (4.3) 0 (0.0) 3 (3.0) 10 (10.0)

Procedures performed for celiac disease-related symptoms

EGD 299 (99.7) 100 (100.0) 99 (99.0) 100 (100.0)

Bone densitometry 76 (25.3) 58 (58.0) 9 (9.0) 9 (9.0)

Colonoscopy 48 (16.0) 11 (11.0) 33 (33.0) 4 (4.0)

Abdominal imaging 9 (3.0) 2 (2.0) 4 (4.0) 3 (3.0)

Enteroscopy 2 (0.7) 1 (1.0) 1 (1.0) 0 (0.0)

Capsule endoscopy 1 (0.3) 1 (1.0) 0 (0.0) 0 (0.0)

Other 15 (5.0) 2 (2.0) 3 (3.0) 10 (10.0)

None 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)

Biopsy results

Normal 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)

Increased intraepithelial lymphocytes 2 (0.7) 1 (1.0) 1 (1.0) 0 (0.0)

Mild/partial villous atrophy 77 (25.7) 22 (22.0) 36 (36.0) 19 (19.0)

Subtotal villous atrophy 115 (38.3) 31 (31.0) 40 (40.0) 44 (44.0)

Total/complete villous atrophy 103 (34.3) 46 (46.0) 22 (22.0) 35 (35.0)

Other 3 (1.0) 0 (0.0) 1 (1.0) 2 (2.0)

1One patient had date of birth missing and is not included in the calculation.2Overall percentage may be greater than 100%, as more than one option could be indicated.



3Extraintestinal manifestations can be found in Table 2.EGD: Esophagogastroduodenoscopy; IQR: Interquartile range; SD: Standard deviation.

the United Kingdom and Norway [n = 22 (22.0%) for both United Kingdom and Norway patients; P < 0.001]. As this procedure is not performed in the gastroen-terology unit, the results of these tests were not routinely available.

A summary of endoscopies with duodenal biopsy performed during the follow-up period, overall and by country, is also presented in Table 3. Of the 300 patients included in this study, 150 (50.0%) had at least one endoscopy with duodenal biopsy during the follow-up period. Of these 150 patients, 116 (77.3%) had a single follow-up endoscopy with biopsy during the follow-up period and most (14.7%, n = 22/150) of the remaining 34 patients had two follow-up endoscopies. A significantly higher proportion of Norway patients received a follow-up biopsy (82.0%, n = 82) compared with patients in the United Kingdom (42.0%, n = 42) and United States (26.0%, n = 26) (P < 0.0001).

The proportion of patients in the four disease state classes at diagnosis and at last follow-up with available data within the study period are presented in Figure 1. Of patients in classes 2 or 4 at diagnosis (n = 295) and who had a follow-up biopsy (n = 150), 53 (36.6%) continued to have villous atrophy (classes 2 or 4) at their last follow-up visit with biopsy data. The proportions were similar for the United Kingdom, United States, and Norway sites, where 39.0% (n = 16), 40.0% (n = 10), and 34.6% (n = 27) of patients, respectively, remained in classes 2 or 4 based on the last available biopsy data within the study period.

Overall, there were 54 patients who were in class 1 (no symptoms and normal duodenal histology) by the last follow-up visit with biopsy data. Of the patients with data available for the classification at diagnosis and at the last follow-up, the proportion of patients in class 1 during the follow-up period was slightly higher in Norwegian patients [n = 34 (43.6%)] compared with patients from the United Kingdom [n = 12 (29.3%)] and the United States [n = 8 (32.0%)].

DISCUSSIONThis real-world study characterizes patients with celiac disease over time, and provides insight into routine monitoring practices from three large referral centers in the United Kingdom, the United States, and Norway. The majority of patients were female, which is consistent with other reports on the demographics of the celiac disease patient population[7,8]. Patients were followed for a mean of 29.9 mo (median 25 mo) and there were, on average, three follow-up visits per patient. Over two thirds of patients had a documentation of gastrointestinal symptoms during the follow-up period, which may indicate inadequate control of celiac disease despite patients being on a GFD. In addition, the fact that a higher proportion of patients from the United Kingdom site presented with extraintestinal manifestations at diagnosis, compared with patients from the United States and Norway sites, indicates that differences may exist in diagnostic or referral practices between different countries. This is particularly true for the United Kingdom site, which was known to see a greater number of patients with neurological manifestations of celiac disease. It is therefore likely that the differences in extraintestinal manifestations at diagnosis between the countries are due to a combination of referral bias and ascertainment bias at the individual sites, such that some manifestations may be evaluated more frequently at some sites than others.

While the study did collect information on extraintestinal manifestations, including liver abnormalities, it did not specifically assess metabolic disorders of patients with celiac disease. Given that an increased risk of non-alcoholic fatty liver disease in patients with celiac disease on a GFD has been reported[9], it would be valuable for future long-term studies to examine such metabolic disorders in this patient population. Country/site-specific differences were also evident in the routine monitoring of patients after diagnosis. While the United States patients had the longest follow-up duration within the study period, compared with Norwegian and United Kingdom patients, a higher proportion of Norwegian patients received a follow-up biopsy, indicating differences in diagnostic or referral practices across the different sites/countries that may not necessarily be reflective of differences in national guidelines.



Table 2 Presentation of gastrointestinal and extraintestinal manifestations at diagnosis and at follow-up visits, by country

All patients (n = 300)

United Kingdom patients (n = 100)

United States patients (n = 100)

Norway patients (n = 100)

At diagnosis, n (%)

At follow-up1, n (%)

At diagnosis, n (%)


At diagnosis, n (%)


At diagnosis, n (%)


Gastrointestinal manifestations2

None3 44 (14.7) 191 (63.7) 26 (26.0) 69 (69.0) 10 (10.0) 56 (56.0) 8 (8.0) 66 (66.0)

Diarrhea 147 (49.0) 100 (33.3) 44 (44.0) 35 (35.0) 54 (54.0) 45 (45.0) 49 (49.0) 20 (20.0)

Abdominal pain 124 (41.3) 93 (31.0) 26 (26.0) 23 (23.0) 46 (46.0) 45 (45.0) 52 (52.0) 25 (25.0)

Abdominal distension 12 (4.0) 11 (3.7) 0 (0.0) 0 (0.0) 7 (7.0) 9 (9.0) 5 (5.0) 2 (2.0)

Poor appetite 5 (1.7) 6 (2.0) 0 (0.0) 1 (1.0) 2 (2.0) 2 (2.0) 3 (3.0) 3 (3.0)

Constipation 38 (12.7) 47 (15.7) 6 (6.0) 6 (6.0) 15 (15.0) 29 (29.0) 17 (17.0) 12 (12.0)

Weight loss 6 (2.0) 33 (11.0) 0 (0.0) 6 (6.0) 5 (5.0) 17 (17.0) 1 (1.0) 10 (10.0)

Weight gain 44 (14.7) 27 (9.0) 10 (10.0) 0 (0.0) 25 (25.0) 24 (24.0) 9 (9.0) 3 (3.0)

Malabsorption 1 (0.3) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 1 (1.0) 0 (0.0)

Bloating 90 (30.0) 76 (25.3) 26 (26.0) 26 (26.0) 31 (31.0) 31 (31.0) 33 (33.0) 19 (19.0)

Hard stools 4 (1.3) 6 (2.0) 0 (0.0) 0 (0.0) 4 (4.0) 4 (4.0) 0 (0.0) 2 (2.0)

Mouth ulcers 3 (1.0) 2 (0.7) 2 (2.0) 0 (0.0) 0 (0.0) 1 (1.0) 1 (1.0) 1 (1.0)

Extraintestinal manifestations2

None3 72 (24.0) 153 (51.0) 37 (37.0) 71 (71.0) 24 (24.0) 38 (38.0) 11 (11.0) 44 (44.0)

Brain fog 11 (3.7) 10 (3.3) 0 (0.0) 0 (0.0) 10 (10.0) 10 (10.0) 1 (1.0) 0 (0.0)

Nutritional deficiency 104 (34.7) 108 (36.0) 27 (27.0) 27 (27.0) 34 (34.0) 59 (59.0) 43 (43.0) 22 (22.0)

Osteoporosis/osteopenia 21 (7.0) 54 (18.0) 5 (5.0) 16 (16.0) 11 (11.0) 26 (26.0) 5 (5.0) 12 (12.0)

Anemia 52 (17.3) 41 (13.7) 29 (29.0) 19 (19.0) 22 (22.0) 20 (20.0) 1 (1.0) 2 (2.0)

Malignancy 2 (0.7) 5 (1.7) 0 (0.0) 0 (0.0) 2 (2.0) 3 (3.0) 0 (0.0) 2 (2.0)

Cardiovascular disease 8 (2.7) 10 (3.3) 0 (0.0) 0 (0.0) 1 (1.0) 2 (2.0) 7 (7.0) 8 (8.0)

Infertility 4 (1.3) 5 (1.7) 0 (0.0) 2 (2.0) 4 (4.0) 3 (3.0) 0 (0.0) 0 (0.0)

Depression 11 (3.7) 12 (4.0) 1 (1.0) 2 (2.0) 8 (8.0) 8 (8.0) 2 (2.0) 2 (2.0)

Anxiety 8 (2.7) 9 (3.0) 1 (1.0) 0 (0.0) 4 (4.0) 8 (8.0) 3 (3.0) 1 (1.0)

Headaches 11 (3.7) 21 (7.0) 2 (2.0) 2 (2.0) 7 (7.0) 15 (15.0) 2 (2.0) 4 (4.0)

Neuropathy 3 (1.0) 4 (1.3) 0 (0.0) 2 (2.0) 3 (3.0) 2 (2.0) 0 (0.0) 0 (0.0)

Autoimmune disease 34 (11.3) 38 (12.7) 2 (2.0) 2 (2.0) 14 (14.0) 16 (16.0) 18 (18.0) 20 (20.0)

Skin and dental conditions 7 (2.3) 7 (2.3) 0 (0.0) 0 (0.0) 3 (3.0) 4 (4.0) 4 (4.0) 3 (3.0)

Hair loss 5 (1.7) 2 (0.7) 0 (0.0) 0 (0.0) 3 (3.0) 2 (2.0) 2 (2.0) 0 (0.0)

Liver abnormalities4 6 (2.0) 11 (3.7) 1 (1.0) 2 (2.0) 3 (3.0) 6 (6.0) 2 (2.0) 3 (3.0)

Musculoskeletal symptoms 28 (9.3) 38 (12.7) 5 (5.0) 7 (7.0) 9 (9.0) 11 (11.0) 14 (14.0) 20 (20.0)

1Denotes presence at any time point during follow-up.2Total percentage may be greater than 100%.3Denotes no presence for the entire follow-up period.4Includes elevated alanine aminotransferase, aspartate aminotransferase, or alkaline phosphatase levels.

In this study, half of patients received at least one follow-up biopsy after diagnosis within the study period, with significant variability between sites. While there is currently no consistent recommendation to perform routine follow-up biopsy on all patients, recent European guidelines suggest a follow-up biopsy in adults one to two



Table 3 Descriptive characteristics of follow-up visits and endoscopies during the follow-up period, overall and by country

Overall United Kingdom United States Norway

Mean (SD) 29.9 (22.1) 26.5 (20.9) 38.7 (22.8) 24.5 (20.1)Length of follow-up time1 (mo)

Median (IQR) 25 (12-47) 21 (9-42) 39.5 (17-60) 16 (12-32.5)

Mean (SD) 3.0 (2.3) 2.7 (2.5) 4.0 (2.4) 2.3 (1.5)Number of follow-up visits (per patient)

Median (IQR) 2 (1-4) 2 (1-3) 4 (2-6) 2 (1-3)

Mean (SD) 0.7 (0.8) 0.5 (0.7) 0.3 (0.5) 1.2 (0.9)Number of follow-up visits with biopsy data (per patient)

Median (IQR) 0.5 (0-1) 0 (0-1) 0 (0-1) 1 (1-1.5)

Endoscopy with duodenal biopsy during follow-up

Yes n (%) 150 (50.0) 42 (42.0) 26 (26.0) 82 (82.0)

No n (%) 150 (50.0) 58 (58.0) 74 (74.0) 18 (18.0)

If ‘Yes’, number of endoscopies with duodenal biopsy during follow-up

1 n (%) 116 (77.3) 37 (88.1) 22 (84.6) 57 (69.5)

2 n (%) 22 (14.7) 3 (7.1) 4 (15.4) 15 (18.3)

3 n (%) 9 (6.0) 1 (2.4) 0 (0.0) 8 (9.8)

> 3 n (%) 3 (2.0) 1 (2.4) 0 (0.0) 2 (2.4)

Timing of endoscopy with duodenal biopsy during follow-up2

< 12 mo n (%) 45 (30.0) 18 (42.9) 3 (11.5) 24 (29.3)

12-24 mo n (%) 63 (42.0) 11 (26.2) 6 (23.1) 46 (56.1)

> 24 mo n (%) 64 (42.7) 19 (45.2) 18 (69.2) 27 (32.9)

For unknown diagnosis days and months, the day was imputed to the 15th, and the month was imputed to June.1Length of time from diagnosis to last follow-up available in charts, within the study period.2Total percentage may be greater than 100%.IQR: Interquartile range; SD: Standard deviation.

years after diagnosis and after starting a GFD to assess mucosal healing, as treatment of ongoing mucosal injury is less well defined and depends on likely etiology[3].

The grouping of patients into four disease state classes in this study allows for examination of the persistence of celiac disease symptoms as well as mucosal recovery/healing. Patients in this study with ongoing mucosal injury likely represent a combination of ongoing gluten exposure, slow recovery post diagnosis, and refractory celiac disease. Analysis of specific etiologies of ongoing villous atrophy, however, is outside the scope of this manuscript. Study results indicated that 36.6% of patients overall had presence of villous atrophy (classes 2 or 4) at the last follow-up visit with available biopsy data, with similar findings across sites. While it is unclear how many of these patients would progress to histologic remission given longer follow-up, these data suggest that a substantial proportion of patients may not be achieving therapeutic goals, even at specialized celiac disease centers. Furthermore, it is important to note that among those with at least one follow-up visit only half of patients had a follow-up biopsy to examine mucosal recovery. While the proportion of patients with persistent villous atrophy may be partially related to referral bias, the inclusion of patients diagnosed only at tertiary centers should have mitigated this. Conversely, patients who are not followed up or who receive care at less well-equipped centers may have even higher rates of inadequate disease control.

The reasons for the variability in follow-up, both within and between centers, are unclear. However, it seems that many of the patients in this study were either not continuing to see their gastroenterologist or not having a follow-up biopsy, which would limit the ability to assess continued presence of symptoms and villous atrophy. Yet, previous studies reported that having a follow-up biopsy did not impact long-term outcomes when compared with those who did not have a follow-up biopsy, possibly due to lack of effective interventions to address ongoing villous atrophy [10,11].



Figure 1 Number of patients grouped into the four disease state classes at diagnosis and at last follow-up with available biopsy data. Class 1: no symptoms and normal duodenal histology; Class 2: No symptoms and abnormal duodenal histology; Class 3: Symptoms and normal duodenal histology; Class 4: Symptoms and abnormal duodenal histology. 1Patients with biopsy result indicated as ‘other’ in the data collection form were excluded from this classification.

Potential country differences in healthcare policies may also be at play here. Indeed one previous study conducted in Norway reported that only 6% of patients had prevalence of villous atrophy after a median follow-up of 8.1 years[10]. The authors of this Norwegian study indicated that this may be partially driven by the fact that, in Norway, patients diagnosed with celiac disease automatically qualify for a reimbur-sement to cover the extra costs associated with following a GFD. In another study, from Australia, rates of mucosal remission and response were 50% and 85% at one and five years, respectively[12]. In addition, Pekki et al[11] reported that 42% (n = 200) of the 476 patients examined in Finland, who had a repeat biopsy, continued to have atrophy after one year of follow-up[11]. In yet another study from Finland, the authors reported that 96% (n = 177) of patients had villous recovery after a mean of 11 years of follow-up while adhering to a GFD[13]. The present study, however, did not find a large difference by country for the proportion of patients with continued presence of villous atrophy during follow-up.

Strengths of this study are the inclusion of patients with biopsy-proven celiac disease, the multinational sample, and the use of consecutively diagnosed patients, which should have reduced selection bias. However, future research may be warranted to examine whether patterns of care are different in community-based compared with tertiary centers, and whether there are potential differences in outcomes for patients diagnosed by serology alone and followed up in general practice. Given that the sites in this study were large gastroenterology referral centers, it is anticipated that they should be reflective of practice patterns in similar centers within the countries studied, and where there were commonalities between the centers, these are likely generalizable. However, as this cannot be tested, it is also likely that the selected sites may not be truly representative of the country, and these findings would need to be confirmed by further research within each country. In addition, patterns of care are reflective of those in gastroenterology referral centers, and may be more rigorous than patterns of care in general practice.

Limitations of this study include the lack of information regarding adherence to a GFD, as this information is often not readily available in patient charts, although most patients (approximately 80%) were referred to a dietician at least once during the follow-up period. Future studies may be able to assess GFD adherence objectively through the presence of gluten immunogenic peptides in the urine[14]. There is also the possibility that variation in pathology assessment and reporting may influence



inter center results; although, good interobserver agreement for the detection of villous atrophy has been reported[15]. In addition, the majority of patients included in this study were diagnosed on the basis of symptoms, with approximately 12% diagnosed by screening alone. While asymptomatic patients may have different outcomes, related in part to GFD adherence, the current study was not designed to address this. However, it would be valuable for future studies to consider and compare outcomes based on whether diagnosis was based on asymptomatic vs symptomatic disease.

Further, it is unclear what proportion of patients in this study were diagnosed elsewhere and referred to one of the participating gastroenterology centers owing to lack of healing. This may have resulted in a higher proportion of patients with villous atrophy compared with a community setting. In addition, this study captured patient visits to the gastroenterologist only, and any continued management with another healthcare provider (e.g., general practitioner, dietician) was not captured. Therefore, the results of this study are reflective of follow-up and outcomes for patients with celiac disease as per their management by the gastroenterologist. While it is expected that most patients will continue to be managed by a gastroenterologist, particularly if they continue to experience symptoms and have no evidence of mucosal healing, management by a general practitioner or other specialist (e.g., dietician) may occur in parallel. In addition, given that the inclusion criteria required selection of patients with at least one follow-up visit within the study period, to report on follow-up patterns and outcomes, the study is unable to provide information on patients who did not return to the gastroenterologist for a follow-up visit during the study period. Further, comparisons made between sites/countries relied on standard parameters assessed across sites including celiac serologies (but heterogeneous in the frequency of retesting), symptoms assessment, GFD adherence and nutritional values. However, differences across the sites and the standard of practice would largely be the driver of follow-up endoscopy/biopsy, and the authors recognize this limitation in adequately comparing outcomes across patients.

There is a lack of clarity in guidelines on types of clinicians who are most appropriate to administer follow-up care and management for patients with celiac disease, and this may be especially important given increasing activity of non-traditional practitioners. Results from a patient survey indicated that 27% of patients had not visited a healthcare provider about celiac disease over the past five years, with almost half of these patients reporting that they felt that they were managing their celiac disease effectively on their own[16]. Therefore, despite the present study focusing specifically on management by gastroenterologists, it may be that some patients choose to manage celiac disease on their own and do not return for regularly scheduled visits.

This study provides valuable insight into the monitoring patterns and outcomes of patients with celiac disease managed at large referral centers in real-world practice. Overall, the monitoring of patients, including the rate of follow-up biopsy, varied across the participating sites, with a higher proportion of Norwegian patients receiving a follow-up biopsy compared with patients in the United Kingdom and United States. Differences were also observed in the presentation of extraintestinal manifestations at diagnosis across the sites. In addition, the study results indicate that a large proportion of patients continue to have villous atrophy and continue to experience symptoms after diagnosis; a finding that was consistent across sites. Pharmacological management may be required for patients who are adherent to a GFD but who still experience symptoms and mucosal injury.

CONCLUSIONIn general, patients are not routinely monitored for the outcome of a GFD on symptoms, which may have an impact on intestinal health and can be a burden to patients. Overall, these data suggest that more routine follow-up assessment of celiac disease activity is needed. The inconsistent rates of mucosal assessment may be of concern, especially as many patients do not achieve histological remission. Novel, less invasive measures for assessment of ongoing villous atrophy, in combination with adjunctive pharmacologic therapy, may be needed to improve outcomes in patients with celiac disease.



ARTICLE HIGHLIGHTSResearch backgroundLong-term outcomes and monitoring patterns for patients with celiac disease in the real world are unknown.

Research motivationLong-term management and regular follow-up of patients with celiac disease is thought to improve adherence to a gluten-free diet, improve mucosal healing, and symptom resolution. However, it is unclear how patients with celiac disease are managed in routine clinical practice. There is anecdotal evidence suggesting that practice patterns vary widely both between countries and between practices.

Research objectivesTo understand real-world clinical practice, patterns of patient follow-up and management, and to characterize patient outcomes related to symptoms and villous atrophy after diagnosis.

Research methodsA retrospective observational study was performed using medical chart data of patients from three gastroenterology referral centers in the United Kingdom, United States, and Norway for patients diagnosed with celiac disease between 2008-2012.

Research results300 patients were followed for a median of 25 mo. During follow-up, 68.4% of patients were recorded as having ongoing gastrointestinal symptoms and 11.0% had ongoing symptoms and enteropathy. 50.0% of patients underwent at least one follow-up duodenal biopsy and 36.6% had continued villous atrophy. Patterns of monitoring varied between sites.

Research conclusionsThis real-world study demonstrates variable follow-up of patients with celiac disease even as most patients continue to have abnormal histology and symptoms after diagnosis.

Research perspectivesThese data suggest that more routine follow-up assessment of celiac disease activity is needed. Novel and less invasive measures to assess ongoing villous atrophy, used in combination with adjunctive pharmacologic therapy, may be needed to improve outcomes in patients with celiac disease.

REFERENCESHall NJ, Rubin G, Charnock A. Systematic review: adherence to a gluten-free diet in adult patients with coeliac disease. Aliment Pharmacol Ther 2009; 30: 315-330 [PMID: 19485977 DOI: 10.1111/j.1365-2036.2009.04053.x]

1

Kurien M, Trott N, Sanders DS. Long-term care for patients with coeliac disease in the UK: a review of the literature and future directions. J Hum Nutr Diet 2016; 29: 617-623 [PMID: 27196331 DOI: 10.1111/jhn.12379]

2

Al-Toma A, Volta U, Auricchio R, Castillejo G, Sanders DS, Cellier C, Mulder CJ, Lundin KEA. European Society for the Study of Coeliac Disease (ESsCD) guideline for coeliac disease and other gluten-related disorders. United European Gastroenterol J 2019; 7: 583-613 [PMID: 31210940 DOI: 10.1177/2050640619844125]

3

Silvester JA, Rashid M. Long-term follow-up of individuals with celiac disease: an evaluation of current practice guidelines. Can J Gastroenterol 2007; 21: 557-564 [PMID: 17853949 DOI: 10.1155/2007/342685]

4

Ludvigsson JF, Bai JC, Biagi F, Card TR, Ciacci C, Ciclitira PJ, Green PH, Hadjivassiliou M, Holdoway A, van Heel DA, Kaukinen K, Leffler DA, Leonard JN, Lundin KE, McGough N, Davidson M, Murray JA, Swift GL, Walker MM, Zingone F, Sanders DS; BSG Coeliac Disease Guidelines Development Group; British Society of Gastroenterology. Diagnosis and management of adult coeliac disease: guidelines from the British Society of Gastroenterology. Gut 2014; 63: 1210-1228 [PMID: 24917550 DOI: 10.1136/gutjnl-2013-306578]

5

Rubio-Tapia A, Hill ID, Kelly CP, Calderwood AH, Murray JA; American College of Gastroenterology. ACG clinical guidelines: diagnosis and management of celiac disease. Am J

6




https://dx.doi.org/10.1111/jhn.12379


https://dx.doi.org/10.1177/2050640619844125


https://dx.doi.org/10.1155/2007/342685





Gastroenterol 2013; 108: 656-76; quiz 677 [PMID: 23609613 DOI: 10.1038/ajg.2013.79]Megiorni F, Mora B, Bonamico M, Barbato M, Montuori M, Viola F, Trabace S, Mazzilli MC. HLA-DQ and susceptibility to celiac disease: evidence for gender differences and parent-of-origin effects. Am J Gastroenterol 2008; 103: 997-1003 [PMID: 18177450 DOI: 10.1111/j.1572-0241.2007.01716.x]

7

Reilly NR, Green PH. Epidemiology and clinical presentations of celiac disease. Semin Immunopathol 2012; 34: 473-478 [PMID: 22526468 DOI: 10.1007/s00281-012-0311-2]

8

Tovoli F, Negrini G, Farì R, Guidetti E, Faggiano C, Napoli L, Bolondi L, Granito A. Increased risk of nonalcoholic fatty liver disease in patients with coeliac disease on a gluten-free diet: beyond traditional metabolic factors. Aliment Pharmacol Ther 2018; 48: 538-546 [PMID: 29984415 DOI: 10.1111/apt.14910]

9

Hære P, Høie O, Schulz T, Schönhardt I, Raki M, Lundin KE. Long-term mucosal recovery and healing in celiac disease is the rule - not the exception. Scand J Gastroenterol 2016; 51: 1439-1446 [PMID: 27534885 DOI: 10.1080/00365521.2016.1218540]

10

Pekki H, Kurppa K, Mäki M, Huhtala H, Laurila K, Ilus T, Kaukinen K. Performing routine follow-up biopsy 1 year after diagnosis does not affect long-term outcomes in coeliac disease. Aliment Pharmacol Ther 2017; 45: 1459-1468 [PMID: 28326597 DOI: 10.1111/apt.14048]

11

Newnham ED, Shepherd SJ, Strauss BJ, Hosking P, Gibson PR. Adherence to the gluten-free diet can achieve the therapeutic goals in almost all patients with coeliac disease: A 5-year longitudinal study from diagnosis. J Gastroenterol Hepatol 2016; 31: 342-349 [PMID: 26212198 DOI: 10.1111/jgh.13060]

12

Tuire I, Marja-Leena L, Teea S, Katri H, Jukka P, Päivi S, Heini H, Markku M, Pekka C, Katri K. Persistent duodenal intraepithelial lymphocytosis despite a long-term strict gluten-free diet in celiac disease. Am J Gastroenterol 2012; 107: 1563-1569 [PMID: 22825364 DOI: 10.1038/ajg.2012.220]

13

Moreno ML, Cebolla Á, Muñoz-Suano A, Carrillo-Carrion C, Comino I, Pizarro Á, León F, Rodríguez-Herrera A, Sousa C. Detection of gluten immunogenic peptides in the urine of patients with coeliac disease reveals transgressions in the gluten-free diet and incomplete mucosal healing. Gut 2017; 66: 250-257 [PMID: 26608460 DOI: 10.1136/gutjnl-2015-310148]

14

Villanacci V, Lorenzi L, Donato F, Auricchio R, Dziechciarz P, Gyimesi J, Koletzko S, Mišak Z, Laguna VM, Polanco I, Ramos D, Shamir R, Troncone R, Vriezinga SL, Mearin ML. Histopathological evaluation of duodenal biopsy in the PreventCD project. An observational interobserver agreement study. APMIS 2018; 126: 208-214 [PMID: 29372596 DOI: 10.1111/apm.12812]

15

Hughey JJ, Ray BK, Lee AR, Voorhees KN, Kelly CP, Schuppan D. Self-reported dietary adherence, disease-specific symptoms, and quality of life are associated with healthcare provider follow-up in celiac disease. BMC Gastroenterol 2017; 17: 156 [PMID: 29228908 DOI: 10.1186/s12876-017-0713-7]

16






https://dx.doi.org/10.1007/s00281-012-0311-2




https://dx.doi.org/10.1080/00365521.2016.1218540




https://dx.doi.org/10.1111/jgh.13060






https://dx.doi.org/10.1111/apm.12812


https://dx.doi.org/10.1186/s12876-017-0713-7


World Journal of



ORIGINAL ARTICLE


Development and validation of a prognostic model for patients with hepatorenal syndrome: A retrospective cohort study

Xin-Yu Sheng, Fei-Yan Lin, Jian Wu, Hong-Cui Cao

ORCID number: Xin-Yu Sheng 0000-0002-0984-2592; Fei-Yan Lin 0000-0002-9064-1033; Jian Wu 0000-0003-0087-3744; Hong-Cui Cao 0000-0002-6604-6867.

Author contributions: Sheng XY designed the research study, collected data, and wrote the manuscript; Lin FY and Wu J contributed to the analysis, conception, design, and manuscript writing; and Cao HC contributed to the design, study supervision, and manuscript writing; All authors read and approved the final manuscript.

Supported by Chinese High Tech Research & Development (863) Program, No. 2013AA020102.

Institutional review board statement: The study was reviewed and approved by the Ethics Committee of The First Affiliated Hospital, College of Medicine, Zhejiang University (No. 2019-1449-1).

Informed consent statement: The researchers only analyzed anonymous data, so informed consent was waived.

Conflict-of-interest statement: All authors have no conflict of interest related to the manuscript.

Xin-Yu Sheng, Fei-Yan Lin, Jian Wu, Hong-Cui Cao, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, Zhejiang Province, China

Xin-Yu Sheng, Fei-Yan Lin, Jian Wu, Hong-Cui Cao, National Clinical Research Center for Infectious Diseases, Hangzhou 310003, Zhejiang Province, China

Corresponding author: Hong-Cui Cao, MD, PhD, Professor, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, The First Affiliated Hospital, Zhejiang University School of Medicine, No. 79 Qingchun Road, Hangzhou 310003, Zhejiang Province, China. [email protected]

AbstractBACKGROUND Hepatorenal syndrome (HRS) is a severe complication of cirrhosis with high mortality, which necessitates accurate clinical decision. However, studies on prognostic factors and scoring systems to predict overall survival of HRS are not enough. Meanwhile, a multicenter cohort study with a long span of time could be more convincing.

AIM To develop a novel and effective prognostic model for patients with HRS and clarify new prognostic factors.

METHODS We retrospectively enrolled 1667 patients from four hospitals, and 371 eligible patients were finally analyzed to develop and validate a novel prognostic model for patients with HRS. Characteristics were compared between survivors and non-survivors, and potential prognostic factors were selected according to the impact on 28-d mortality. Accuracy in predicting 28-d mortality was compared between the novel and other scoring systems, including Model for End-Stage Liver Disease (MELD), Chronic Liver Failure-Sequential Organ Failure Assessment (CLIF-SOFA), and Chinese Group on the Study of Severe Hepatitis B-Acute-on-Chronic Liver Failure (COSSH-ACLF).

RESULTS Five prognostic factors, comprised of gender, international normalized ratio, mean corpuscular hemoglobin concentration, neutrophil percentage, and stage,



http://orcid.org/0000-0002-0984-2592

http://orcid.org/0000-0002-0984-2592

http://orcid.org/0000-0002-0984-2592

http://orcid.org/0000-0002-9064-1033

http://orcid.org/0000-0002-9064-1033

http://orcid.org/0000-0002-9064-1033

http://orcid.org/0000-0003-0087-3744

http://orcid.org/0000-0003-0087-3744

http://orcid.org/0000-0003-0087-3744

http://orcid.org/0000-0002-6604-6867

http://orcid.org/0000-0002-6604-6867

http://orcid.org/0000-0002-6604-6867


Sheng XY et al. Prognostic model for HRS


Data sharing statement: No additional data are available.

STROBE statement: The authors have read the STROBE Statement-checklist of items, and the manuscript was prepared and revised according to the STROBE Statement-checklist of items.





Peer-review report’s scientific quality classificationGrade A (Excellent): 0 Grade B (Very good): B, B, B, B, B, B Grade C (Good): C Grade D (Fair): 0 Grade E (Poor): 0

Received: November 13, 2020 Peer-review started: November 13, 2020 First decision: February 11, 2021 Revised: February 23, 2021 Accepted: April 8, 2021 Article in press: April 8, 2021 Published online: May 28, 2021

P-Reviewer: Chaiteerakij R, Ho KY, Kao JT, Lipiński P, Morales-González JA, Uhlmann D S-Editor: Liu M L-Editor: Filipodia P-Editor: Liu JH

were integrated into a new score, GIMNS; stage is a binary variable defined by the number of failed organs. GIMNS was positively correlated with MELD, CLIF-SOFA, and COSSH-ACLF. Additionally, it had better accuracy [area under the receiver operating characteristic curve (AUROC): 0.830] than MELD (AUROC: 0.759), CLIF-SOFA (AUROC: 0.767), and COSSH-ACLF (AUROC: 0.759) in the derivation cohort (P < 0.05). It performed better than MELD and CLIF-SOFA in the validation cohort (P < 0.050) and had a higher AUROC than COSSH-ACLF (P = 0.122).

CONCLUSION We have developed a new scoring system, GIMNS, to predict 28-d mortality of HRS patients. Mean corpuscular hemoglobin concentration and stage were first proposed and found to be related to the mortality of HRS. Additionally, the GIMNS score showed better accuracy than MELD and CLIF-SOFA, and the AUROC was higher than that of COSSH-ACLF.

Key Words: Hepatorenal syndrome; Prognostic factor; Mean corpuscular hemoglobin concentration; Mortality; Scoring system; Cohort study


Core Tip: This multicenter retrospective cohort study investigated the prognostic factors for patients with hepatorenal syndrome and developed a novel prognostic model, named GIMNS. GIMNS contains five prognostic factors, comprised of gender, interna-tional normalized ratio, mean corpuscular hemoglobin concentration, neutrophil percentage, and stage, which had different expression levels between survivors and non-survivors. Stage, defined according to the number of organ failures and mean corpuscular hemoglobin concentration, was found to be an effective prognostic factor for the first time. The area under the operating characteristic curve reached 0.830 for 28-d mortality. The GIMNS score showed better accuracy than Model for End-Stage Liver Disease, Chronic Liver Failure-Sequential Organ Failure Assessment, and Chinese Group on the Study of Severe Hepatitis B-Acute-on-Chronic Liver Failure.

Citation: Sheng XY, Lin FY, Wu J, Cao HC. Development and validation of a prognostic model for patients with hepatorenal syndrome: A retrospective cohort study. World J Gastroenterol 2021; 27(20): 2615-2629URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2615.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2615

INTRODUCTIONHepatorenal syndrome (HRS) is a severe complication of cirrhosis characterized by increased splanchnic blood flow and rapid kidney dysfunction excluding other reasons[1]. It has high mortality within 2 wk, and the most effective treatment is liver transplantation, either liver only or simultaneous liver–kidney transplantation. Although pharmacological therapy, mainly vasoconstrictive (terlipressin or norepinephrine) plus albumin, can improve kidney function, it may make no difference to the percentage of patients who die or develop serious complications or the percentage of complications of any severity[2]. A retrospective study demonstrated that with present medical tools, there is still significantly high mortality in HRS despite guideline-based treatment[3]. According to another, single center study, occurrence of HRS is associated with a significantly worse outcome after living donor liver transplantation[4]. The focus of current HRS research is mainly on diagnosis and treatment, and there has been only a small number of studies on the prognosis of HRS. Thus, it is critical to develop a predictive tool to predict the mortality of HRS and to manage liver graft allocation on waiting lists.

A prospective study has demonstrated that age, serum bilirubin, and no reversal of creatinine after volume expansion independently predict mortality[5]. Another study has claimed that factors associated with poor prognosis are baseline bilirubin, no








reversibility of type-1 HRS, lack of resolution of the infection, and sepsis after diagnosis of type-1 HRS[6]. Acute-on-chronic liver failure (ACLF) grade is defined by the number of organ failures. A retrospective analysis from centers in Europe found that ACLF grade is the largest determinant of response and could affect the survival of HRS[7]. Among the predictors, the Model for End-stage Liver Disease (MELD) score has been widely applied for evaluating the severity of advanced liver diseases and liver graft allocation[8]. However, it has a limitation of failure to predict mortality after liver transplantation[9]. Therefore, a new effective model is urgently needed.

In this study, we aimed to construct a new prognostic model to predict mortality of HRS through a multicenter cohort study and retrospectively enrolled HRS patients from 2011 to 2019.

MATERIALS AND METHODSPatient studiesStudy population and data collection: In this retrospective cohort study we enrolled patients from four hospitals: First affiliated Hospital of Zhejiang University, Shulan Hospital, People’s Hospital of Zhejiang Province, and People’s Hospital of Shengzhou City. We studied a cohort with decompensated cirrhosis with acute renal injury from January 2011 to October 2019. Patients from the First Affiliated Hospital of Zhejiang University formed the derivation cohort and patients from the other three hospitals were the validation cohort. Demographic data (age and gender), history (vital signs and treatment), and laboratory parameters were available from medical records or the hospital database. The 28-d mortality was obtained from medical records or by directly contacting the patients or their families. All assays for serum biochemical parameters were routinely performed at the Central Clinical Laboratory of the four hospitals with the same type of testing equipment. The study was approved by the Ethics Committee of the First Affiliated Hospital, College of Medicine, Zhejiang University (No. 2019-1449-1), and complied with the ethical guidelines of the Declaration of Helsinki. The researchers only analyzed anonymous data, so informed consent was waived.

Inclusion and exclusion criteria: HRS was diagnosed based on criteria proposed by the International Club of Ascites in 2015. Acute kidney injury (AKI) was identified according to the standard Kidney Disease: Improving Global Outcomes criteria. The exclusion criteria were: (1) absence of ascites; (2) hepatocellular carcinoma; (3) other types of tumors; (4) chronic renal diseases; (5) liver transplantation; (6) age < 18 years; (7) thyroid diseases; (8) severe immunosuppression; (9) hospital stay < 1 wk; and (10) incomplete information. Finally, we excluded patients who were lost to follow-up (Figure 1).

Scoring models: The MELD score was calculated by the following formula: MELD = 3.78 ln [Total bilirubin (mg/dL)] + 11.2 ln (INR) + ln [serum creatinine (mg/dL)] + 6.43 (INR is international normalized ratio). The Chronic Liver Failure-Sequential Organ Failure Assessment (CLIF-SOFA) score summed the severity grades of organ failures. Liver failure was defined as bilirubin ≥ 12 mg/dL, coagulation failure as INR ≥ 2.5, brain failure as hepatic encephalopathy grade ≥ 3 (West Haven criteria), respiratory failure as a pulse oximetric saturation/fraction of inspired oxygen ratio ≤ 214 or arterial partial pressure of oxygen/fraction of inspired oxygen ratio ≤ 200 or the need for mechanical ventilation, and circulatory failure as the need for vasopressor therapy to achieve an adequate mean arterial pressure (MAP)[10]. The Chinese Group on the Study of Severe Hepatitis B-ACLF (COSSH-ACLF) was calculated as 0.741 INR + 0.523 hepatitis B virus (HBV)-SOFA + 0.026 age + 0.003 Total bilirubin. Stage was defined according to the number of organ failures, except renal failure. Stage 0 contained patients with zero or one organ failure, while stage 1 contained patients with more than two organ failures.

Statistical analysis: Clinicopathological features were summarized using medians and interquartile ranges or frequencies with percentages for normally distributed factors while using means with standard deviation for skewed distributed factors. Continuous variables were compared using t test or Mann–Whitney U test, and categorical variables were compared using the χ2 test or Fisher’s exact test. Patients were followed up from the day of diagnosis with HRS–AKI and ended at death or last follow–up at day 28. Univariate and multivariate Cox regression analyses were performed to establish the association between prognostic factors and overall survival of HRS. The



Figure 1 Workflow. The eligible cohort selection process after applying the inclusion and exclusion criteria from four hospitals. HRS: Hepatorenal syndrome.

Kaplan–Meier method was used to evaluate the survival probability of HRS, and a log–rank test was used to assess differences between groups. Hazard ratios were estimated using the Cox regression model. Statistical analyses were performed with R version 3.6.2 (R Foundation for Statistical Computing, Vienna, Austria). All tests were two-sided, and P < 0.05 was considered statistically significant. The statistical methods of this study were reviewed by a member of the Biostatistics Service from the First Affiliated Hospital, College of Medicine, Zhejiang University.

RESULTSClinical characteristicsWe retrospectively enrolled 1667 patients from four large general hospitals in Zhejiang province between January 2011 and October 2019. After inclusion and exclusion criteria, we had 383 eligible patients. Twelve patients were lost to follow-up due to a lack of contact information or survival information. Finally, we had 371 HRS patients for analysis: 248 in the derivation cohort and 123 in the validation cohort.

There were no significant differences in characteristics including demographic and clinical information between the derivation and validation cohorts (P > 0.05) (Table 1). Until the end of follow-up, 278 (74.9%) patients died, comprising 188 in the derivation cohort and 90 in the validation cohort. Male patients made up 76.8% (285) of the cohort, with 76.6% (190) and 77.2% (95) male in the derivation and validation cohorts, respectively. MAP, INR, alanine aminotransferase, serum bilirubin, number of failed organs, and the MELD, CLIF-SOFA, and COSSH-ACLF scores differed significantly between non-survivors and survivors in both cohorts (all P < 0.05). Age, gender, cause of cirrhosis, neutrophil percentage, hemoglobin, mean corpuscular hemoglobin (MCH) and MCH concentration (MCHC) differed between non-survivors and survivors only in the derivation cohort. Cirrhosis caused by HBV or HBV plus other reasons accounted for most of the patients (63.7% in the derivation cohort and 52.0% in the validation cohort). MAP was significantly higher in survivors than in non-survivors. Non-survivors had higher levels of INR and alanine aminotransferase compared with survivors. The MELD, CLIF-SOFA, and COSSH-ACLF scores were significantly higher in non-survivors in both cohorts (P < 0.05). Regarding treatment, all the eligible patients accepted rehydration treatment with albumin. In total, 21.3% (79) of patients received terlipressin plus albumin, and treatment was the same among the four hospitals (P > 0.05).



Table 1 Characteristics of patients with hepatorenal syndrome, n (%)

Variables Derivation cohort (n = 248) Validation cohort (n = 123)

Non-survivor (n = 188) Survivor (n = 60) P value Non-survivor (n

= 90) Survivor (n = 33) P value1P value

Age (y) 59.11 ± 12.61 55.43 ± 12.04 0.048 56.94 ± 12.17 57.21 ± 11.81 0.913 0.379

Gender (male/female) 138/50 52/8 0.031 69/21 26/7 0.995 0.795

Causes of cirrhosis 0.046 0.741 0.096

HBV/HBV plus others 124 (66.0) 34 (56.7) 45 (50.0) 19 (57.6)

Alcoholic 27 (14.4) 17 (28.3) 21 (23.3) 7 (21.2)

Others 37 (19.7) 9 (15.0) 24 (26.7) 7 (21.2)

MAP (mmHg) 95.16 ± 16.42 100.31 ± 18.44 0.041 93.1 ± 15.01 100.58 ± 18.33 0.023 0.483

INR 1.90 (1.70-2.60) 1.60 (1.30-1.90) < 0.001 1.96 (1.54-2.40) 1.50 (1.31-1.90) < 0.001 0.305

Neutrophil 75.68 ± 11.99 71.21 ± 13.03 0.015 76.03 ± 10.98 72.95 ± 10.97 0.170 0.647

RBC (× 109/L) 3.15 ± 0.85 3.04 ± 0.85 0.404 3.16 ± 1.05 2.99 ± 0.78 0.388 0.950

Hemoglobin (g/L) 103.13 ± 25.48 95.14 ± 26.89 0.044 102.55 ± 29.63 95.41 ± 25.60 0.229 0.840

MCV (fl) 94.20 ± 9.16 92.21 ± 8.07 0.146 95.28 ± 11.72 94.00 ± 9.71 0.580 0.267

MCH (pg) 33.19 ± 3.78 31.45 ± 3.21 0.002 33.17 ± 3.93 32.21 ± 3.66 0.233 0.756

MCHC (g/L) 351.81 ± 18.15 341.77 ± 19.35 < 0.001 349.14 ± 19.13 342.83 ± 18.25 0.104 0.355

Platelet (× 109/L) 66.5 (42.0-97.5) 69.0 (45.5-98.0) 0.419 64.0 (36.0-103.0) 75.0 (46.0-118.0) 0.250 0.737

Albumin (g/L) 29.12 ± 5.48 28.77 ± 5.37 0.668 29.22 ± 5.41 30.99 ± 5.61 0.115 0.274

A/G 1.00 (0.78-1.40) 0.92 (0.73-1.30) 0.267 1.13 ± 0.56 1.19 ± 0.51 0.620 0.400

ALT 60.00 (31.00-137.00)

32.50 (18.25-66.00) 0.001 59.00 (26.50-135.00)

36.00 (16.00-104.00) 0.035 0.673

Serum bilirubin 15.00 (5.32-27.50) 3.92 (1.10-12.80) < 0.001 18.30 (6.75-25.30) 2.30 (1.17-23.30) 0.003 0.779

Cholinesterase (U/L) 1963.00 (1399.00-2559.00)

2264.00 (1410.50-3081.50)

0.197 1937.50 (1585.75-2637.20)

2376.00 (1716.00-3903.00)

0.152 0.198

Creatinine (mg/dL) 1.60 (0.92-2.70) 1.84 (1.14-2.80) 0.206 1.43 (0.85-2.20) 2.10 (1.10-2.90) 0.046 0.291

Sodium (mmol/L) 133.74 ± 6.79 135.66 ± 6.34 0.053 132.88 ± 5.89 136.02 ± 4.90 0.009 0.499

Potassium (mmol/L) 4.35 ± 1.01 4.18 ± 0.74 0.217 4.16 ± 0.80 4.31 ± 0.65 0.352 0.288

Number of failed organs

1 (0.75-2.00) 0 (0-1.00) < 0.001 1 (0-2.00) 0 (0-1.00) < 0.001 0.433

One organ failed 65 (34.6) 19 (31.7) 0.797 28 (31.1) 11 (33.3) 0.987 0.764

Two organs failed 50 (26.6) 1 (1.7) < 0.001 26 (28.9) 3 (9.1) 0.040 0.596

≥ Three organs failed 26 (13.8) 1 (1.7) 0.017 12 (13.3) 1 (3.0) 0.188 0.990

MELD 29.50 ± 9.06 22.97 ± 8.00 < 0.001 28.04 ± 8.74 23.46 ± 9.47 0.013 0.276

CLIF-SOFA 10.79 ± 3.29 8.18 ± 3.08 < 0.001 10.86 ± 3.75 8.09 ± 3.37 < 0.001 0.912

COSSH-ACLF 7.57 ± 1.84 6.05 ± 1.27 < 0.001 7.57 ± 1.86 6.14 ± 1.49 < 0.001 0.958

Treatment 0.261 0.842 0.125

Terlipressin plus albumin

41 (21.8) 18 (30.0) 15 (16.7) 5 (15.2)

Albumin only 147 (78.2) 42 (70.0) 75 (83.3) 28 (84.8)

P non-survivor vs survivor.1P derivation cohort vs validation cohort. A/G: Albumin/globulin; ALT: Alanine aminotransferase; CLIF-SOFA: Chronic Liver Failure-Sequential Organ Failure Assessment; COSSH-ACLF: Chinese Group on the Study of Severe Hepatitis B-Acute-on-Chronic Liver Failure; HBV: Hepatitis B virus; INR: International normalized ratio; MAP: Mean arterial pressure; MCH: Mean cell hemoglobin; MCHC: Mean corpuscular hemoglobin concentration; MCV:



Mean cell volume; MELD: Model for End-Stage Liver Disease; RBC: Red blood cells.

Univariate and multivariate Cox regressionIn the derivation cohort, gender, MAP, INR, neutrophil percentage, hemoglobin, MCH, MCHC, albumin/globulin, serum bilirubin, sodium, stage (defined above), and the MELD, CLIF-SOFA, and COSSH-ACLF scores were significantly associated with overall survival by univariate Cox regression (Table 2). We incorporated the above variables into multivariate Cox regression and finally five prognostic factors, including gender, INR, neutrophil percentage, MCHC and stage, were independently associated with overall survival.

The hazard ratio of stage was 2.724 (95% confidence interval, 1.877–3.952), and the hazard ratios of gender, INR, neutrophil percentage, and MCHC were 0.532, 1.408, 1.031, and 1.014, respectively. The risk of death in female patients was approximately 1.5 times that of male patients. Patients with ≥ two organ failures had up to 2.7 times greater mortality risk compared with those with ≤ one organ failure. Creatinine was not associated with overall survival of HRS patients, although AKI stage was defined by creatinine level.

GIMNS score development According to the multivariate logistic regression based on 28-d mortality, the GIMNS score was developed. The formula is:

GIMNS = -1.412 gender + 0.053 neutrophil percentage + 0.014 MCHC + 2.073 ln (INR) + 1.231 stage - 9.217.

Mortality at 28 d in stratified HRS patients showed that patients with GIMNS > 2 were all dead, while those with GIMNS < 0 had the lowest mortality (30.34%) (Table 3). The mortality rates of patients with score 1–2 and 0–1 were 73.81% and 57.14%, respectively. Figure 2A showed the importance of each prognostic factor based on the logistic regression, and stage was the most important factor. Cumulative percentage also demonstrated this.

Characteristics in the GIMNS score The GIMNS score contains five variables: gender, INR, MCHC, neutrophil percentage, and stage. The mosaic plot acquired from the derivation cohort could intuitively sense the distribution of the patients’ survival statuses among different variables and establish the relationship between the variables (Figure 2B). From the mosaic plot, HBV-related cirrhosis accounted for the majority of the patients. For patients with HBV-related cirrhosis, male patients accounted for most, and 97.7% of the alcoholic HRS patients were male. As for patients with HRS of other causes, the proportion of female patients was larger than that of male patients. Although for every part of the whole mosaic plot, the number of patients with stage 1 disease was smaller than those with stage 0, the mortality was higher.

For the prognostic factors in the GIMNS score, we converted continuous variables into categorical variables by upper and lower quartiles (Figure 3A-C). INR was negatively correlated with survival probability. The survival probability of the upper quartile of INR was < 20%, which was lower compared with that of the others (P < 0.001). Similarly, patients with the lower quartile of neutrophil percentage had better overall survival compared with the others. Stratified MCHC also had different prognoses (P = 0.003).

Stage based on the number of organ failures (stage 0: 0–1, stage 1: ≥ 2) was the most important independent risk factor affecting overall survival of patients with HRS. The survival probability of stage 1 patients was < 10% while that of stage 0 patients was > 30%, which emphasized the essential role in predicting the prognosis of HRS (Figure 3D). Although female patients accounted for a small part of the whole cohort, they had a worse prognosis (P < 0.01) (Figure 3E).

GIMNS score and other scoresThe GIMNS score was significantly higher in non-survivors in the derivation and validation cohorts (Figure 4A). The GIMNS score was divided into three parts according to the upper and lower quartiles. According to the Kaplan–Meier curves, mortality increased with GIMNS score (P < 0.001) (Figure 4B). The GIMNS score had a superior discrimination of risk of mortality of HRS patients. The GIMNS score was positively correlated with the MELD, CLIF-SOFA, and COSSH-ACLF scores (Figure 5A-C).



Table 2 Univariate and multivariate Cox regression.

Univariate Cox regression Multivariate Cox regressionVariables

HR (95%CI) P value HR (95%CI) P value

Age 1.007 (0.995-1.019) 0.244

Gender 0.586 (0.425-0.807) 0.001 0.532 (0.384-0.737) < 0.001

Causes of cirrhosis 0.946 (0.785-1.140) 0.557

MAP 0.990 (0.981-0.999) 0.024

INR 1.842 (1.599-2.122) < 0.001 1.408 (1.173-1.691) < 0.001

Neutrophil 1.028 (1.015-1.041) < 0.001 1.031 (1.018-1.044) < 0.001

RBC 1.137 (0.959-1.348) 0.139

Hemoglobin 1.007 (1.002-1.012) 0.012

MCV 1.004 (0.988-1.021) 0.599

MCH 1.066 (1.022-1.111) 0.003

MCHC 1.019 (1.010-1.027) < 0.001 1.014 (1.005-1.022) 0.002

Platelet 0.997 (0.994-1.000) 0.090

Albumin 1.000 (0.973-1.028) 0.999

A/G 1.257 (1.093-1.445) 0.001

ALT 1.000 (1.000-1.001) 0.094

Serum bilirubin 1.002 (1.001-1.003) < 0.001

Cholinesterase 1.000 (0.999-1.000) 0.160

Creatinine 1.000 (0.999-1.001) 0.704

Sodium 0.960 (0.939-0.982) < 0.001

Potassium 1.215 (1.032-1.430) 0.019

Stage 3.805 (2.800-5.170) < 0.001 2.724 (1.877-3.952) < 0.001

MELD 1.077 (1.058-1.096) < 0.001

CLIF-SOFA 1.221 (1.166-1.278) < 0.001

COSSH-ACLF 1.467 (1.361-1.581) < 0.001

A/G: Albumin/globulin; ALT: Alanine aminotransferase; CI: Confidence interval; CLIF-SOFA: Chronic liver failure-sequential organ failure assessment; COSSH-ACLF: Chinese Group on the study of severe hepatitis B-acute-on-chronic liver failure; HR: Hazard ratio; INR: International normalized ratio; MAP: Mean arterial pressure; MCH: Mean cell hemoglobin; MCHC: Mean corpuscular hemoglobin concentration; MCV: Mean cell volume; MELD: Model for end-stage liver disease; RBC: Red blood cells.

Table 3 Mortality at 28 d in patients with hepatorenal syndrome stratified according to the GIMNS score

GIMNS score 28-d mortality in HRS patients (%)

≥ 2 100.0

1-2 73.8

0-1 57.1

< 0 30.3

HRS: Hepatorenal syndrome.

Accuracy of the GIMNS scoreThe area under the receiver operating characteristic curve (AUROC) of the GIMNS score for 28-d mortality was 0.830, with a sensitivity of 0.735 and specificity of 0.787 at a cut-off value of 0.36 (Table 4). The mortality rate of the upper quartile GIMNS score



Table 4 Accuracy of different scoring systems in derivation and validation cohorts

P valueModels Cut-off Sensitivity (%) Specificity (%) Youden AUROC (95%CI) P value compared with

GIMNS

Derivation cohort (n = 248)

GIMNS 0.4 73.5 78.7 0.522 0.830 (0.778-0.882) < 0.001 -

MELD 24.8 61.4 76.6 0.380 0.759 (0.670-0.821) < 0.001 0.029

CLIF-SOFA 9.0 72.7 68.1 0.408 0.767 (0.706-0.827) < 0.001 0.045

COSSH-ACLF 7.5 56.8 84.0 0.408 0.759 (0.696-0.821) < 0.001 0.026

Validation cohort (n = 123)

GIMNS 0.7 57.8 79.6 0.374 0.732 (0.642-0.821) < 0.001 -

MELD 26.6 65.6 63.0 0.286 0.623 (0.520-0.726) < 0.050 0.013

CLIF-SOFA 11.0 29.7 94.4 0.241 0.661 (0.563-0.758) < 0.010 0.049

COSSH-ACLF 7.9 51.6 81.5 0.331 0.674 (0.577-0.770) < 0.001 0.122

AUROC: Area under the receiver operating characteristic curve; CLIF-SOFA: Chronic liver failure-sequential organ failure assessment; COSSH-ACLF: Chinese Group on the study of severe hepatitis B-acute-on-chronic liver failure; MELD: Model for end-stage liver disease; CI: Confidence interval.

group was significantly higher than that of the lower quartile GIMNS score group. In the derivation cohort, the GIMNS score had significantly higher predictive power for the outcome of HRS patients than did the MELD, CLIF-SOFA, and COSSH-ACLF scores (P < 0.05) (Figure 5D). The AUROCs of GIMNS and other scoring systems are compared in Table 4.

Validation of performance of GIMNS scoreThe clinical and laboratory characteristics of the derivation and validation cohorts are shown in Table 1, which verified that there was no difference in the distribution of the characteristics between the cohorts. The mortality rate of the upper quartile GIMNS score group was significantly higher than that of the lower quartile GIMNS score group (P < 0.001). The AUROC of the GIMNS score model for 28-d mortality was 0.732, which was significantly better than that of the MELD (P = 0.038) and CLIF-SOFA (P = 0.049) scores and tended to be better than that of the COSSH-ACLF score (P = 0.122) (Figure 5E). The AUROCs of GIMNS and the other scoring systems were compared in Table 4.

DISCUSSIONIn this study, we developed a novel model for prediction of prognosis of HRS called GIMNS, which had better accuracy than the traditional scoring systems MELD and CLIF-SOFA and a larger AUROC than COSSH-ACLF. GIMNS contains five prognostic factors: gender, INR, MCHC, neutrophil percentage, and stage.

INR is normally used in scoring systems to evaluate the severity of advanced liver diseases. In our study, different quartiles of INR indeed resulted in significantly different survival rates. Patients with higher INR have problems with blood coagulation and have a worse outcome. There was an obvious survival benefit in male patients, although the proportion of female patients was smaller, accounting for only 23.2% of the whole cohort. This suggests that female HRS patients should be given more attention than usual[11].

MCHC has traditionally been a subordinate process to diagnose the type of anemia combined with MCH and mean corpuscular volume. However, the association between MCHC and outcome of HRS, especially the role of MCHC in predicting survival, has been ignored. We found that when adjusted by other confounding factors in multivariate Cox analysis, the mortality risk of patients in the lower quartile of MCHC was nearly twice that of patients in the upper quartile (data not shown).



Figure 2 Characteristics associated with survival. A: Importance of each characteristic of GIMNS; B: Mosaic plot. Survival status of patients with different causes of cirrhosis, gender, and stage. HBV: Hepatitis B virus; INR: International normalized ratio; MCHC: Mean corpuscular hemoglobin concentration.

It has been shown that advanced cirrhosis is linked to increased serum levels of proinflammatory cytokines, which are common triggers of HRS[12,13]. These cytokines somehow cause kidney dysfunction. Neutrophil percentage reflects inflam-matory state. Also, different neutrophil percentages divided by upper and lower quartiles resulted in different outcomes, meaning that patients with lower neutrophil percentage achieved a survival benefit.

For the first time, we incorporated the number of organ failures into the prognostic scoring system and divided the entire cohort into two, stage 0 with zero or one organ failure, except renal failure, and stage 1 with ≥ two organ failures. From the multivariate Cox analysis, we found that the 28-d mortality was significantly higher in stage 1 HRS, and the risk of stage 1 was almost three times that of stage 0 HRS. Organ failures have been used in the CLIF-SOFA score to evaluate the severity of advanced liver diseases[14]. Studies on the relationship between organ failure and prognosis of HRS are rare. It is essential to take this into consideration when developing a prognostic tool, either predicting the mortality or managing the liver transplantation waiting list.

The GIMNS score considered five prognostic factors and had better accuracy than the MELD and CLIF-SOFA scores in the derivation and validation cohorts. It also had a trend toward greater accuracy than the COSSH-ACLF had. The MELD score has been widely used in advanced liver diseases and plays an important role in liver graft allocation policy[15]. The MELD score is associated with 3-mo survival and can alter overall graft and outcome of HRS after transplantation. There is no doubt that the MELD score has been the first choice when evaluating prognosis of HRS and managing waiting lists for liver transplantation. However, with deeper understanding of HRS, more prognostic factors associated with HRS have been included, and a new score for predicting the mortality of HRS should be developed.



Figure 3 Kaplan–Meier curves of variables of GIMNS. A: Stratified international normalized ratio; B: Stratified neutrophil percentage; C: Stratified mean corpuscular hemoglobin concentration; D: Stage; E. Gender. INR: International normalized ratio; MCHC: Mean corpuscular hemoglobin concentration.

Preceding studies have found that gender is not associated with AKI, including HRS. However, we found that female patients were in more danger compared with male patients. Previous research has shown that predictors of AKI improvement are absence of alcoholic hepatitis, baseline creatinine, and male gender[11]. Male patients performed better in AKI, including HRS. In the derivation cohort, 72.2% of male patients and 86.8% of female patients died. The mortality risk in female patients was almost twice that in male patients. Thus, female patients have been ignored for a long time and more attention should be paid to this group.

INR has long been used as an indicator to evaluate advanced liver diseases, and it is widely used in different scoring systems, such as MELD, CLIF-SOFA, and COSSH-ACLF. INR can reflect coagulation function, and according to the Kaplan–Meier curves, patients with higher INR can have poor outcome. If the INR level is > 2.50, mortality is 95.0%, while if it is < 1.84, mortality can decrease to 64.0%. MCHC can also predict 28-d mortality of HRS partly because of kidney and liver dysfunction. Although the underlying pathological mechanism of MCHC remains unclear, it is associated with outcome of HRS and thus can act as a prognostic factor. Erythrocyte indices contain MCHC, MCH, and mean corpuscular volume. MCH and mean



Figure 4 Distributions of the GIMNS score. A: The GIMNS score distribution between survivors and non-survivors in the derivation and validation cohorts; B: Kaplan–Meier curves of GIMNS.

corpuscular volume have no relationship with the prognosis of HRS. A previous retrospective study proposed that MCHC is independently associated with occurrence of HRS[16]. However, study on the association between MCHC and prognosis of HRS has not continued.

The GIMNS score is the first system to incorporate organ failure. It has been utilized in the ACLF score to stratify patients. It is classed as ACLF-1, ACLF-2, and ACLF-3 according to the number of organ failures, and this grade is significantly associated with outcome of ACLF. As HRS is a severe complication of end-stage cirrhosis, we previously proposed that stratification by the number of organ failures, like ACLF grade, could be useful for predicting prognosis of HRS. Finally, stage was found to be an important prognostic factor.

There have been some studies on factors associated with outcome of HRS, mainly mortality or reversal of HRS. One retrospective study has shown a death risk prediction score model included four independent risk factors: liver cancer, neutrophil > 70%, alanine aminotransferase > 40 U/L, and creatinine > 127 mmol/L[17]. It confirmed the essential role of neutrophil percentage in predicting mortality of HRS, although liver cancer should be excluded as it could be a competitive factor for mortality. We compared the creatinine levels between non-survivors and survivors in the derivation and validation cohorts, and there was no significant difference. Consistently, creatinine level was not associated with 28-d mortality. Another study showed that serum creatinine and urinary sodium at the time of diagnosis were associated with survival in univariate Cox analysis, but when adjusted by other confounding factors, serum creatinine had no relationship with outcome of HRS[18]. All the above results confirm that, although creatinine level defines the severity of AKI, it cannot help to predict the prognosis of HRS.

A retrospective study from a tertiary center created a new score that was an extension of the MELD score, including changes in serum bilirubin, creatinine, and albumin level during admission, to predict prognosis of type-1 HRS[19]. Although the concept of HRS-1 has been abandoned, it still helps to construct better prediction models. The MELD score has been widely used to evaluate advanced liver diseases, especially for liver graft allocation policy. However, it fails to predict outcomes of HRS after transplantation and could be a little outdated considering more prognostic factors have been found[20]. The new score derived from MELD proves this point.

Prognostic models for the short-term prognosis of HRS are not enough. Some commonly used prognostic models for end-stage liver disease, such as MELD, MELD-Na, COSSH-ACLF, and CLIF-SOFA, are developed for all end-stage liver diseases including severe hepatitis and cirrhosis. The novel model, GIMNS, takes HRS as the target disease. Moreover, our HRS patients were enrolled according to the latest diagnostic criteria of the International Club of Ascites in 2015, and the sample size is larger than other studies, which can increase the credibility of the results. We operated



Figure 5 Relationship between GIMNS and other scoring systems and receiver operating characteristic curve curves. A: GIMNS and Model for End-stage Liver Disease; B: GIMNS and Chronic Liver Failure-Sequential Organ Failure Assessment; C: GIMNS and Chinese Group on the Study of Severe Hepatitis B-Acute-on-Chronic Liver Failure; D: Receiver operating characteristic curve curves in derivation cohort; E: Receiver operating characteristic curve in validation cohort. CLIF-SOFA: Chronic Liver Failure-Sequential Organ Failure Assessment; COSSH-ACLF: Chinese Group on the Study of Severe Hepatitis B-Acute-on-Chronic Liver Failure; MELD: Model for End-stage Liver Disease; AUC: Area under the curve.

external validation in three other hospitals to prevent overfitting. Finally, GIMNS performed better in both the derivation and validation cohorts. As its prognostic efficacy is better than MELD, COSSH-ACLF, and CLIF-SOFA, it has the value of clinical application.

Our new GIMNS score was based on all previous studies and showed better accuracy than MELD and the other commonly used scoring systems. Also, it is inexpensive and convenient to use, and MCHC and stage are given more attention. Our study had some limitations. First, this was a retrospective study, and therefore had a lower level of evidence compared with randomized controlled trials. We tried to balance the bias through external validation in three hospitals throughout Zhejiang Province, while developing the GIMNS score in one hospital. This could increase the credibility of our study. Second, it was hard to distinguish between HRS and other complications of AKI with cirrhosis, especially acute tubular necrosis. Thus, our cohort may have contained acute tubular necrosis patients, which could have interfered with our model development. Pathological sections are needed to separate HRS from acute



tubular necrosis. Finally, this study did not include urine output in the AKI diagnosis, which may have missed some eligible patients.

In summary, we developed a new scoring system, GIMNS, to predict 28-d mortality of HRS patients. The GIMNS score contains five elements: Gender, INR, MCHC, neutrophil percentage, and stage. Some of these elements have been widely used and others are recent additions. The GIMNS score has better accuracy than MELD and CLIF-SOFA, and the AUROC is larger than that of COSSH-ACLF. Prospective studies are needed for confirmation.

CONCLUSIONWe developed a new scoring system, GIMNS, to predict 28-d mortality of HRS patients. Two of the five prognostic indicators, MCHC and stage, were proposed and found to be related to the mortality of HRS for the first time. Additionally, the GIMNS score showed better accuracy than MELD and CLIF-SOFA, and the AUROC was higher than that of COSSH-ACLF.

ARTICLE HIGHLIGHTSResearch backgroundHepatorenal syndrome (HRS) is a severe complication of cirrhosis with high mortality. It has a unique in vivo pathological environment compared with other end-stage liver diseases. Meanwhile, prognostic factors associated with prognosis of HRS are not rich enough, and current prognostic scoring systems for end-stage liver diseases need to be updated to fit the application in HRS.

Research motivationAs the diagnostic criteria for HRS were confusing before 2015, a multicenter retrospective study, strictly following the criteria proposed by the International Club of Ascites in 2015, was designed to systematically evaluate the evolution of HRS over the past decade at our institution and to figure out what affects the prognosis of HRS.

Research objectivesOur goal was to develop a novel scoring system to predict 28-d mortality of HRS and evaluate the accuracy, sensitivity, and specificity compared with other scoring systems, including Model for End-stage Liver Disease, Chronic Liver Failure-Sequential Organ Failure Assessment, and Chinese Group on the Study of Severe Hepatitis B-Acute-on-Chronic Liver Failure.

Research methodsDemographic/clinical variables and medical records of HRS patients from January 2011 to October 2019 were collected from four tertiary medical centers, and strict eligibility criteria were applied. The final analysis included 371 patients with HRS. Univariate and multivariate Cox regression analyses were performed for prediction modeling of HRS.

Research resultsWe found that five indicators can be used as prognostic factors for HRS, including gender, international normalized ratio, mean corpuscular hemoglobin concentration, neutrophil percentage, and stage (defined as a binary variable by the number of failed organs). They formed a new score, GIMNS, based on the weight coefficient. As the GIMNS increases, the risk of mortality gets higher. The sensitivity and specificity of GIMNS were much higher than those of Model for End-stage Liver Disease, Chronic Liver Failure-Sequential Organ Failure Assessment, and Chinese Group on the Study of Severe Hepatitis B-Acute-on-Chronic Liver Failure in both the derivation and validation cohorts.

Research conclusionsWe innovatively defined a new variable, stage (based on the number of failed organs), to predict the prognosis of HRS. Importantly, we developed a new scoring system, GIMNS, which was specifically for patients with HRS and performed better than



Model for End-stage Liver Disease, Chronic Liver Failure-Sequential Organ Failure Assessment, and Chinese Group on the Study of Severe Hepatitis B-Acute-on-Chronic Liver Failure in accuracy, sensitivity, and specificity. As the five prognostic factors in GIMNS are easily available, clinical application of GIMNS could be more convenient.

Research perspectivesThe GIMNS scoring system should be validated in future prospective studies to get a more complete assessment.

ACKNOWLEDGEMENTSWe thank The First Affiliated Hospital of Zhejiang University, Shulan Hospital, People’s Hospital of Zhejiang Province, and People’s Hospital of Shengzhou City for providing data.

REFERENCESYulyana Y, Ho IA, Sia KC, Newman JP, Toh XY, Endaya BB, Chan JK, Gnecchi M, Huynh H, Chung AY, Lim KH, Leong HS, Iyer NG, Hui KM, Lam PY. Paracrine factors of human fetal MSCs inhibit liver cancer growth through reduced activation of IGF-1R/PI3K/Akt signaling. Mol Ther 2015; 23: 746-756 [PMID: 25619723 DOI: 10.1038/mt.2015.13]

1

Best LM, Freeman SC, Sutton AJ, Cooper NJ, Tng EL, Csenar M, Hawkins N, Pavlov CS, Davidson BR, Thorburn D, Cowlin M, Milne EJ, Tsochatzis E, Gurusamy KS. Treatment for hepatorenal syndrome in people with decompensated liver cirrhosis: a network meta-analysis. Cochrane Database Syst Rev 2019; 9: CD013103 [PMID: 31513287 DOI: 10.1002/14651858.CD013103.pub2]

2

Bashir MH, Iqbal S, Miller R, Singh J, Mubarak G, Likhtshteyn M, Bigajer E, Gallagher B, Hurairah A, Stefanov D, McFarlane SI, Ferstenberg R. Management and outcomes of hepatorenal syndrome at an urban academic medical center: a retrospective study. Eur J Gastroenterol Hepatol 2019; 31: 1545-1549 [PMID: 31169567 DOI: 10.1097/MEG.0000000000001462]

3

Okamura Y, Hata K, Inamoto O, Kubota T, Hirao H, Tanaka H, Fujimoto Y, Ogawa K, Mori A, Okajima H, Kaido T, Uemoto S. Influence of hepatorenal syndrome on outcome of living donor liver transplantation: A single-center experience in 357 patients. Hepatol Res 2017; 47: 425-434 [PMID: 27323334 DOI: 10.1111/hepr.12764]

4

Salerno F, Cazzaniga M, Merli M, Spinzi G, Saibeni S, Salmi A, Fagiuoli S, Spadaccini A, Trotta E, Laffi G, Koch M, Riggio O, Boccia S, Felder M, Balzani S, Bruno S, Angeli P; Italian Association of the Hospital Gastroenterologists (AIGO) investigators. Diagnosis, treatment and survival of patients with hepatorenal syndrome: a survey on daily medical practice. J Hepatol 2011; 55: 1241-1248 [PMID: 21703199 DOI: 10.1016/j.jhep.2011.03.012]

5

Barreto R, Fagundes C, Guevara M, Solà E, Pereira G, Rodríguez E, Graupera I, Martín-Llahí M, Ariza X, Cárdenas A, Fernández J, Rodés J, Arroyo V, Ginès P. Type-1 hepatorenal syndrome associated with infections in cirrhosis: natural history, outcome of kidney function, and survival. Hepatology 2014; 59: 1505-1513 [PMID: 24037970 DOI: 10.1002/hep.26687]

6

Piano S, Schmidt HH, Ariza X, Amoros A, Romano A, Hüsing-Kabar A, Solà E, Gerbes A, Bernardi M, Alessandria C, Scheiner B, Tonon M, Maschmeier M, Solè C, Trebicka J, Gustot T, Nevens F, Arroyo V, Gines P, Angeli P; EASL CLIF Consortium, European Foundation for the Study of Chronic Liver Failure (EF Clif). Association Between Grade of Acute on Chronic Liver Failure and Response to Terlipressin and Albumin in Patients With Hepatorenal Syndrome. Clin Gastroenterol Hepatol 2018; 16: 1792-1800. e3 [PMID: 29391267 DOI: 10.1016/j.cgh.2018.01.035]

7

Wu J, Guo N, Zhang X, Xiong C, Liu J, Xu Y, Fan J, Yu J, Zhao X, Liu B, Wang W, Zhang J, Cao H, Li L. HEV-LFS : A novel scoring model for patients with hepatitis E virus-related liver failure. J Viral Hepat 2019; 26: 1334-1343 [PMID: 31294523 DOI: 10.1111/jvh.13174]

8

Angeli P, Gines P. Hepatorenal syndrome, MELD score and liver transplantation: an evolving issue with relevant implications for clinical practice. J Hepatol 2012; 57: 1135-1140 [PMID: 22749942 DOI: 10.1016/j.jhep.2012.06.024]

9

Organization Committee of 13th Asia-Pacific Congress of Clinical Microbiology and Infection. 13th Asia-Pacific Congress of Clinical Microbiology and Infection Consensus Guidelines for diagnosis and treatment of liver failure. Hepatobiliary Pancreat Dis Int 2013; 12: 346-354 [PMID: 23924491 DOI: 10.1016/s1499-3872(13)60055-7]

10

Wong F, Boyer TD, Sanyal AJ, Pappas SC, Escalante S, Jamil K. Reduction in acute kidney injury stage predicts survival in patients with type-1 hepatorenal syndrome. Nephrol Dial Transplant 2020; 35: 1554-1561 [PMID: 30887050 DOI: 10.1093/ndt/gfz048]

11

Solé C, Solà E, Huelin P, Carol M, Moreira R, Cereijo U, Mas JM, Graupera I, Pose E, Napoleone L, dePrada G, Juanola A, Fabrellas N, Torres F, Morales-Ruiz M, Farrés J, Jiménez W, Ginès P. Characterization of inflammatory response in hepatorenal syndrome: Relationship with kidney

12


https://dx.doi.org/10.1038/mt.2015.13


















https://dx.doi.org/10.1016/s1499-3872(13)60055-7


https://dx.doi.org/10.1093/ndt/gfz048



outcome and survival. Liver Int 2019; 39: 1246-1255 [PMID: 30597709 DOI: 10.1111/Liv.14037]Ginès P, Solà E, Angeli P, Wong F, Nadim MK, Kamath PS. Hepatorenal syndrome. Nat Rev Dis Primers 2018; 4: 23 [PMID: 30213943 DOI: 10.1038/s41572-018-0022-7]

13

Wu T, Li J, Shao L, Xin J, Jiang L, Zhou Q, Shi D, Jiang J, Sun S, Jin L, Ye P, Yang L, Lu Y, Li T, Huang J, Xu X, Chen J, Hao S, Chen Y, Xin S, Gao Z, Duan Z, Han T, Wang Y, Gan J, Feng T, Pan C, Li H, Huang Y, Xie Q, Lin S, Li L; Chinese Group on the Study of Severe Hepatitis B (COSSH). Development of diagnostic criteria and a prognostic score for hepatitis B virus-related acute-on-chronic liver failure. Gut 2018; 67: 2181-2191 [PMID: 28928275 DOI: 10.1136/gutjnl-2017-314641]

14

Alessandria C, Ozdogan O, Guevara M, Restuccia T, Jiménez W, Arroyo V, Rodés J, Ginès P. MELD score and clinical type predict prognosis in hepatorenal syndrome: relevance to liver transplantation. Hepatology 2005; 41: 1282-1289 [PMID: 15834937 DOI: 10.1002/hep.20687]

15

Koola JD, Chen G, Malin BA, Fabbri D, Siew ED, Ho SB, Patterson OV, Matheny ME. A clinical risk prediction model to identify patients with hepatorenal syndrome at hospital admission. Int J Clin Pract 2019; 73: e13393 [PMID: 31347754 DOI: 10.1111/ijcp.13393]

16

Zhang S, He LL, Wang XH, Dang ZB, Liu XL, Li MG, Wang XB, Yang ZY. A novel scoring model for predicting mortality risk in patients with cirrhosis and hepatorenal syndrome. Eur J Gastroenterol Hepatol 2018; 30: 938-943 [PMID: 29634664 DOI: 10.1097/MEG.0000000000001127]

17

Olivera-Martinez M, Sayles H, Vivekanandan R, D' Souza S, Florescu MC. Hepatorenal syndrome: are we missing some prognostic factors? Dig Dis Sci 2012; 57: 210-214 [PMID: 21850494 DOI: 10.1007/s10620-011-1861-1]

18

Yang YW, Wu CH, Hu RH, Ho MC, Tsai MK, Wu YM, Lee PH. Longitudinal assessment of prognostic factors for patients with hepatorenal syndrome in a tertiary center. Hepatol Int 2010; 4: 507-510 [PMID: 20827408 DOI: 10.1007/s12072-010-9180-8]

19

Sibulesky L, Leca N, Blosser C, Rahnemai-Azar AA, Bhattacharya R, Reyes J. Is MELD score failing patients with liver disease and hepatorenal syndrome? World J Hepatol 2016; 8: 1155-1156 [PMID: 27721921 DOI: 10.4254/wjh.v8.i27.1155]

20


https://dx.doi.org/10.1111/Liv.14037


https://dx.doi.org/10.1038/s41572-018-0022-7






https://dx.doi.org/10.1111/ijcp.13393




https://dx.doi.org/10.1007/s10620-011-1861-1


https://dx.doi.org/10.1007/s12072-010-9180-8


https://dx.doi.org/10.4254/wjh.v8.i27.1155


World Journal of



ORIGINAL ARTICLE

Observational Study

Inflammatory bowel disease in Tuzla Canton, Bosnia-Herzegovina: A prospective 10-year follow-up

Emir Tulumović, Nermin Salkić, Denijal Tulumović

ORCID number: Emir Tulumović 0000-0001-9064-8625; Nermin Salkić 0000-0003-4727-9267; Denijal Tulumović 0000-0002-6635-9679.

Author contributions: Tulumović E contributed to data collection and analysis, and writing of the first draft of the paper; Salkić N contributed to study design and patient recruitment; Tulumović D contributed to data analysis and writing of the first draft of the paper; Salkić N and Tulumović D revised the article critically for important intellectual content.

Institutional review board statement: The study was reviewed and approved by the Ethical Board of University Clinical Center Tuzla by the declaration 02-09/2-78/20, on the 01.12.2020.

Informed consent statement: All study participants, or their legal guardian, provided informed written consent prior to study enrollment.

Conflict-of-interest statement: The authors declare that there is no conflict of interest related to this manuscript.

Data sharing statement: The data underlying this article cannot be shared publicly, as it is provided with permission of University

Emir Tulumović, Nermin Salkić, Department of Gastroenterology and Hepatology, University Clinical Center Tuzla, Tuzla 75000, Tuzlanski Kanton, Bosnia and Herzegovina

Denijal Tulumović, Department of Nephrology, Dialysis and Transplantation, University Clinical Center Tuzla, Tuzla 75000, Tuzlanski Kanton, Bosnia and Herzegovina

Corresponding author: Emir Tulumović, MD, Doctor, Department of Gastroenterology and Hepatology, University Clinical Center Tuzla, Ibre Pašića bb, Tuzla 75000, Tuzlanski Kanton, Bosnia and Herzegovina. [email protected]

AbstractBACKGROUND The incidence and prevalence of inflammatory bowel disease (IBD) vary between regions but have risen globally in recent decades. A lack of data from developing nations limits the understanding of IBD epidemiology.

AIM To perform a follow-up review of IBD epidemiology in the Tuzla Canton of Bosnia-Herzegovina during a 10-year period (2009-2019).

METHODS We prospectively evaluated the hospital records of both IBD inpatients and outpatients residing in Tuzla Canton for the specified period of time between January 1, 2009 and December 31, 2019. Since all our patients had undergone proximal and distal endoscopic evaluations at the hospital endoscopy unit, we used the hospital’s database as a primary data source, alongside an additional cross-relational search of the database. Both adult and pediatric patients were included in the study. Patients were grouped by IBD type, phenotype, age, and gender. Incidence rates were calculated with age standardization using the European standard population. Trends in incidence and prevalence were evaluated as a 3-year moving average and average annual percentage change rates.

RESULTS During the 10-year follow-up period, 651 patients diagnosed with IBD were monitored (of whom 334, or 51.3%, were males, and 317, or 48.7%, were females). Of all the patients, 346 (53.1%) had been diagnosed with ulcerative colitis (UC), 292 (44.9%) with Crohn’s disease (CD), and 13 (2%) with indeterminate colitis (IC).



http://orcid.org/0000-0001-9064-8625

http://orcid.org/0000-0001-9064-8625

http://orcid.org/0000-0003-4727-9267

http://orcid.org/0000-0003-4727-9267

http://orcid.org/0000-0002-6635-9679

http://orcid.org/0000-0002-6635-9679


Tulumović E et al. Ten-year follow-up of IBD in Tuzla Canton


Clinical Center Tuzla. The data will be shared on reasonable request to the corresponding author with the permission of University Clinical Center Tuzla.

STROBE statement: The authors have read the STROBE Statement-checklist of items, and the manuscript was prepared and revised according to the STROBE Statement-checklist of items.




Country/Territory of origin: Bosnia and Herzegovina

Peer-review report’s scientific quality classificationGrade A (Excellent): 0 Grade B (Very good): B, B Grade C (Good): 0 Grade D (Fair): D, D Grade E (Poor): 0

Received: January 24, 2021 Peer-review started: January 24, 2021 First decision: March 7, 2021 Revised: March 15, 2021 Accepted: April 22, 2021 Article in press: April 22, 2021 Published online: May 28, 2021

P-Reviewer: Luo HS, Rawat K, Zhu YF S-Editor: Gao CC L-Editor: Wang TQ P-Editor: Liu JH

We observed 440 newly diagnosed patients with IBD: 240 (54.5%) with UC, 190 (43.2%) with CD, and 10 (2.3%) with IC. The mean annual crude incidence rates were found to be 9.01/100000 population for IBD [95% confidence interval (CI): 8.17-9.85], with 4.91/100000 (95%CI: 4.29-5.54) for UC and 3.89/100000 (95%CI: 3.34-4.44) for CD. Calculated IBD prevalence in 2019 was 146.64/100000 (95%CI: 128.09-165.19), with 77.94/100000 (95%CI: 68.08-87.70) for UC and 65.77/100000 (95%CI: 54.45-74.1) for CD. The average annual IBD percentage change was 0.79% (95%CI: 0.60-0.88), with -2.82% (95%CI: -2.67 to -2.97) for UC and 6.92% (95%CI: 6.64-7.20) for CD. During the study period, 24,509 distal endoscopic procedures were performed. The incidence of IBD was 3.16/100 examinations (95%CI: 2.86-3.45) or 1.72/100 examinations (95%CI: 1.5-1.94) for UC and 1.36/100 examin-ations (95%CI: 1.17-1.56) for CD.

CONCLUSION Trends in the incidence and prevalence of IBD in Tuzla Canton are similar to Eastern European averages, although there are significant epidemiological differences within geographically close and demographically similar areas.

Key Words: Inflammatory bowel disease; Crohn's disease; Ulcerative colitis; Indeterminate colitis; Epidemiology; Endoscopy


Core Tip: The greatest gap in knowledge about the epidemiological picture of inflam-matory bowel disease relates to developing countries, such as those in Eastern Europe. The heterogeneity between the results among the few available epidemiological studies concerning Eastern Europe is intriguing. Therefore, we aimed to perform a follow-up review of inflammatory bowel disease epidemiology in the Tuzla Canton of Bosnia-Herzegovina during a 10-year period (2009-2019).

Citation: Tulumović E, Salkić N, Tulumović D. Inflammatory bowel disease in Tuzla Canton, Bosnia-Herzegovina: A prospective 10-year follow-up. World J Gastroenterol 2021; 27(20): 2630-2642URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2630.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2630

INTRODUCTIONInflammatory bowel disease (IBD) is a group of diseases sharing characteristics of chronic and relapsing-remitting immune activation and inflammation within the gastrointestinal tract; this group shares several clinical and epidemiological character-istics[1].

Traditionally, a higher incidence of IBD is found in developed and industrialized countries. Recent reports indicate a stabilization of the incidence in most Western European countries, with the highest reported incidence in Netherlands [ulcerative colitis (UC) 17.2/100.000 inhabitants; Crohn’s disease (CD) 10.5/100.000 inhabitants] and North American countries, with the highest reported incidence in Nova Scotia, Canada (UC 23.14/100.000 inhabitants; CD 23.82/100.000 inhabitants), and suggest an increase in northern European countries[2-4]. On the other hand, there is a trend of increasing incidence in areas with traditionally lower incidence rates, such as the countries of Eastern Europe[2].

The heterogeneity of results among the few available epidemiological studies concerning Eastern Europe is intriguing[5-11]. It is apparent that the epidemiological picture of the region needs to be investigated if we are to further our understanding of the nature of IBD. Building on the results of previous studies conducted in Tuzla Canton[10,11], this study aimed to contribute to understanding the epidemiological picture of IBD in the region of Southeastern Europe by evaluating the epidemiological characteristics of IBD in the Tuzla Canton of Bosnia-Herzegovina over a follow-up period of 10 years (2009-2019).








MATERIALS AND METHODSStudy population and data collectionWe prospectively evaluated both IBD inpatients and outpatients residing in Tuzla Canton in northeast Bosnia-Herzegovina between January 1, 2009 and December 31, 2019 at the Department of Gastroenterology and Hepatology, University Clinical Center Tuzla. Both adult and pediatric patients were included in the study.

Tuzla Canton is one of ten cantons of the Federation of Bosnia-Herzegovina, the one of two entities in Bosnia-Herzegovina. As the most populous region with 12.5% of the nation’s population, Tuzla Canton is also an industrial center. The population in the study area is homogenous in terms of ethnicity and genetic background.

Since all our patients had undergone proximal and distal endoscopic evaluations at the hospital endoscopy unit, we used the hospital’s database as a primary data source, alongside an additional cross-relational search of the database. The study included only patients with a definitive diagnosis, based on widely accepted diagnostic criteria consisting of history data, clinical examinations, laboratory tests, radiology, endoscopy, and histology[12,13]; those with an uncertain diagnosis or without permanent residency in Tuzla Canton were excluded from the study.

Patients were grouped by IBD type, phenotype, age, and gender. As UC is cate-gorized according to endoscopic findings relating to the disease localization, it was divided into proctitis, left-sided colitis, and extensive colitis[12]. CD was categorized according to the Montreal classification[13]. Indeterminate colitis (IC) was diagnosed in those patients for whom endoscopic, radiological, and pathohistological evaluations could not confirm either of the two main forms of IBD[1].

Statistical analysisStatistical analyses were performed with SPSS software, version 26 (SPSS, IBM, United States). Calculations of disease incidence were performed based on the 2013 census and data from the Statistical Office of the Federation of Bosnia-Herzegovina.

Descriptive statistical parameters were used to determine the basic characteristics of the study population. The year of diagnosis was used to calculate incidence. Incidence and prevalence calculations were performed using census data from the Statistical Office of the Federation of Bosnia-Herzegovina. Ninety-five percent confidence intervals (95%CIs) for the incidence rate were calculated assuming a Poisson distri-bution of cases.

Crude annual incidence rates for both genders were calculated based on the number of diagnosed patients and the number of inhabitants, while the average incidence rate during the observed period was calculated based on the number of years of the study. Incidence rates were standardized using standard European age groups for each of the standardized age groups[14].

Morbidity trends from 2009 to 2019 were determined by moving 3-year averages with 95%CI and by calculating annual average percentage change. The annual incidence of new cases detected per 100 distal endoscopic procedures and colono-scopies was estimated. Trends in incidence and prevalence were estimated using a linear regression model, where applicable. Prevalence estimates during the observed period were produced based on the total number of detected cases and the number of inhabitants.

The statistical level of 95% (P < 0.05) was considered significant for all statistical tests. The statistic review of the study was performed by a biomedical statistician.

RESULTSFrom January 1, 2009 to December 31, 2019, 651 patients diagnosed with IBD were monitored (of whom 334, or 51.3%, were males, and 317, or 48.7%, were females). Of these, 346 (53.1%) had been diagnosed with UC, 292 (44.9%) with CD, and the remaining 13 (2%) with IC.

In total, 440 new IBD patients were diagnosed: 240 (54.5%) with UC, 190 (43.2%) with CD, and 10 (2.3%) with IC. Among the newly diagnosed cases, 230 (52.3%) were males, and 210 (47.7%) were females.

Age distributionThe mean age (± SD) of all monitored patients was 46.2 ± 16.6 years, with a median of 47 years (25th and 75th percentiles: 33 and 58 years). The mean age (± SD) of newly diagnosed patients was 45.26 ± 17.38 years, with a median age of 47 years (25th and 75th



percentiles: 30 and 59 years). The mean age of newly diagnosed patients with UC was 48.16 ± 16.31 years, with a median age of 50 years (25th and 75th percentiles: 35 and 60 years), among whom the youngest patient was 15 and the oldest, 81 years old. The mean age of newly diagnosed patients with CD was 41.18 ± 18.01 years, with a median age of 40 years (25th and 75th percentiles: 26 and 55 years), among whom the youngest patient was 8 and the oldest, 80 years old. In total, 27 (6.1%) newly diagnosed patients younger than 19 years of age were registered. UC was diagnosed in five of these (18.5%), while CD was diagnosed in the remaining 22 (81.5%).

Distribution by genderThere were slightly more males among the newly diagnosed patients than women, with a ratio of 52.3% to 47.7%. Males were more likely to suffer from both major forms of IBD. Of the new cases of UC diagnosed, 124 were male (51.7%) and 116 (48.3%) were female (a ratio of 1.07:1). In total, 102 new cases of CD among the male population were registered (53.7%) and 88 (46.3%) new cases among the female population (a ratio of 1.16:1). Among the ten new cases of IC, the gender distribution was equal.

In general, no statistically significant difference in age was observed between genders at the time of diagnosis. The mean age (± SD) among male patients was 45.22 ± 18.01 (8-81) years, while the mean age of female patients was 45.3 ± 16.7 (11-80) years (t = 0.046, d.f. = 438, P = 0.964). A statistically significant difference in age between genders was observed among the patients diagnosed with UC. Males were older, with a mean age of 50.98 ± 18.12 (17-81) years, compared to females, with a mean age of 45.15 ± 16.05 (15-79) years (t = 2.805, d.f. = 238, P = 0.005). The same observation was true for IC.

Among the patients diagnosed with CD, a statistically significant difference between genders was observed in terms of age at the time of diagnosis, and females were older than males. Therefore, the average age of males was 37.53 ± 17.47 (8-76) years, while the average age of females was 45.41 ± 17.79 (11-80) years (t = 3, d.f. = 188, P = 0.002).

IncidenceThe mean annual crude incidence of IBD during the study period was 9.01/100000 inhabitants (95%CI: 8.17-9.85), with an incidence of 9.64/100000 inhabitants (95%CI: 8.5-10.78) in the male population and 8.41/100000 inhabitants (95%CI: 7.16-9.65) in the female population. The age-standardized mean annual incidence was 8.9/100000 inhabitants (95%CI: 8.07-9.72). In general, the ratio between the incidence of individual forms (UC:CD:IC) of the disease during the study was 54.55%:43.18%:2.27%.

The mean crude annual incidence of UC was 4.91/100000 inhabitants per year (95%CI: 4.29-5.54), with an incidence of 5.16/100000 (95%CI: 4.25-6.07) among males, and 4.68/100000 (95%CI: 3.83-5.53) among females. The age-standardized incidence for the observed period was 4.9/100000 inhabitants per year (95%CI: 4.3-5.54). The average annual incidence during the last 5 years of the study (2015-2019) was 4.91/100000 inhabitants (95%CI: 4-5.83).

The mean crude annual incidence of CD was 3.89/100000 inhabitants (95%CI: 3.34-4.44), with a prevalence of 4.28/100000 (95%CI: 3.54-5.01) for males and 3.52/100000 (95%CI: 2.69-4.35) for females. The age-standardized incidence for the observed period was 3.76/100000 inhabitants (95%CI: 3.22-4.29). The average annual incidence for the last 5 years of the study was 4.05/100000 inhabitants (95%CI: 3.22-4.89).

The mean crude annual incidence of IC was 0.2/100000 inhabitants (95%CI: 0.08-0.33).

The mean crude annual incidence of IBD in the pediatric and adolescent population was 2.49/100000 inhabitants (95%CI: 1.55-3.43), with 0.46/100000 inhabitants (95%CI: 0.06-0.87) for UC and 2.03/100000 inhabitants (95%CI: 1.26-2.79) for CD.

Figure 1 presents the mean crude annual incidence of IBD clustered by age and gender. Overall, the highest incidence of IBD occurred in three age groups in the male population: Patients between 55 and 64 years (14/100000 inhabitants), between 25 and 34 years (10.8/100000 inhabitants), and between 45 and 54 years (10.6/100000 inhabitants).

In general, a higher incidence was recorded both in the younger female age groups and in the older male age groups among patients diagnosed with UC. The highest incidence was recorded among males aged between 55 and 64 (11.2/100000 inhabitants) and in the age group between 55 and 64 years among the female population, which was 6.9/100000 inhabitants.



Figure 1 Inflammatory bowel diseases incidence clustered by age and gender.

The highest incidence of CD was recorded among age groups in the male population comprising patients between 15 and 34 years: 7/100000 inhabitants. Among the female population, the disease was most often diagnosed in patients between 45 and 54 years of age, which amounts to 5.4/100000 inhabitants.

Based on the 3-year average incidence rate, which is presented in Figure 2, the incidence of both major forms of IBD occurred fairly uniformly throughout the study. The average annual incidence percentage variation was 0.79% (95%CI: 0.60-0.88) or -2.82% (95%CI: -2.67 - -2.97) for UC and 6.92% (95%CI: 6.64-7.20) for CD. Summarizing the above, we concluded that the UC incidence trend is stable; however, the same cannot be concluded for CD, where the incidence trend is increasing (Figure 2).

During the study period, 24509 distal endoscopic procedures were performed (an average of 2228 per year). Figure 3 shows the trend of incidence per 100 colonoscopies. Any patient suspected of having IBD underwent a detailed endoscopic evaluation in the form of a gastroscopy and colonoscopy with—if the disease phenotype allowed—intubation of the terminal ileum. The incidence of IBD was 3.16/100 examin-ations (95%CI: 2.86-3.45) or 1.72/100 examinations (95%CI: 1.5-1.94) for UC and 1.36/100 examinations (95%CI: 1.17-1.56) for CD.

A summarized review of the 3-year average incidence rate reported in previous studies[10,11] and a recent study is presented in Figure 4. Data from the last year of the previous study (2006) and the first year of our study (2009) were used to fill in for missing years by averaging the values of these 2 years.

PrevalenceIn the last year of the study, the prevalence of IBD was 147.44/100000 inhabitants (95%CI: 136.1-158.8): 154.48/100000 inhabitants (95%CI: 137.8-171.1) in the male population and 140.73/100000 inhabitants (95%CI: 125.2-156.2) in the female population. The age-standardized prevalence was 147.3/100000 inhabitants (95%CI: 136-158.7).

Clinical and phenotypical characteristics of IBDThe types of individual forms of IBD according to the Montreal classification at diagnosis are presented in Tables 1 and 2. In eight (0.12%) patients, a change in the previously made diagnosis was recorded.

Among 651 patients, 49 (7.53%) required surgical treatment. In total, 8 (2.31%) patients treated for UC and 41 (14%) treated for CD underwent surgical treatment due



Table 1 Clinical phenotypes of ulcerative colitis clustered by gender

Males Females Total

Proctitis, n (%) 32 (13.3) 37 (15.4) 69 (28.7)

Left-sided colitis, n (%) 68 (28.3) 64 (26.7) 132 (55)

Extensive colitis, n (%) 24 (10) 15 (6.3) 39 (16.3)

Table 2 Clinical phenotypes of Crohn’s disease according to Vienna classification

n (%)

Age

A1 13 (6.8)

A2 117 (61.6)

A3 60 (31.6)

Location

L1 79 (41.5)

L2 52 (27.4)

L3 59 (31.1)

Behaviour

B1 124 (65.3)

B2 49 (25.8)

B3 17 (8.9)

Perianal disease

Present 12 (6.3)

Absent 178 (93.7)

to complications. The most common indication was lumen stenosis, which occurred in 23 (46.94%) patients. Multiple surgical treatments were performed in four (8.16%) patients. Comorbidities, primarily gastroenterological and autoimmune diseases, were also monitored during the study. Primary sclerosing cholangitis was diagnosed in four (1.37%) patients treated for CD and one (0.03%) treated for UC. Primary biliary cirrhosis, followed by liver transplantation, was noted in one (0.15%) patient. Autoimmune hepatitis was diagnosed in two (0.3%) patients, one of whom was treated for UC and the other for CD. Familial adenomatous polyposis or ankylosing spondylitis were diagnosed in two patients. Malignant colon disease occurred in five (0.76%) patients, four of whom were treated for UC (1.16%) and one for CD (0.03%). Fatal outcome occurred in six (0.9%) patients. Two patients died due to the consequences of malignant colon disease, two immediately after demanding surgical procedures, while one patient died due to the consequences of toxic megacolon accompanied by massive pulmonary embolism.

DISCUSSIONUntil recently, scholars have considered IBD to be a group of diseases of developed Western societies. Recent research into the global epidemiological situation confirms a stabilization of the IBD incidence in these areas, with far higher prevalence than in less developed societies, such as in Eastern Europe[4]. The GBD (Global Burden of Disease) study observed the highest age-standardized prevalence in societies with the highest socio-demographic index (SDI)[15]. The study found the highest prevalence in the region of North America (422/100000 inhabitants) and Western Europe, especially the United Kingdom (449.6/100000 inhabitants). Bosnia-Herzegovina is currently classified as a country with a high middle SDI; according to the study, the values of



Figure 2 Annual incidence rates of two major forms of inflammatory bowel diseases presented as moving 3-yr average with a 95% confidence interval.

Figure 3 The trend of the annual incidence of inflammatory bowel diseases per 100 colonoscopies performed.

age-standardized prevalence do not differ significantly compared to regions and countries of a similar standard.

When it comes to neighbouring countries, the GBD study ranked Bosnia-Herzegovina among countries with an estimated age-standardized prevalence between 100 and 120/100000 inhabitants, which is still slightly lower than the results of our research. Among the former Yugoslavian republics, neighbouring Croatia and Slovenia lead, with an estimated prevalence of between 180 and 200/100000 inhabitants[15]. These amounts do not apply to Hungary; since the 1970s, several extensive studies have shown that Hungary hosts the highest incidence of IBD in



Figure 4 Summarized review of 3-yr average incidence rates on both previous and recent study. Since the time period 2006-2009 was not evaluated, we used the calculated average of 2006 and 2009.

Eastern Europe[16,17].Our study recorded the highest prevalence among males over the age of 75, which is

similar to the GBD study results. The highest prevalence among the female population in our study occurred among slightly younger age groups (44-54 years) than those globally[15].

Paralysis of the healthcare system caused by the war in our country at the end of the last century required a longer recovery, which is certainly reflected in the volume of diagnostic tests and the results of scientific research. When compared with incidence rates from 1995-2006[10,11], a stable trend can be seen in IBD incidence over the study’s 10-year period. According to a study by Salkic et al[11] of the period between 1995 and 2006, an average of 397 colonoscopies per year were performed, which is 3.2 times fewer than the average number of diagnostic procedures performed annually during our study. The frequency of diagnosing UC per 100 colonoscopies performed during the previous study was 3.2 times higher than now; CD was diagnosed 2.2 times more often according to the given number of procedures performed.

The incidence rate among developing countries and regions such as Eastern Europe can be seen to have significantly increased when compared to previous reports[4]. Earlier studies from developing regions have shown an increase in the incidence of CD compared to UC[18], which was (based on the average percentile annual variation) also found to be the case in our study. Conducted in 2010 among 22 European countries, the ECCO-EpiCom cohort study[3] compared epidemiological differences between diseases in both Western and Eastern Europe, presenting the clearest and most concise epidemiological picture of IBD in Eastern Europe. It showed that the annual crude incidence of IBD in Eastern Europe was half that in Western Europe, with an average of 8.1/100000 inhabitants. The incidence of UC was 4.6/100000 inhabitants in 2010, while the incidence of CD was 3.3/100000 inhabitants, which are fairly uniform results, compared to the results presented in our study.

When it comes to the area of Southeastern Europe, the epidemiological situation is—except for Greece, Croatia, and Bosnia-Herzegovina[5-11] — poorly researched or unknown. Epidemiological studies among the pediatric population have been performed only in Slovenia[8]. The incidence calculated during our study period is somewhat lower compared to that in this region. A 10-year study (2000-2010) of both major forms of IBD in Zadar county has found a trend of increasing incidence, similar to previous studies from the Tuzla Canton[5,6,10,11]. The other recent Croatian study (conducted in Split-Dalmatia county), on the other hand, observed a decline in UC incidence and stabilization of CD incidence[7]. With the results of our study showing



stabilization of UC incidence alongside an increasing trend in CD incidence, we came to the conclusion that there are significant epidemiological differences within geographically close and demographically similar areas.

The population of Tuzla Canton is generally older than it was during the previous studies, which ultimately affected the results in terms of the epidemiological situation. Although the canton’s population is still younger than the average European population, the patients involved in the study are generally older than the patients analysed in the ECCO-EpiCom study[3]. Despalatović et al[7] reported the highest incidence of UC within the 18-30 and 51-60 age groups. Their results are similar to ours and the results from a previous study from Tuzla Canton[11]. One can consider age structure to be a bimodal distribution of incidence. Similarities in the findings of these two studies can also be seen in relation to CD incidence, with the results showing a higher incidence of CD among the younger population[7,10].

When it comes to the pediatric and adolescent population, our study recorded a significantly lower incidence of IBD than research conducted in Croatia, Slovenia, and most of European countries[5,8,19], probably due to a relatively small number of colonoscopies performed in the pediatric and adolescent population in our center.

The main forms of IBD are characterized by a relapsing and remitting course, with a tendency to change localization and phenotype over time, although the results among studies are inconsistent[20-23]. The trend of localization change over time is more pronounced in UC than in CD. We registered a change in the localization of CD in 3.53% of patients during our study, which is significantly lower than those of the studies of Louis et al[21] and Lo et al[20]. Phenotype change was recorded in 16.47% of patients, which is close to the result of Lo et al[20] but drastically lower than that of Louis et al[21]’s study, which observed a change in phenotype in 45.9% of patients. Considering the new modalities of CD treatment—primarily in terms of the introduction of anti-TNF therapy—the reduction in phenotype change frequency is unsurprising. Modification of the early diagnosis was observed far less frequently than in the IBSEN study[24], in which the diagnosis of 9% of patients changed over time.

In comparison with an earlier study by Salkic et al[11], we observed a significantly higher incidence of proctitis at the expense of left-sided colitis. Given the significantly greater number of endoscopic procedures, our observation might have resulted from earlier detection of the disease. Interestingly, both studies detected the same incidence of extensive colitis, which was diagnosed in 16.3% of patients. The results of our study are close to the Eastern European average, with a higher incidence of left-sided colitis but a lower incidence of proctitis and extensive colitis. The localization and phenotype of CD do not differ significantly between Eastern and Western Europe. Except for a somewhat more frequent incidence of colic and stenotic types of CD, the results of our study do not differ significantly from the (Eastern) European average. The results of our study showed a higher incidence of the ileocolic form among the younger population, while the colic form was found to affect the older population more frequently (results which are analogous to previous studies)[25,26].

Despite the fact that patients with IBD develop a significantly higher risk of malignancy over time compared to the rest of the population[27], the results of our study show that a large number of newly diagnosed cases (128, or 37%, cases newly diagnosed with UC) undergo endoscopic examination less than two times. Similar results were obtained by Vienne et al[28] during a 7-year multicenter study in France: As many as 54% of patients with extensive colitis underwent endoscopic examination only once.

Etiological factors related to the onset or worsening of the course of the disease are numerous. IBD is mostly associated with industrialization, urbanization, and the Western way of life[2,4]. Tuzla Canton is the industrial and electric power center of Bosnia-Herzegovina, with coal as the primary raw material used for electricity production. Considering that the University Clinical Center Tuzla is the largest medical center in northeastern Bosnia-Herzegovina, it is not unrealistic to expect that the region suffers one of the highest IBD incidence rates in the country.

Population-based study in Bosnia-Herzegovina estimates that nearly half of all adults consume tobacco products on a daily basis[29]. Given the fact that tobacco consumption in public places in Bosnia-Herzegovina is not legally restricted, even a larger proportion of the population is exposed to tobacco smoke. The number of breastfeeding mothers in Bosnia-Herzegovina is far below the European average[30], which is alarming if the benefits of breastfeeding as a protective factor against the development of IBD are taken into account[31].

The design of our study was conceived according to a follow-up of patients over a 10-year period. Given the lack of an adequate information system from 2006 to 2009, 2009 was accepted as satisfactory for our study’s beginning point. Our study is



retrospective-prospective in nature, based on respondents evaluated and treated at the University Clinical Center Tuzla, which is the only clinical center in the region and is a meeting place for all IBD-treated patients in the region and beyond. Given that the largest gap in the epidemiological picture of IBD relates to regions and developing countries[32] that include Bosnia-Herzegovina, we believe that our study (along with earlier reports[10,11]) can serve as a template for understanding the epidemiological picture in developing countries. In future, a multicenter study of the territory of Bosnia-Herzegovina and the countries of Southeastern Europe would greatly help us to understand this issue.

Conducted in a clinical center, this was a hospital study and, as such, has its limitations. The University Clinical Center Tuzla was the only institution capable of providing a comprehensive evaluation of patients with IBD in Tuzla Canton. Additionally, since all patients undergoing any form of IBD treatment must have a written recommendation for treatment in order to obtain reimbursement for the costs, any IBD patient from Tuzla Canton must eventually be evaluated in our center. We therefore feel safe in our assumption that despite it being single-hospital-based, our study is a confident representation of the epidemiological status of IBD in our region.

The need to conduct the study was made apparent by previous research conducted on this topic during the 11-year period from 1995 to 2006. Given the already mentioned technical limitations, the study did not follow the period from the end of 2006 to 2009. However, by monitoring prevalence, we can conclude that the incidence data throughout this period are close to data from the end of the previous study and from the beginning of our study.

CONCLUSIONThe epidemiological image of IBD in Tuzla Canton is similar to the Eastern European average. Stagnation of the incidence of UC was registered, as well as an increase in the incidence of CD during the observed period. The prevalence of both major forms of IBD tends to increase. The incidence and prevalence of IC are similar to those of the Eastern European and the world average. The highest incidence of UC is observed in slightly older age groups, comparable to the average in Eastern Europe and the rest of the world. The highest incidence of CD is observed among younger age groups, between the second and fourth decades of life, as it follows Eastern European and world trends. Gender distribution among newly diagnosed patients during the study was approximately the same, with a slightly higher incidence among the males for all three most common forms of the disease. The results of the study did not show significant deviations according to the clinical and endoscopic characteristics of the disease compared to the European average.

ARTICLE HIGHLIGHTSResearch backgroundRecent epidemiological studies conducted in Southeastern Europe show increasing incidence of inflammatory bowel diseases (IBD) in areas previously characterized as low-incidence areas. However, the results are still heterogenous as studies were conducted in areas heterogenous in terms of ethnicity, genetics background and lifestyle.

Research motivationThe region of Eastern Europe, especially Balkan region, is poorly described in terms of epidemiology of inflammatory bowel disease. This obvious gap in our understanding of IBD epidemiology in this region of Europe was previously partially described, and this study represents a continuation of one of the longest continuous surveys in Europe at all with nearly 25 years of epidemiological data.

Research objectivesThe authors sought to publish an epidemiological evaluation of IBD through a detailed 10-year follow-up of the clinical and epidemiological characteristics of the major forms of the disease.



Research methodsThe study evaluated both IBD inpatients and outpatients residing in Tuzla Canton, and is single-hospital-based. Descriptive statistical parameters were used to determine basic characteristics of the study population. Incidence and prevalence calculations were performed using census data from the Statistical Office of the Federation of Bosnia-Herzegovina. The statistical level of 95% (P < 0.05) was considered significant for all statistical tests.

Research resultsWe observed 440 newly diagnosed patients with IBD: 240 (54.5%) with ulcerative colitis (UC), 190 (43.2%) with Crohn’s disease (CD), and 10 (2.3%) with indeterminate colitis (IC). The mean annual crude incidence rates were found to be 9.01/100000 population for IBD [95% confidence interval (CI): 8.17-9.85], with 4.91/100000 (95%CI: 4.29-5.54) for UC and 3.89/100000 (95%CI: 3.34-4.44) for CD. Calculated IBD prevalence in 2019 was 146.64/100000 (95%CI: 128.09-165.19), with 77.94/100000 (95%CI: 68.08-87.70) for UC and 65.77/100000 (95%CI: 54.45-74.1) for CD. The average annual IBD percentage change was 0.79% (95%CI: 0.60-0.88), with -2.82% (95%CI: -2.67 - -2.97) for UC and 6.92% (95%CI: 6.64-7.20) for CD.

Research conclusionsWe conclude that our region of Europe has a relatively stable incidence of UC (5/100000) and CD (4/100000), which is in line with previous predictions that IBD incidence in Eastern Europe is approximately half of the incidence in Western Europe. Whether this is the result of lower ascertainment or there are other factors included remains to be seen in future research.

Research perspectivesThe most important factor for future research is why the incidence is lower in Eastern Europe. There are numerous factors to be considered, which include environmental, genetic, and sociological peculiarities of the region and its population. All of these factors should be taken in consideration during the design of future research.

ACKNOWLEDGEMENTSThe authors would like to thank the staff of the Endoscopy Unit and the Department of gastroenterology and hepatology, University Clinical Center Tuzla for their valuable contribution and technical help.

REFERENCESFeldman M, Friedman LS, Brandt LJ. Sleisenger and Fordtran's gastrointestinal and liver disease: Pathophysiology/diagnosis/management. 10th ed. Philadelphia, PA: Saunders/Elsevier, 2016: 115-116

1

Loftus EV Jr. Clinical epidemiology of inflammatory bowel disease: Incidence, prevalence, and environmental influences. Gastroenterology 2004; 126: 1504-1517 [PMID: 15168363 DOI: 10.1053/j.gastro.2004.01.063]

2

Burisch J, Pedersen N, Čuković-Čavka S, Brinar M, Kaimakliotis I, Duricova D, Shonová O, Vind I, Avnstrøm S, Thorsgaard N, Andersen V, Krabbe S, Dahlerup JF, Salupere R, Nielsen KR, Olsen J, Manninen P, Collin P, Tsianos EV, Katsanos KH, Ladefoged K, Lakatos L, Björnsson E, Ragnarsson G, Bailey Y, Odes S, Schwartz D, Martinato M, Lupinacci G, Milla M, De Padova A, D'Incà R, Beltrami M, Kupcinskas L, Kiudelis G, Turcan S, Tighineanu O, Mihu I, Magro F, Barros LF, Goldis A, Lazar D, Belousova E, Nikulina I, Hernandez V, Martinez-Ares D, Almer S, Zhulina Y, Halfvarson J, Arebi N, Sebastian S, Lakatos PL, Langholz E, Munkholm P; EpiCom-group. East-West gradient in the incidence of inflammatory bowel disease in Europe: the ECCO-EpiCom inception cohort. Gut 2014; 63: 588-597 [PMID: 23604131 DOI: 10.1136/gutjnl-2013-304636]

3

Ng SC, Shi HY, Hamidi N, Underwood FE, Tang W, Benchimol EI, Panaccione R, Ghosh S, Wu JCY, Chan FKL, Sung JJY, Kaplan GG. Worldwide incidence and prevalence of inflammatory bowel disease in the 21st century: a systematic review of population-based studies. Lancet 2017; 390: 2769-2778 [PMID: 29050646 DOI: 10.1016/S0140-6736(17)32448-0]

4

Sincić BM, Vucelić B, Persić M, Brncić N, Erzen DJ, Radaković B, Mićović V, Stimac D. Incidence of inflammatory bowel disease in Primorsko-goranska County, Croatia, 2000-2004: A prospective population-based study. Scand J Gastroenterol 2006; 41: 437-444 [PMID: 16635912 DOI: 10.1080/00365520500320094]

5

Klarin I, Majnarić L, Jovanović Z, Nakić D, Vcev I, Vcev A. Epidemiology and clinical presentation 6






https://dx.doi.org/10.1016/S0140-6736(17)32448-0


https://dx.doi.org/10.1080/00365520500320094



of inflammatory bowel disease in Zadar County, Croatia. Coll Antropol 2013; 37: 1161-1170 [PMID: 24611329]Despalatović BR, Bratanić A, Radić M, Jurišić Z, Tonkić A. Epidemiological trends of inflammatory bowel disease (IBD) in Split-Dalmatia County, Croatia from 2006 to 2014. Eur J Intern Med 2017; 46: e17-e19 [PMID: 28882418 DOI: 10.1016/j.ejim.2017.08.026]

7

Urlep D, Blagus R, Orel R. Incidence Trends and Geographical Variability of Pediatric Inflammatory Bowel Disease in Slovenia: A Nationwide Study. Biomed Res Int 2015; 2015: 921730 [PMID: 26688822 DOI: 10.1155/2015/921730]

8

Malakos Z, Saridi M, Latsou D, Katsanos K, Bakasis AD, Albani E, Christodoulou D. Increasing Incidence of Crohn’s Disease in Epirus, Greece: A 30-Year Prospective Study from a Referral Center. [cited 5 May 2020]. In: Ecronicon [Internet]. Available from: https://www.ecronicon.com/ecgds/pdf/ECGDS-05-00211.pdf

9

Pavlovic-Calic N, Salkic NN, Gegic A, Smajic M, Alibegovic E. Crohn's disease in Tuzla region of Bosnia and Herzegovina: a 12-year study (1995-2006). Int J Colorectal Dis 2008; 23: 957-964 [PMID: 18548259 DOI: 10.1007/s00384-008-0493-1]

10

Salkic NN, Pavlovic-Calic N, Gegic A, Jovanovic P, Basic M. Ulcerative colitis in the Tuzla region of Bosnia and Herzegovina between 1995 and 2006: epidemiological and clinical characteristics. Eur J Gastroenterol Hepatol 2010; 22: 346-353 [PMID: 19474740 DOI: 10.1097/MEG.0b013e32832bfe05]

11

Stange EF, Travis SP, Vermeire S, Reinisch W, Geboes K, Barakauskiene A, Feakins R, Fléjou JF, Herfarth H, Hommes DW, Kupcinskas L, Lakatos PL, Mantzaris GJ, Schreiber S, Villanacci V, Warren BF; European Crohn's and Colitis Organisation (ECCO). European evidence-based Consensus on the diagnosis and management of ulcerative colitis: Definitions and diagnosis. J Crohns Colitis 2008; 2: 1-23 [PMID: 21172194 DOI: 10.1016/j.crohns.2007.11.001]

12

Silverberg MS, Satsangi J, Ahmad T, Arnott ID, Bernstein CN, Brant SR, Caprilli R, Colombel JF, Gasche C, Geboes K, Jewell DP, Karban A, Loftus EV Jr, Peña AS, Riddell RH, Sachar DB, Schreiber S, Steinhart AH, Targan SR, Vermeire S, Warren BF. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 2005; 19 Suppl A: 5A-36A [PMID: 16151544 DOI: 10.1155/2005/269076]

13

Pace M, Lanzieri G, Glickman M, Županič T. Revision of the European Standard Population: report of Eurostat's task force. Publications O_ce of the European Union. 2013. [cited 21 June 2020]. In: ec.europa.eu [Internet]. Available from: https://ec.europa.eu/eurostat/documents/3859598/5926869/KS-RA-13-028-EN.PDF/e713fa79-1add-44e8-b23d-5e8fa09b3f8f

14

GBD 2017 Inflammatory Bowel Disease Collaborators. The global, regional, and national burden of inflammatory bowel disease in 195 countries and territories, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Gastroenterol Hepatol 2020; 5: 17-30 [PMID: 31648971 DOI: 10.1016/S2468-1253(19)30333-4]

15

Lakatos L, Kiss LS, David G, Pandur T, Erdelyi Z, Mester G, Balogh M, Szipocs I, Molnar C, Komaromi E, Lakatos PL. Incidence, disease phenotype at diagnosis, and early disease course in inflammatory bowel diseases in Western Hungary, 2002-2006. Inflamm Bowel Dis 2011; 17: 2558-2565 [PMID: 22072315 DOI: 10.1002/ibd.21607]

16

Kurti Z, Vegh Z, Golovics PA, Fadgyas-Freyler P, Gecse KB, Gonczi L, Gimesi-Orszagh J, Lovasz BD, Lakatos PL. Nationwide prevalence and drug treatment practices of inflammatory bowel diseases in Hungary: A population-based study based on the National Health Insurance Fund database. Dig Liver Dis 2016; 48: 1302-1307 [PMID: 27481587 DOI: 10.1016/j.dld.2016.07.012]

17

Cosnes J, Gower-Rousseau C, Seksik P, Cortot A. Epidemiology and natural history of inflammatory bowel diseases. Gastroenterology 2011; 140: 1785-1794 [PMID: 21530745 DOI: 10.1053/j.gastro.2011.01.055]

18

Sýkora J, Pomahačová R, Kreslová M, Cvalínová D, Štych P, Schwarz J. Current global trends in the incidence of pediatric-onset inflammatory bowel disease. World J Gastroenterol 2018; 24: 2741-2763 [PMID: 29991879 DOI: 10.3748/wjg.v24.i25.2741]

19

Lo B, Vester-Andersen MK, Vind I, Prosberg M, Dubinsky M, Siegel CA, Bendtsen F, Burisch J. Changes in Disease Behaviour and Location in Patients With Crohn's Disease After Seven Years of Follow-Up: A Danish Population-based Inception Cohort. J Crohns Colitis 2018; 12: 265-272 [PMID: 29506105 DOI: 10.1093/ecco-jcc/jjx138]

20

Louis E, Collard A, Oger AF, Degroote E, Aboul Nasr El Yafi FA, Belaiche J. Behaviour of Crohn's disease according to the Vienna classification: changing pattern over the course of the disease. Gut 2001; 49: 777-782 [PMID: 11709511 DOI: 10.1136/gut.49.6.777]

21

Lovasz BD, Lakatos L, Horvath A, Szita I, Pandur T, Mandel M, Vegh Z, Golovics PA, Mester G, Balogh M, Molnar C, Komaromi E, Kiss LS, Lakatos PL. Evolution of disease phenotype in adult and pediatric onset Crohn's disease in a population-based cohort. World J Gastroenterol 2013; 19: 2217-2226 [PMID: 23599648 DOI: 10.3748/wjg.v19.i14.2217]

22

Vester-Andersen MK, Prosberg MV, Jess T, Andersson M, Bengtsson BG, Blixt T, Munkholm P, Bendtsen F, Vind I. Disease course and surgery rates in inflammatory bowel disease: a population-based, 7-year follow-up study in the era of immunomodulating therapy. Am J Gastroenterol 2014; 109: 705-714 [PMID: 24642581 DOI: 10.1038/ajg.2014.45]

23

Henriksen M, Jahnsen J, Lygren I, Sauar J, Schulz T, Stray N, Vatn MH, Moum B; Ibsen Study 24



https://dx.doi.org/10.1016/j.ejim.2017.08.026


https://dx.doi.org/10.1155/2015/921730

https://www.ecronicon.com/ecgds/pdf/ECGDS-05-00211.pdf


https://dx.doi.org/10.1007/s00384-008-0493-1


https://dx.doi.org/10.1097/MEG.0b013e32832bfe05


https://dx.doi.org/10.1016/j.crohns.2007.11.001


https://dx.doi.org/10.1155/2005/269076

https://ec.europa.eu/eurostat/documents/3859598/5926869/KS-RA-13-028-EN.PDF/e713fa79-1add-44e8-b23d-5e8fa09b3f8f

https://ec.europa.eu/eurostat/documents/3859598/5926869/KS-RA-13-028-EN.PDF/e713fa79-1add-44e8-b23d-5e8fa09b3f8f


https://dx.doi.org/10.1016/S2468-1253(19)30333-4










https://dx.doi.org/10.1093/ecco-jcc/jjx138









Group. Change of diagnosis during the first five years after onset of inflammatory bowel disease: results of a prospective follow-up study (the IBSEN Study). Scand J Gastroenterol 2006; 41: 1037-1043 [PMID: 16938716 DOI: 10.1080/00365520600554527]Castro M, Papadatou B, Baldassare M, Balli F, Barabino A, Barbera C, Barca S, Barera G, Bascietto F, Berni Canani R, Calacoci M, Campanozzi A, Castellucci G, Catassi C, Colombo M, Covoni MR, Cucchiara S, D'Altilia MR, De Angelis GL, De Virgilis S, Di Ciommo V, Fontana M, Guariso G, Knafelz D, Lambertini A, Licciardi S, Lionetti P, Liotta L, Lombardi G, Maestri L, Martelossi S, Mastella G, Oderda G, Perini R, Pesce F, Ravelli A, Roggero P, Romano C, Rotolo N, Rutigliano V, Scotta S, Sferlazzas C, Staiano A, Ventura A, Zaniboni MG. Inflammatory bowel disease in children and adolescents in Italy: data from the pediatric national IBD register (1996-2003). Inflamm Bowel Dis 2008; 14: 1246-1252 [PMID: 18521916 DOI: 10.1002/ibd.20470]

25

Charpentier C, Salleron J, Savoye G, Fumery M, Merle V, Laberenne JE, Vasseur F, Dupas JL, Cortot A, Dauchet L, Peyrin-Biroulet L, Lerebours E, Colombel JF, Gower-Rousseau C. Natural history of elderly-onset inflammatory bowel disease: a population-based cohort study. Gut 2014; 63: 423-432 [PMID: 23408350 DOI: 10.1136/gutjnl-2012-303864]

26

Axelrad JE, Lichtiger S, Yajnik V. Inflammatory bowel disease and cancer: The role of inflammation, immunosuppression, and cancer treatment. World J Gastroenterol 2016; 22: 4794-4801 [PMID: 27239106 DOI: 10.3748/wjg.v22.i20.4794]

27

Vienne A, Simon T, Cosnes J, Baudry C, Bouhnik Y, Soulé JC, Chaussade S, Marteau P, Jian R, Delchier JC, Coffin B, Admane H, Carrat F, Drouet E, Beaugerie L. Low prevalence of colonoscopic surveillance of inflammatory bowel disease patients with longstanding extensive colitis: a clinical practice survey nested in the CESAME cohort. Aliment Pharmacol Ther 2011; 34: 188-195 [PMID: 21615760 DOI: 10.1111/j.1365-2036.2011.04711.x]

28

Gligorić D, Petković S, Pepić A, Ateljević J, Vukojević B. The elasticity of tobacco products in BiH – macrodata analysis. Tob Prev Cessation 2019; 5: A97 [DOI: 10.18332/tpc/105226]

29

Šumanović-Glamuzina D. Breastfeeding practice in Bosnia and Herzegovina. J Pediatr Neonat Individual Med 2013; 2: 55-62 [DOI: 10.7363/020117]

30

Xu L, Lochhead P, Ko Y, Claggett B, Leong RW, Ananthakrishnan AN. Systematic review with meta-analysis: breastfeeding and the risk of Crohn's disease and ulcerative colitis. Aliment Pharmacol Ther 2017; 46: 780-789 [PMID: 28892171 DOI: 10.1111/apt.14291]

31

Kaplan GG, Ng SC. Understanding and Preventing the Global Increase of Inflammatory Bowel Disease. Gastroenterology 2017; 152: 313-321. e2 [PMID: 27793607 DOI: 10.1053/j.gastro.2016.10.020]

32


https://dx.doi.org/10.1080/00365520600554527









https://dx.doi.org/10.18332/tpc/105226

https://dx.doi.org/10.7363/020117






World Journal of



META-ANALYSIS

Association between oral contraceptive use and pancreatic cancer risk: A systematic review and meta-analysis

Milena Ilic, Biljana Milicic, Irena Ilic

ORCID number: Milena Ilic 0000-0003-3229-4990; Biljana Milicic 0000-0001-8091-2461; Irena Ilic 0000-0001-5347-3264.

Author contributions: All authors equally contributed to this paper with conception and design of the study, literature review and analysis, drafting and critical revision and editing, and final approval of the final version.

Supported by Ministry of Education, Science and Technological development, Republic of Serbia, 2011–2020, No. 175042.

Conflict-of-interest statement: The authors have no conflicts of interest to declare.

PRISMA 2009 Checklist statement: The authors have read the PRISMA 2009 Checklist, and the manuscript was prepared and revised according to the PRISMA 2009 Checklist.

Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build

Milena Ilic, Department of Epidemiology, Faculty of Medical Sciences, University of Kragujevac, Kragujevac 34000, Serbia

Biljana Milicic, Department of Medical Statistics and Informatics, Faculty of Dental Medicine, University of Belgrade, Belgrade 11000, Serbia

Irena Ilic, Faculty of Medicine, University of Belgrade, Belgrade 11000, Serbia

Corresponding author: Milena Ilic, MD, PhD, Professor, Department of Epidemiology, Faculty of Medical Sciences, University of Kragujevac, S. Markovica 69, Kragujevac 34000, Serbia. [email protected]

AbstractBACKGROUND Studies on the association of oral contraceptive (OC) use and pancreatic cancer showed inconsistent findings.

AIM To evaluate the relationship between OC use and pancreatic cancer risk.

METHODS A literature search for observational studies (case-control and cohort studies) was conducted up to December 2020. A meta-analysis was performed by calculating pooled relative risks (RRs) and 95% confidence intervals (CIs). Heterogeneity was assessed using Cochran’s chi-square test and I2 statistic. Subgroup analyses were performed by study design, source of controls in case-control studies, number of cases of pancreatic cancers, study quality according to Newcastle-Ottawa Scale score, geographical region and menopausal status. All analyses were performed using Review Manager 5.3 (RevMan 5.3).

RESULTS A total of 21 studies (10 case-control studies and 11 cohort studies) were finally included in the present meta-analysis, comprising 7700 cases of pancreatic cancer in total. A significant association was observed between the ever use of OC and pancreatic cancer risk in the overall analysis (RR = 0.85; 95%CI = 0.73-0.98; P = 0.03). Duration of OC use (< 1 year, < 5 years, 5-10 years, > 10 years) was not significantly associated with the risk of pancreatic cancer. Subgroup analyses revealed a statistically significant subgroup difference for the geographic region in which the study was conducted (Europe vs Americas vs Asia; P = 0.07).



http://orcid.org/0000-0003-3229-4990

http://orcid.org/0000-0003-3229-4990

http://orcid.org/0000-0003-3229-4990

http://orcid.org/0000-0001-8091-2461

http://orcid.org/0000-0001-8091-2461

http://orcid.org/0000-0001-5347-3264

http://orcid.org/0000-0001-5347-3264


Ilic M et al. OC and pancreatic cancer risk


upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/Licenses/by-nc/4.0/



Country/Territory of origin: Serbia

Peer-review report’s scientific quality classificationGrade A (Excellent): 0 Grade B (Very good): B, B Grade C (Good): 0 Grade D (Fair): 0 Grade E (Poor): 0

Received: February 6, 2021 Peer-review started: February 6, 2021 First decision: February 27, 2021 Revised: March 13, 2021 Accepted: April 23, 2021 Article in press: April 23, 2021 Published online: May 28, 2021

P-Reviewer: Sergi C S-Editor: Fan JR L-Editor: Filipodia P-Editor: Li JH

Subgroup analyses showed a statistically significant decrease in pancreatic cancer risk and OC use in high-quality studies, studies conducted in Europe, and in postmenopausal women.

CONCLUSION Despite the suggested protective effects of OC use in this meta-analysis, further epidemiological studies are warranted to fully elucidate the association between the use of OC and pancreatic cancer risk.

Key Words: Pancreatic cancer; Oral contraceptives; Risk factors; Risk assessment; Meta-analysis; Review


Core Tip: Although the understanding of the etiology of pancreatic cancer has improved dramatically over the past decades, the link between pancreatic cancer risk and oral contraceptive (OC) use is still insufficiently known. This meta-analysis showed a significant association between OC use and pancreatic cancer risk (overall relative risk = 0.85; 95% confidence interval = 0.73-0.98). A better understanding of the risks of pancreatic cancer occurrence in women who use OC may be relevant for pancreatic cancer prevention strategy.

Citation: Ilic M, Milicic B, Ilic I. Association between oral contraceptive use and pancreatic cancer risk: A systematic review and meta-analysis. World J Gastroenterol 2021; 27(20): 2643-2656URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2643.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2643

INTRODUCTIONPancreatic cancer is the seventh most common cause of death among malignant tumors in females, with about 220000 deaths worldwide in 2018[1,2]. Pancreatic cancer is one of the deadliest types of cancer, with an estimated overall 5-year survival rate less than 10%[1]. Pancreatic cancer mortality is characterized by a dramatic increase after the age of 30 years, reaching the highest burden in women at about 80-years-old[1,3].

An understanding of the etiology of pancreatic cancer has improved dramatically over the past decades and certain risk factors have been established including tobacco use, obesity, diabetes mellitus, chronic pancreatitis, positive family history and inherited genetic syndromes, high alcohol consumption, dietary factors, physical inactivity, workplace exposure to some chemicals, infections[4-7]. Although some risk factors have been identified, the causes of pancreatic cancer are still insufficiently known.

Regarding the link between pancreatic cancer risk and oral contraceptive (OC) use, epidemiological studies have shown conflicting results: Some findings have shown positive associations with risk of pancreatic cancer[8,9], whereas some studies have indicated inverse associations[10,11]. One previous meta-analysis of observational studies did not support the hypothesis that OC use is associated with pancreatic cancer risk (the pooled relative risk [RR] = 1.09, 95% confidence interval [CI] 0.96–1.23)[12].

A better understanding of risks of pancreatic cancer occurrence in women who use OC may be relevant for pancreatic cancer prevention strategy. The purpose of this study was to evaluate the relationship between the use of OC and pancreatic cancer risk by performing a meta-analysis of case-control and cohort studies.

MATERIALS AND METHODSThis meta-analysis was performed following the Preferred Reporting Items for








Systematic Reviews and Meta-Analyses guidelines[13].

Ethics statement This study is a part of a research project approved by the Ethics Committee of the Faculty of Medical Sciences, University of Kragujevac (No. 01-14321).

Literature searchWe searched PubMed from inception through December 2020 using combinations of keywords: (“oral contraceptives” or “birth control pills”) and (“pancreatic cancer” or “pancreatic tumor” or “pancreatic neoplasm”) and (“risk” or “risk factors” or “risk assessment”). Additionally, references of retrieved studies and reviews were hand-searched to identify additional relevant studies (up to 31 December 2020). No language restrictions were applied in the search.

Inclusion and exclusion criteriaTwo authors (II and MI) independently screened the titles and abstracts of studies retrieved by literature search. Subsequently, the full texts of articles that were identified as relevant were assessed. Any disagreements between the reviewers were resolved through consensus. Studies which reported on the association between the use of OC (the exposure of interest) and risk of pancreatic cancer (the outcome of interest) and that were designed as case-control studies and cohort studies were included. In cases of multiple publications reporting results from the same population, the most recent report and the one with the most data was used. Case reports, case-series, reviews, letters, animal studies and studies with incomplete data were excluded. Studies that did not report separate data for OC use, but instead reported “any hormone therapy” were excluded.

Data extraction and quality appraisal of included studiesData extraction and quality appraisal were performed independently by two authors (II and MI). Details regarding the study’s author and year of publication, study design, sample size, methods of exposure assessment, methods of outcome assessment, and main findings regarding the investigated outcome were extracted. Methodological quality of studies was assessed using the Newcastle-Ottawa Scale (NOS) for quality assessment of case-control studies and cohort-studies[14]. This tool rates three categories: Selection, comparability and exposure (in case-control studies) or outcome (in cohort studies) in observational studies using a star-system. We considered studies with 8 and 9 stars as high quality, 6 and 7 stars as medium quality, and ≤ 5 stars as low quality.

Statistical analysisA meta-analysis of the comparison of nonusers vs users of OC was performed. Odds ratios (ORs), RRs, and hazard ratios (HRs) were extracted from included studies and transformed into RRs. It can be considered that OR and HR approximate RR because the absolute risk of pancreatic cancer is low[15]. For studies that did not report risk estimates for the comparison of ever vs never use of OC, we calculated ORs and RRs based on the available published data. We pooled risk estimates for pancreatic cancer and calculated overall RRs with 95%CIs. Risk effects were combined using the random-effects model[16]. The P < 0.05 was considered significant.

Heterogeneity was quantified using Cochran’s chi-square test and I2 statistic, with I2

values of 30%-60%, 50%-90%, and 75%-100% indicating moderate, substantial and considerable heterogeneity, respectively[17]. To explore heterogeneity, we performed subgroup analyses by study design (case-control and cohort), source of controls in case-control studies (population and hospital), number of cases of pancreatic cancers (< 200 and ≥ 200), and assessed study quality (NOS score ≤ 7 and > 7), geographical region (Europe, Americas, Asia), and menopausal status (premenopausal and postmenopausal). Statistically significant subgroup differences were considered for P < 0.1[18].

The pooled RRs with corresponding 95%CIs were presented graphically with forest plots. Publication bias was assessed with a funnel plot. All statistical analyses were performed using Review Manager 5 software (RevMan version 5.3, The Nordic Cochrane Centre, The Cochrane Collaboration)[19].



RESULTSLiterature search results and characteristics of included studiesThe results of the literature search are presented in Figure 1. We identified 21 studies that investigated the association between the use of OC and risk of pancreatic cancer and that fulfilled the inclusion criteria[8-11,20-36]. There were 10 case-control studies and 11 cohort studies included in the meta-analysis. Characteristics of the included studies are presented in Table 1. In total, the included studies comprised 7700 cases of pancreatic cancer. Among the case-control studies, six had population-based controls and four had hospital based-controls. The use of OC was assessed via interviews with trained interviewers in 11 studies by means of self-reported questionnaires in 9 studies, and one study used data from the National Register of Medicinal Products. Outcome assessment was verified through cancer and hospital registries, and in most of the studies, there was pathohistological verification of the diagnosis of pancreatic cancer. Seven studies were conducted in the European region, ten in the Americas (the United States and Canada), three in Asia, and one study was multicentric and conducted in all three regions. According to NOS scores, 12 studies were of high quality (3 case-control and 9 cohort), 8 of moderate quality (6 case-control and 2 cohort), and 1 case-control study was of low quality.

Ever use of OC and risk of pancreatic cancerPooled risk estimates of all included studies showed that the ever use of OC was statistically significantly associated with a decreased risk of pancreatic cancer (RR = 0.85, 95%CI: 0.73-0.98; P = 0.03) (Figure 2). There was substantial heterogeneity for this estimate (I2 = 78%; P < 0.00001). Pooled risk estimates from case-control studies were not statistically significant (RR = 0.85, 95%CI: 0.64-1.14), while the association between ever use of OC and decreased risk of pancreatic cancer was borderline significant in cohort studies (RR = 0.84, 95%CI: 0.70-1.00). Visual inspection of funnel plot (Figure 3) did not indicate any apparent presence of publication bias.

Duration of OC use and risk of pancreatic cancerEver use vs never use of OC was chosen as the primary assessment of exposure because some studies have reported that it was not possible to assess the years of oral contraceptive use in their sample[10] or simply did not inquire participants about the duration of use[8-11,23,29,34]. However, we pooled the results from the subset of studies that investigated the duration of use of OC and risk of pancreatic cancer and that reported comparable cut-off periods. Eight studies (2 case-control and 6 cohort) assessed the < 1 year duration of OC use and risk of pancreatic cancer and the pooled risk estimate was not statistically significant (RR = 1.08, 95%CI: 0.91-1.29; P = 0.38) (Figure 4A). Similarly, across one case-control and six cohort studies, the results were not significant for durations of use: < 5 years (RR = 1.07, 95%CI: 0.95-1.19; P = 0.27) (Figure 4B), 5-10 years (RR = 1.08, 95%CI: 0938-1.26; P = 0.29) (Figure 4C) and > 10 years (RR = 1.16, 95%CI: 0.92-1.45; P = 0.20) (Figure 4D).

Subgroup analysesResults of the subgroup analyses by selected characteristics are presented in Table 2. Statistically significant results for subgroup differences were noted only for geographic region (Europe vs Americas vs Asia) where studies were conducted (P = 0.07), while there were no significant subgroup differences noted for study design (case-control vs cohort studies), source of controls in case-control studies (population vs hospital), number of pancreatic cancer cases (< 200 vs ≥ 200), study quality (NOS ≤ 7 vs NOS > 7), or menopausal status (premenopausal vs postmenopausal). Finally, subgroup analyses revealed a statistically significant association between the use of OC and decreased risk of pancreatic cancer in studies of high quality (RR = 0.80, 95%CI: 0.66-0.98; P = 0.03), studies conducted in Europe (OR = 0.67, 95%CI: 0.51-0.88; P = 0.004) and in postmenopausal women (r = 0.88, 95%CI: 0.79-0.98; P = 0.02).

DISCUSSIONIn this systematic review and meta-analysis, we identified 10 case-control and 11 cohort studies that investigated the association between the use of OC and pancreatic cancer. Our pooled analysis of these 21 studies, which comprised 7700 cases of pancreatic cancer, showed that the ever use of OC was statistically significantly



Table 1 Characteristics of included studies

Study design Ref. Country Sample size1 Exposure

assessment Outcome assessmentRisk estimate (95%CI)2

Adjustments NOS score

Case-control

Bueno de Mesquita et al[20], 1992

Netherlands 82 cases and 252 controls

Interviewer administered questionnaire

Clinical diagnosis (histological verification, clinicians, laboratory records, registries)

OR = 0.83 (0.31-2.23)

Age, response status and life-time smoking of cigarettes 7

Ji et al[21], 1996 China 184 cases and 680 controls


Rapid reporting system within cancer registry

OR = 1.78 (0.91-3.47)

Age, income, education, smoking, BMI, green tea drinking, respondent status, age at first birth, intake of dietary vitamin C

7

Kreiger et al[10], 2001

Canada 52 cases and 233 controls

Mailed questionnaire Cancer registry OR = 0.36 (0.13-0.96); P = 0.031

Age, smoking status, BMI, tofu, dietary fat, coffee consumption, age at menarche, age at menopause, parity, estrogen replacement therapy age at first full term pregnancy

8

Duell et al[22], 2005

United States 241 cases and 818 controls


Physician, SEER and histologic confirmation

OR = 0.95 (0.65-1.4)

Age, education, smoking 8


Australia, Canada, The Netherlands, Poland

367 cases and 821 controls


Clinicians, hospital records, pathology records, cancer registries

OR = 0.74 (0.43-1.26)

Smoking, schooling, age, center, type of interview 8

Zhang et al[24], 2010

United States 284 cases and 1096 controls


Pathology report confirmation 3 - 6

Lucenteforte et al[8], 2011

Italy 285 cases and 713 controls


Histologically confirmed OR = 1.04 (0.55-1.98)

Study/center, age, education, area of residence, year of interview, history of diabetes, tobacco smoking

7

Azeem et al[11], 2015

Czech Republic 129 cases and 97 controls


Hospital diagnosis OR = 0.21 (0.07-0.69)

Not specified (other monitored factors) 5

Masoudi et al[9], 2017

Iran 153 cases and 202 controls


Pathological reports OR = 1.07 (0.62-1.84)

Smoking status, BMI, diabetes 6

Archibugi et al[25], 2020

Italy 253 cases and 506 controls


Histologically proven diagnosis OR = 0.52 (0.31-0.89)

Age, BMI, first degree family history of pancreatic cancer, history of diabetes > 1 yr, history of chronic pancreatitis, heavy alcohol intake, smoking habit

7

Cohort

Skinner et al[26], 243 cases and Self-reported confirmed via medical RR = 1.21 Age, time period, cigarette smoking, diabetes, BMI, height, United States Mailed questionnaire 8



2003 115474 non-cases records, pathology reports, death certificates, physicians, next of kin interview

(0.91-1.61) parity

Teras et al[27], 2005

United States 1959 cases among 387981 participants

Self-administered questionnaire

National Death Index and death certificates

3 - 8

Navarro Silvera et al[28], 2005

Canada 187 cases and 89645 noncases


Canadian Cancer Database and the National mortality database

HR = 1.10 (0.81-1.50)

Age, cigarette smoking intensity, cigarette smoking duration, BMI, height, study center, randomization group, parity

8

Prizment et al[29], 2007

United States 228 cases and 37231 noncases


National Death Index and State Health Registry of Iowa

HR = 0.90 (0.62-1.30)

Age 7

Dorjgochoo et al[30], 2009

China 78 cases among 66661 participants

In-person interview and a self-reported questionnaire

Shanghai Cancer registry and Shanghai Vital Statistics Registry

HR = 0.83 (0.45-1.55)

Education, age at menarche, number of live births, cumulative breast feeding months, BMI, exercised regularly in past 5 yr, smoking, menopausal status, first-degree family history of cancer, other contraceptive methods

9


Denmark, France, Germany, Greece, Italy, Netherlands, Norway, Spain, Sweden, United Kingdom

304 cases among 328610 participants


Population cancer registries, health insurance records, hospital-based and pathology registries

3 - 8

Lee et al[32], 2013

United States 323 cases among 118164 participants


California Cancer Registry 3 - 8

Kabat et al[33], 2017

United States 1003 cases and 157295 non-cases


Self-reports verified by physician adjudicators through records of hospitalizations, surgeries, pathology reports and procedures

HR = 0.92 (0.80-1.06)

Age, smoking status, pack-years of smoking, BMI, educational level, ethnicity, allocation to study component, diabetes

7

Andersson et al[34],2018

Sweden 110 cases and 16921 noncases

On-site questionnaires and examinations

Swedish Cancer register confirmed by pathology records, autopsy

HR = 0.68 (0.44-1.06)

Age, smoking, alcohol consumption, BMI 8

Butt et al[35], 2018

Denmark 235 cases among 1.9 million women

National Register of Medicinal Product Statistics

Danish Cancer Register and Danish National Patient Register

RR = 0.90 (0.68-1.19)

Age, year, education, PCOS, endometriosis, parity 9

Michels et al[36], 2018

United States 1000 cases and 195536 noncases

Mailed questionnaire Cancer registries HR = 1.11 (0.97-1.28)

Age, race, BMI, smoking status, alcohol use, number of cigarettes smoked per day

8

1For samples including both sexes only data for women were presented.2Ever use of oral contraceptives vs never use.3Not reported, derived from available published data.CI: Confidence interval; NOS: Newcastle-Ottawa Scale; OR: Odds ratio; RR: Relative risk; HR: Hazard ratio; BMI: Body mass index; PCOS: Polycystic ovary syndrome.

associated with a decreased risk of pancreatic cancer. However, the association was not significant when the duration of OC use less than 1 year, less than 5 years, 5-10 years, and longer than 10 years was assessed in relation to pancreatic cancer risk. A



Table 2 Association between the use of oral contraceptives and risk of pancreatic cancer: Subgroup analysis

Characteristic Subgroup Number of studies RR (95%CI) I2 within each

subgroupTest for subgroups differences

Case-control 10 0.85 (0.64-1.14)

60% (P = 0.008)Study design

Cohort 11 0.84 (0.70-1.00)

85% (P < 0.00001)

P = 0.92

Population 6 0.75 (0.47-1.21)

63% (P = 0.02)Source of controls in case-control studies

Hospital 4 0.94 (0.64-1.39)

61% (P = 0.06)

P = 0.47

< 200 8 0.80 (0.56-1.14)

66% (P = 0.004)Number of pancreatic cancer cases

≥ 200 13 0.85 (0.71-1.00)

83% (P < 0.00001)

P = 0.77

NOS ≤ 7 9 0.93 (0.75-1.17)

56% (P = 0.02)Assessed study quality

NOS > 7 12 0.80 (0.66-0.98)

84% (P < 0.00001)

P = 0.33

Europe 7 0.67 (0.51-0.88)

55% (P = 0.04)

Americas 10 0.91 (0.74-1.11)

86% (P < 0.00001)

Geographic region1

Asia 3 1.14 (0.76-1.73)

27% (P = 0.25)

P = 0.07

Premenopausal 1 0.90 (0.68-1.19)

N/AMenopausal status

Postmenopausal 4 0.88 (0.79-0.98)

17% (P = 0.31)

0.90

1One study was multicentric and conducted in all three geographic regions so it was not included in the subgroup analysis by geographic region.CI: Confidence interval; NOS: Newcastle-Ottawa Scale; RR: Relative risk.

significantly reduced risk of pancreatic cancer in women using OC was noted in higher quality studies, studies conducted in Europe, and in postmenopausal women.

Differences in pancreatic cancer incidence rates between sexes, namely a higher incidence of pancreatic cancer in men than in women[7], have led to investigations into possible reasons behind these differences. Apart from the possible influence of environmental factors, it was hypothesized that female sex hormones could be responsible for a lower incidence of pancreatic cancer in women. In vitro and in vivo studies have shown that the pancreas contains estrogen and androgen receptors and that estrogen inhibits and testosterone promotes the occurrence of some pancreatic cancers[37].

Numerous observational studies have investigated the role of the use of OC and risk of pancreatic cancer; however, the results were not consistent. While some authors have reported an inverse relationship between the use of OC and risk of pancreatic cancer[10,11,25,31,32], other studies have not confirmed these findings. However, none of the published studies found a significant positive relationship between the ever use of OC and pancreatic cancer. With regard to the duration of OC use, our pooled analyses did not identify a significant association with pancreatic cancer risk. Still, one cohort study revealed a significant increase in pancreatic cancer risk in women using OC < 1 year (HR = 1.65, 95%CI: 1.08-2.50)[28], but the number of cases of pancreatic cancer in the group of women who were taking OC less than 1 year was small. Also, one hospital-based case-control study (NOS score assessed as 6) found a borderline positive association for the duration of use of OC of 5-10 years and > 10 years and risk of pancreatic cancer[24], and, despite the small numbers of cases in these groups, the P for trend was significant (< 0.01). In contrast, Kreiger et al[10] found a significant decrease in pancreatic cancer risk in women using OC longer than 6 mo. These discrepancies in results across the studies might be explained by differences in study design, study population, assessment of exposure assessment, definitions of exposure,



Figure 1 Flow-diagram of literature search.

and different cut-offs for duration of use of OC. Additionally, most of the studies have reported risk estimates adjusted for known and potential pancreatic cancer risk factors (age, diabetes, cigarette smoking, obesity), but with fewer studies also providing estimates adjusted for history of pancreatitis, positive family history of pancreatic cancer and high level of alcohol consumption[25,30,34,36] and one study reporting estimates adjusted only for age[29]. While some of the included studies have adjusted for factors which refer to diet, this most often involved body mass index (BMI)[9,10,21,25,26,28,30,33,34,36], with only two studies investigating nutrition variables such as green tea drinking and intake of dietary vitamin C[21], and coffee and tofu consumption and dietary fat intake[10]. Obesity is a risk factor for pancreatic cancer; however, the mechanisms are not fully known and may involve sex hormones[21]. High BMI might reflect high intake of dietary fat, although the findings regarding its association with pancreatic cancer risk are inconsistent[21,34]. Notably, adipose tissue produces estrogens and might have a protective role[34]. Therefore, dietary factors could confound the association between the risk for pancreatic cancer and the use of OC[10]. Similarly, studies investigating nutrition and pancreatic cancer risk should adjust for reproductive factors such as the use of OC.

Subgroup analyses identified a significantly lower risk of pancreatic cancer in women in European region who used OC (RR = 0.67, 95%CI: 0.51-0.88). Worldwide, highest incidence of pancreatic cancer in women was noted in Northern America, Western and Northern Europe[7]. Prevalence of OC use was the highest in Northern America, followed by Europe and Asia at 75%, 69% and 68%, respectively[38]. The differences in pancreatic cancer risk associated with the use of OC between different geographic regions could be explained by differences in exposure to environmental risk factors, genetic or cultural differences. Notably, most of the studies pooled in this meta-analysis have included relevant known or potential risk factors as covariates in the adjustments of risk estimates. It is also possible that regional differences in diagnosis and outcome assessment and reporting could contribute to the observed significant difference in subgroup analyses by geographic region. Possible explan-ations for the observed inverse relationship between the use of OC and pancreatic cancer in postmenopausal women could be related to the age of menopause, duration of menopause, duration of use of OC, and formulation of used OC.

Our literature search revealed one previously published meta-analysis that assessed the association between pancreatic cancer risk and female hormonal and menstrual factors[12], and one pooled analysis from the international pancreatic cancer case-control consortium[39]. In contrast to our results, a previous meta-analysis found no



Figure 2 Forest plot of the association between the use of oral contraceptives and risk of pancreatic cancer. CI: Confidence interval.

Figure 3 Funnel plot of studies investigating the use of oral contraceptives and risk of pancreatic cancer. RR: Relative risk.

significant associations between the risk of pancreatic cancer and OC use–pooled RR from six case-control studies and eight cohort studies was 1.09 (95%CI: 0.96-1.23)[12]. Subsequently, the authors noted that their subgroup analyses by study design showed a marginally significant increased risk of pancreatic cancer associated with OC use in cohort studies (RR = 1.09, 95%CI: 1.00-1.29). Our subgroup analyses by study design identified the opposite, namely, a borderline insignificant result for inverse association (RR = 0.84, 95%CI: 0.70-1.00). However, the authors identified publication bias for studies on exposure to OC, which could mask the true association. Our meta-analysis included an additional 4 case-control studies and 3 cohort studies, totaling 7700 cases of pancreatic cancer vs 5084 in the previous meta-analysis, and our analysis did not identify publication bias. Also, our study did not have language limitations in the





Figure 4 Forest plot of the association between pancreatic cancer risk and duration of oral contraceptives use. A: Less than 1 year; B: Less than 5 years; C: 5-10 years; D: Longer than 10 years. CI: Confidence interval.

literature search in contrast to previously conducted meta-analyses. A previous meta-analysis did not assess the quality of included studies, unlike our study that also explored subgroup differences in relation to study quality and found a statistically significant reduction in pancreatic cancer risk associated with the use of OC (RR = 0.67, 95%CI: 0.51-0.88) when pooling results from studies of high quality. The previous pooled analysis that included only case-control studies did not find a significant association between the ever use of OC and pancreatic cancer (OR = 0.83, 95%CI: 0.69-1.01).

Finally, studies included in this meta-analysis have been conducted in different time periods and the influence of different formulations of OC cannot be excluded. However, only one of the included studies inquired about the type of OC used by women[35], and did not find a significant association between pancreatic cancer and use of different types of hormonal contraceptives (RR = 0.92, 95%CI: 0.62-1.36 for oral combined 20-40 ug ethinyl estradiol, and RR = 1.16, 95%CI: 0.71-1.89 for progestin only). Further on, two studies made assumptions regarding the dose of OC by using the calendar year of use as a proxy for whether women were taking high-dose or low dose formulations[32,34]. Andersson et al[34] did not find a significant association between the risk of pancreatic cancer and use of OC in 1960-1970, after 1970, 1960-1980 or after 1980. However, Lee et al[32] found that a duration dose was present for high-dose use of OC (women who stopped using OC before 1974, P for trend 0.027), and in particular the risk for pancreatic cancer was increased in women using high-dose OC ≥ 10 years (RR = 2.08, 95%CI: 1.05-4.12). Due to the inconsistent results obtained by case-control and cohort studies, it is important to address this issue when planning future observational studies in order to provide further insight into the association between the use of OC and risk of pancreatic cancer.

Strengths and limitationsTo the best of our knowledge, this represents the most comprehensive meta-analysis of observational studies which have investigated the association between the use of OC and risk of pancreatic cancer to date. Our analysis included 21 published studies with 7700 from various geographic regions which could add to the generalizability of the presented results. Also, the quality of included studies was relatively high and assessment of outcome involved pathohistological confirmation in majority of the cases.

However, our study had several limitations. First, this was a meta-analysis of observational studies and inherit limitations of study design of included studies cannot be excluded. Second, assessment of exposure was mostly based on self-report or obtained through interview, with only one study investigating the use of OC through a national medicinal registry. Further on, even though we applied a random-effects model in our meta-analyses, a high level of heterogeneity was found in some comparisons in our analyses, which we tried to explore by performing subgroup



analyses. We did not assess the association between pancreatic cancer and age at initiation of OC use, years since last use and intensity of use because only two studies reported this data and used different cut-offs. Our analyses pooled risk estimates adjusted for most potential cofactors as available in original studies, however it is not possible to exclude the influence of some confounding factors. Namely, the use of OC could be related with a higher socio-economic status which can in turn be related to a healthier diet[25]. Presence of bias due to the lack of confounding control cannot be excluded in included studies which have not considered dietary factors as possible confounding factors when investigating the association between the use of OC and risk for pancreatic cancer. Further on, for a few studies for which we needed to derive the necessary risk estimates there were no available adjusted risk estimates. Finally, an analysis of individual-patient data would have provided more precise results regarding the association between the use of OC and risk of pancreatic cancer.

CONCLUSIONEver use of OC was associated with a decreased risk of pancreatic cancer in the present meta-analysis. However, more well-designed and detailed epidemiological studies are necessary in order to fully elucidate the association between the use of OC and pancreatic cancer.

ARTICLE HIGHLIGHTSResearch backgroundPancreatic cancer is the seventh most common cause of death among malignant tumors in women. It represents one of the deadliest types of cancer with overall 5-year survival rate < 10%.

Research motivationAlthough the understanding of the etiology of pancreatic cancer has improved over the past decades and certain risk factors have been established, the causes of pancreatic cancer are still insufficiently known. Results of epidemiological studies show conflicting results regarding the association of the use of oral contraceptives (OC) and risk for pancreatic cancer.

Research objectivesThe aim of this study was to evaluate the relationship between the use of OC and risk for pancreatic cancer.

Research methodsA comprehensive literature search was performed based on defined inclusion and exclusion criteria. Quality of included observational studies was assessed and data was extracted. A meta-analysis of ever-use vs never-use of OC and risk for pancreatic cancer was performed using Review Manager 5.3. In addition, the association between the duration of use of OC and pancreatic cancer risk was also assessed, and a subgroup analysis was performed.

Research resultsA total of 7700 cases of pancreatic cancer from 21 studies (10 case-control and 11 cohort) were included in this meta-analysis. A significant association was observed between the ever-use of OC and pancreatic cancer risk (relative risk = 0.85; 95% confidence interval: 0.73-0.98), while the duration of use (< 1 year, < 5 years, 5-10 years, > 10 years) did not show a significant association. Subgroup analysis revealed a statistically significant decrease in pancreatic cancer and use of OC in high quality studies, studies conducted in Europe and in postmenopausal women.

Research conclusionsThis meta-analysis suggests a protective effect of the use of OC and pancreatic cancer occurrence, however more epidemiological studies are necessary to fully elucidate this association.



Research perspectivesFurther epidemiological studies are warranted to fully assess the association between the use of OC and risk for pancreatic cancer. These future studies investigating the risk for pancreatic cancer should be well-designed and include detailed questions regarding the use of OC.

REFERENCESInternational Agency for Research on Cancer. Cancer Incidence in Five Continents. Vol. XI (electronic version). Lyon: International Agency for Research on Cancer. [cited 20 January 2021]. Available from: http://ci5.iarc.fr

1

Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018; 68: 394-424 [PMID: 30207593 DOI: 10.3322/caac.21492]

2

Ilić M, Vlajinac H, Marinković J, Kocev N. Pancreatic cancer mortality in Serbia from 1991-2010 - a joinpoint analysis. Croat Med J 2013; 54: 369-375 [PMID: 23986278 DOI: 10.3325/cmj.2013.54.369]

3

GBD 2017 Pancreatic Cancer Collaborators. The global, regional, and national burden of pancreatic cancer and its attributable risk factors in 195 countries and territories, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Gastroenterol Hepatol 2019; 4: 934-947 [PMID: 31648972 DOI: 10.1016/S2468-1253(19)30347-4]

4

Zanini S, Renzi S, Limongi AR, Bellavite P, Giovinazzo F, Bermano G. A review of lifestyle and environment risk factors for pancreatic cancer. Eur J Cancer 2021; 145: 53-70 [PMID: 33423007 DOI: 10.1016/j.ejca.2020.11.040]

5

Lin QJ, Yang F, Jin C, Fu DL. Current status and progress of pancreatic cancer in China. World J Gastroenterol 2015; 21: 7988-8003 [PMID: 26185370 DOI: 10.3748/wjg.v21.i26.7988]

6

Ilic M, Ilic I. Epidemiology of pancreatic cancer. World J Gastroenterol 2016; 22: 9694-9705 [PMID: 27956793 DOI: 10.3748/wjg.v22.i44.9694]

7

Lucenteforte E, Zucchetto A, Bosetti C, Talamini R, Negri E, Serraino D, Franceschi S, Lipworth L, La Vecchia C. Reproductive and hormonal factors and pancreatic cancer risk in women. Pancreas 2011; 40: 460-463 [PMID: 21343831 DOI: 10.1097/MPA.0b013e31820bf986]

8

Masoudi S, Momayez Sanat Z, Mahmud Saleh A, Nozari N, Ghamarzad N, Pourshams A. Menstrual and Reproductive Factors and Risk of Pancreatic Cancer in Women. Middle East J Dig Dis 2017; 9: 146-149 [PMID: 28894516 DOI: 10.15171/mejdd.2017.65]

9

Kreiger N, Lacroix J, Sloan M. Hormonal factors and pancreatic cancer in women. Ann Epidemiol 2001; 11: 563-567 [PMID: 11709276 DOI: 10.1016/s1047-2797(01)00219-8]

10

Azeem K, Horáková D, Tomášková H, Ševčíková J, Vlčková J, Pastucha D, Procházka V, Shonová O, Martínek A, Janout V, Žídková V, Kollárová H. [A multifactor epidemiological analysis of risk factors for pancreatic cancer in women]. Epidemiol Mikrobiol Imunol 2015; 64: 34-40 [PMID: 25872994]

11

Tang B, Lv J, Li Y, Yuan S, Wang Z, He S. Relationship between female hormonal and menstrual factors and pancreatic cancer: a meta-analysis of observational studies. Medicine (Baltimore) 2015; 94: e177 [PMID: 25700305 DOI: 10.1097/MD.0000000000000177]

12

Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009; 6: e1000097 [PMID: 19621072 DOI: 10.1371/journal.pmed.1000097]

13

Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, Tugwell P. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Ottawa, ON: Ottawa Hospital Research Institute, 2014. [cited 20 January 2021]. Available from: https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp

14

Greenland S. Quantitative methods in the review of epidemiologic literature. Epidemiol Rev 1987; 9: 1-30 [PMID: 3678409 DOI: 10.1093/oxfordjournals.epirev.a036298]

15

DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 177-188 [PMID: 3802833 DOI: 10.1016/0197-2456(86)90046-2]

16

Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. The Cochrane Collaboration, 2011. [cited 20 January 2021]. Available from: https://www.handbook.cochrane.org

17

Richardson M, Garner P, Donegan S. Interpretation of subgroup analyses in systematic reviews: a tutorial. Clin Epidemiol Glob Health 2019; 7: 192-198 [DOI: 10.1016/j.cegh.2018.05.005]

18

RevMan. 2014 [Computer program] Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). Version 5.3. Copenhagen: Nordic Cochrane Centre, The Cochrane Collaboration, 2014

19

Bueno de Mesquita HB, Maisonneuve P, Moerman CJ, Walker AM. Anthropometric and reproductive variables and exocrine carcinoma of the pancreas: a population-based case-control study in The Netherlands. Int J Cancer 1992; 52: 24-29 [PMID: 1500223 DOI: 10.1002/ijc.2910520106]

20

Ji BT, Hatch MC, Chow WH, McLaughlin JK, Dai Q, Howe GR, Gao YT, Fraumeni JF Jr. 21

http://ci5.iarc.fr




https://dx.doi.org/10.3325/cmj.2013.54.369


https://dx.doi.org/10.1016/S2468-1253(19)30347-4


https://dx.doi.org/10.1016/j.ejca.2020.11.040






https://dx.doi.org/10.1097/MPA.0b013e31820bf986


https://dx.doi.org/10.15171/mejdd.2017.65


https://dx.doi.org/10.1016/s1047-2797(01)00219-8





https://dx.doi.org/10.1371/journal.pmed.1000097

https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp


https://dx.doi.org/10.1093/oxfordjournals.epirev.a036298


https://dx.doi.org/10.1016/0197-2456(86)90046-2

https://www.handbook.cochrane.org

https://dx.doi.org/10.1016/j.cegh.2018.05.005


https://dx.doi.org/10.1002/ijc.2910520106



Anthropometric and reproductive factors and the risk of pancreatic cancer: a case-control study in Shanghai, China. Int J Cancer 1996; 66: 432-437 [PMID: 8635856 DOI: 10.1002/(SICI)1097-0215(19960516)66:4<432::AID-IJC4>3.0.CO;2-X]Duell EJ, Holly EA. Reproductive and menstrual risk factors for pancreatic cancer: a population-based study of San Francisco Bay Area women. Am J Epidemiol 2005; 161: 741-747 [PMID: 15800266 DOI: 10.1093/aje/kwi104]

22

Duell EJ, Maisonneuve P, Baghurst PA, Bueno-de-Mesquita HB, Ghadirian P, Miller AB, Zatonski W, Vrieling A, Boffetta P, Boyle P. Menstrual and reproductive factors and pancreatic cancer in the SEARCH program of the IARC. Cancer Causes Control 2009; 20: 1757-1762 [PMID: 19653108 DOI: 10.1007/s10552-009-9408-x]

23

Zhang Y, Coogan PF, Palmer JR, Strom BL, Rosenberg L. A case-control study of reproductive factors, female hormone use, and risk of pancreatic cancer. Cancer Causes Control 2010; 21: 473-478 [PMID: 19941157 DOI: 10.1007/s10552-009-9478-9]

24

Archibugi L, Graglia B, Valente R, Stigliano S, Roberto M, Capalbo C, Marchetti P, Nigri G, Capurso G. Gynecological and reproductive factors and the risk of pancreatic cancer: A case-control study. Pancreatology 2020; 20: 1149-1154 [PMID: 32800645 DOI: 10.1016/j.pan.2020.07.398]

25

Skinner HG, Michaud DS, Colditz GA, Giovannucci EL, Stampfer MJ, Willett WC, Fuchs CS. Parity, reproductive factors, and the risk of pancreatic cancer in women. Cancer Epidemiol Biomarkers Prev 2003; 12: 433-438 [PMID: 12750238]

26

Teras LR, Patel AV, Rodriguez C, Thun MJ, Calle EE. Parity, other reproductive factors, and risk of pancreatic cancer mortality in a large cohort of U.S. women (United States). Cancer Causes Control 2005; 16: 1035-1040 [PMID: 16184468 DOI: 10.1007/s10552-005-0332-4]

27

Navarro Silvera SA, Miller AB, Rohan TE. Hormonal and reproductive factors and pancreatic cancer risk: a prospective cohort study. Pancreas 2005; 30: 369-374 [PMID: 15841050 DOI: 10.1097/01.mpa.0000160301.59319.ba]

28

Prizment AE, Anderson KE, Hong CP, Folsom AR. Pancreatic cancer incidence in relation to female reproductive factors: Iowa Women's Health Study. JOP 2007; 8: 16-27 [PMID: 17228129]

29

Dorjgochoo T, Shu XO, Li HL, Qian HZ, Yang G, Cai H, Gao YT, Zheng W. Use of oral contraceptives, intrauterine devices and tubal sterilization and cancer risk in a large prospective study, from 1996 to 2006. Int J Cancer 2009; 124: 2442-2449 [PMID: 19170208 DOI: 10.1002/ijc.24232]

30

Duell EJ, Travier N, Lujan-Barroso L, Dossus L, Boutron-Ruault MC, Clavel-Chapelon F, Tumino R, Masala G, Krogh V, Panico S, Ricceri F, Redondo ML, Dorronsoro M, Molina-Montes E, Huerta JM, Barricarte A, Khaw KT, Wareham NJ, Allen NE, Travis R, Siersema PD, Peeters PH, Trichopoulou A, Fragogeorgi E, Oikonomou E, Boeing H, Schuetze M, Canzian F, Lukanova A, Tjønneland A, Roswall N, Overvad K, Weiderpass E, Gram IT, Lund E, Lindkvist B, Johansen D, Ye W, Sund M, Fedirko V, Jenab M, Michaud DS, Riboli E, Bueno-de-Mesquita HB. Menstrual and reproductive factors in women, genetic variation in CYP17A1, and pancreatic cancer risk in the European prospective investigation into cancer and nutrition (EPIC) cohort. Int J Cancer 2013; 132: 2164-2175 [PMID: 23015357 DOI: 10.1002/ijc.27875]

31

Lee E, Horn-Ross PL, Rull RP, Neuhausen SL, Anton-Culver H, Ursin G, Henderson KD, Bernstein L. Reproductive factors, exogenous hormones, and pancreatic cancer risk in the CTS. Am J Epidemiol 2013; 178: 1403-1413 [PMID: 24008905 DOI: 10.1093/aje/kwt154]

32

Kabat GC, Kamensky V, Rohan TE. Reproductive factors, exogenous hormone use, and risk of pancreatic cancer in postmenopausal women. Cancer Epidemiol 2017; 49: 1-7 [PMID: 28521283 DOI: 10.1016/j.canep.2017.05.002]

33

Andersson G, Borgquist S, Jirström K. Hormonal factors and pancreatic cancer risk in women: The Malmö Diet and Cancer Study. Int J Cancer 2018; 143: 52-62 [PMID: 29424426 DOI: 10.1002/ijc.31302]

34

Butt SA, Lidegaardi Ø, Skovlund C, Hannaford PC, Iversen L, Fielding S, Mørch LS. Hormonal contraceptive use and risk of pancreatic cancer-A cohort study among premenopausal women. PLoS One 2018; 13: e0206358 [PMID: 30376560 DOI: 10.1371/journal.pone.0206358]

35

Michels KA, Brinton LA, Pfeiffer RM, Trabert B. Oral Contraceptive Use and Risks of Cancer in the NIH-AARP Diet and Health Study. Am J Epidemiol 2018; 187: 1630-1641 [PMID: 29394309 DOI: 10.1093/aje/kwx388]

36

Wahi MM, Shah N, Schrock CE, Rosemurgy AS 2nd, Goldin SB. Reproductive factors and risk of pancreatic cancer in women: a review of the literature. Ann Epidemiol 2009; 19: 103-111 [PMID: 19185803 DOI: 10.1016/j.annepidem.2008.11.003]

37

United Nations. United Nations, Department of Economic and Social Affairs, Population Division (2015). Trends in Contraceptive Use Worldwide 2015 (ST/ESA/SER.A/349). [cited 20 January 2021]. Available from: https://www.un.org/development/desa/pd/sites/www.un.org.development.desa.pd/files/undesa_pd_report_2015_trends_contraceptive_use.pdf

38

Lujan-Barroso L, Zhang W, Olson SH, Gao YT, Yu H, Baghurst PA, Bracci PM, Bueno-de-Mesquita HB, Foretová L, Gallinger S, Holcatova I, Janout V, Ji BT, Kurtz RC, La Vecchia C, Lagiou P, Li D, Miller AB, Serraino D, Zatonski W, Risch HA, Duell EJ. Menstrual and Reproductive Factors, Hormone Use, and Risk of Pancreatic Cancer: Analysis From the International Pancreatic Cancer Case-Control Consortium (PanC4). Pancreas 2016; 45: 1401-1410 [PMID: 27088489 DOI: 10.1097/MPA.0000000000000635]

39


https://dx.doi.org/10.1002/(SICI)1097-0215(19960516)66:4<432::AID-IJC4>3.0.CO;2-X


https://dx.doi.org/10.1093/aje/kwi104




https://dx.doi.org/10.1007/s10552-009-9478-9


https://dx.doi.org/10.1016/j.pan.2020.07.398



https://dx.doi.org/10.1007/s10552-005-0332-4


https://dx.doi.org/10.1097/01.mpa.0000160301.59319.ba







https://dx.doi.org/10.1093/aje/kwt154


https://dx.doi.org/10.1016/j.canep.2017.05.002






https://dx.doi.org/10.1093/aje/kwx388


https://dx.doi.org/10.1016/j.annepidem.2008.11.003

https://www.un.org/development/desa/pd/sites/www.un.org.development.desa.pd/files/undesa_pd_report_2015_trends_contraceptive_use.pdf

https://www.un.org/development/desa/pd/sites/www.un.org.development.desa.pd/files/undesa_pd_report_2015_trends_contraceptive_use.pdf




World Journal of



CASE REPORT

Cyclophosphamide-associated enteritis presenting with severe protein-losing enteropathy in granulomatosis with polyangiitis: A case report

Hiroko Sato, Tsuyoshi Shirai, Hiroshi Fujii, Tomonori Ishii, Hideo Harigae

ORCID number: Hiroko Sato 0000-0002-7139-7792; Tsuyoshi Shirai 0000-0002-6295-3494; Hiroshi Fujii 0000-0002-6885-6492; Tomonori Ishii 0000-0001-5361-5824; Hideo Harigae 0000-0003-4849-7442.

Author contributions: Sato H and Shirai T performed the literature review, and wrote the manuscript; Shirai T, Fujii H, Ishii T, and Harigae H were involved in the clinical management.

Supported by Funding for Scientific Research (Funding for Academic Research), No. 18K16136.

Informed consent statement: Written informed consent for the publication of this case report was obtained from the patient.

Conflict-of-interest statement: The authors declare that they have no conflict of interest.

CARE Checklist (2016) statement: The authors have read the CARE Checklist (2016), and the manuscript was prepared and revised according to the CARE Checklist (2016).

Open-Access: This article is an open-access article that was selected by an in-house editor and

Hiroko Sato, Tsuyoshi Shirai, Hiroshi Fujii, Tomonori Ishii, Hideo Harigae, Department of Hematology and Rheumatology, Tohoku University Graduate School of Medicine, Sendai 9808574, Japan

Corresponding author: Tsuyoshi Shirai, MD, PhD, Assistant Professor, Doctor, Department of Hematology and Rheumatology, Tohoku University Graduate School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai 9808574, Japan. [email protected]

AbstractBACKGROUND Although cyclophosphamide (CPA) is the key drug for the treatment of autoimmune diseases including vasculitides, it has some well-known adverse effects, such as myelosuppression, hemorrhagic cystitis, infertility, and infection. However, CPA-associated severe enteritis is a rare adverse effect, and only one case with a lethal clinical course has been reported. Therefore, the appropriate management of patients with CPA-associated severe enteritis is unclear.

CASE SUMMARY We present the case of a 61-year-old woman diagnosed with granulomatosis with polyangiitis based on the presence of symptoms in ear, lung, and, kidney with positive myeloperoxidase-antineutrophil cytoplasmic antibody. She received pulsed methylprednisolone followed by prednisolone 55 mg/d and intravenous CPA at a dose of 500 mg/mo. Ten days after the second course of intravenous CPA, she developed nausea, vomiting, and diarrhea, and was admitted to the hospital. Laboratory testing revealed hypoalbuminemia, suggesting protein-losing enteropathy. Computed tomography revealed wall thickening of the stomach, small intestine, and colon with contrast enhancement on the lumen side. Antibiotics and immunosuppressive therapy were not effective, and the patient’s enteritis did not improve for > 4 mo. Because her condition became seriously exhausted, corticosteroids were tapered and supportive therapies including intravenous hyperalimentation, replenishment of albumin and gamma globulin, plasma exchange, and infection control were continued. These supportive therapies improved her condition, and her enteritis gradually regressed. She was finally discharged 7 mo later.

CONCLUSION



http://orcid.org/0000-0002-7139-7792

http://orcid.org/0000-0002-7139-7792

http://orcid.org/0000-0002-7139-7792

http://orcid.org/0000-0002-6295-3494

http://orcid.org/0000-0002-6295-3494

http://orcid.org/0000-0002-6885-6492

http://orcid.org/0000-0002-6885-6492

http://orcid.org/0000-0001-5361-5824

http://orcid.org/0000-0001-5361-5824

http://orcid.org/0000-0003-4849-7442

http://orcid.org/0000-0003-4849-7442


Sato H et al. Cyclophosphamide-associated enteritis in granulomatosis with polyangiitis


fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/Licenses/by-nc/4.0/



Country/Territory of origin: Japan

Peer-review report’s scientific quality classificationGrade A (Excellent): A Grade B (Very good): B, B Grade C (Good): C, C Grade D (Fair): 0 Grade E (Poor): 0

Received: January 25, 2021 Peer-review started: January 25, 2021 First decision: February 27, 2021 Revised: March 9, 2021 Accepted: May 7, 2021 Article in press: May 7, 2021 Published online: May 28, 2021

P-Reviewer: Abe Y, Mattar MC, Rawat K S-Editor: Zhang L L-Editor: A P-Editor: Li JH

Immediate discontinuation of CPA and intensive supportive therapy are crucial for the survival of patients with CPA-associated severe enteritis.

Key Words: Antineutrophil cytoplasmic antibody; Cyclophosphamide; Enteritis; Granulomatosis with polyangiitis; Plasma exchange; Vasculitis; Case report


Core Tip: Cyclophosphamide-associated enteritis is rare, but a fatal complication of the treatment for the vasculitides. This is the first successful report of the treatment for the severe cyclophosphamide-associated enteritis, and indicates the importance of immediate discontinuation of cyclophosphamide and intensive supportive therapy.

Citation: Sato H, Shirai T, Fujii H, Ishii T, Harigae H. Cyclophosphamide-associated enteritis presenting with severe protein-losing enteropathy in granulomatosis with polyangiitis: A case report. World J Gastroenterol 2021; 27(20): 2657-2663URL: https://www.wjgnet.com/1007-9327/full/v27/i20/2657.htmDOI: https://dx.doi.org/10.3748/wjg.v27.i20.2657

INTRODUCTIONCyclophosphamide (CPA) is an alkylating agent that is frequently used in vasculitides as an induction therapy. Particularly in antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis, the addition of immunosuppressants is recommended for patients with a higher five-factor score[1]. CPA is metabolized into the active form in the liver and covalently binds to the nucleus[2]. Thereafter, it manifests antiprolif-erative effects by inhibiting deoxyribonucleic acid replication[2]. Although CPA has an established therapeutic effect in vasculitides, it sometimes results in adverse effects including myelosuppression, hemorrhagic cystitis, increased susceptibility to infection, and infertility[3]. However, the development of enteritis due to CPA use in patients with autoimmune diseases is rare, with only one case report in which Yang et al[4] presented a lethal case of CPA-associated enteritis in a patient with microscopic polyangiitis. Here, we report a severe case of CPA-associated enteritis in a patient who presented with prominent protein-losing enteropathy, who was successfully treated with supportive therapy including plasma exchange (PE).

CASE PRESENTATIONChief complaintsA 61-year-old Japanese woman developed nausea, vomiting, and diarrhea.

History of present illnessThree months before the first admission, she experienced nasal congestion, aural fullness, and auditory disturbance, and was diagnosed with otitis media. Also, two months before presentation, she experienced cough worsening, and chest radiography revealed a pulmonary infiltrative shadow. Two months before presentation, she was referred to our hospital for the evaluation of pulmonary consolidation with positive myeloperoxidase-ANCA (MPO-ANCA). She was admitted to the previous hospital. She had a body temperature of 38 °C and elevated C-reactive protein level (7.7 mg/dL). Although she was treated with antibiotics, her symptoms, inflammatory markers, and chest infiltrates did not improve. Laboratory examination showed a high titer of MPO-ANCA (Table 1), and she was referred to our hospital. After admission, her creatinine level increased with proteinuria, glomerular hematuria, and granular casts, indicating the presence of rapidly progressive glomerulonephritis (RPGN). Granulomatosis with polyangiitis (GPA) was diagnosed on the basis of the presence of otitis media, sinusitis, pulmonary nodule, RPGN, and positive MPO-ANCA. She received pulsed methylprednisolone followed by prednisolone (PSL) 55 mg/d in








Table 1 Laboratory findings

First admission Second admission

Urinalysis

Protein 1+ 2+

Occult blood 1+ -

Red blood cell 10-29/HPF1 < 4/HPF

Casts +2 +3

Spot protein-creatinine ratio (g/g Cr) 0.38 0.20

WBC (/µL) 14500 6100

Hb (g/dL) 10.9 14.9

Plt (104/µL) 43.4 35.1

T-Bil (mg/dL) 0.3 0.8

AST (U/L) 25 20

ALT (U/L) 30 18

γ-GTP (U/L) 24 18

ALP (U/L) 239 102

LDH (U/L) 207 200

CK (U/L) 58 37

TP (g/dL) 7.1 4.3

Alb (g/dL) 2.6 2.5

BUN (mg/dL) 11 35

Cr (mg/dL) 0.5 0.7

CRP (mg/dL) 14.2 0.2

MPO-ANCA (U/mL) 194.0 1.0

1Deformed erythrocytes.2Casts: waxy casts, leukocyte casts.3Casts: granular casts.Alb: Albumin; ANCA: Antineutrophil cytoplasmic antibody; ALP: Alkaline phosphatase; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BUN: Blood urea nitrogen; CK: Creatine kinase; Cr: Creatinine; CRP: C-reactive protein; Hb: Hemoglobin; HPF: High-power field; LDH: Lactate dehydrogenase; MPO: Myeloperoxidase; Plt: Platelets; T-Bil: Total bilirubin; TP: Total protein; WBC: White blood cells.

combination with intravenous CPA at a dose of 500 mg/mo. Her condition significantly improved, and she was discharged 9 d after the second course of intravenous CPA when the PSL dose was 45 mg/d. She developed nausea, vomiting, and diarrhea the day after discharge, and was admitted to our hospital.

History of past illnessShe had been diagnosed with Grave’s disease since the age of 40.

Personal and family historyThe patient had no family history.

Physical examinationOn admission, her consciousness was clear, body temperature was 36.3 °C, and body pressure was 106/75 mmHg. Her abdomen was soft and flat, with pain in the left lower quadrant without defense or rebound tenderness.

Laboratory examinationsLaboratory testing indicated hypoalbuminemia and hypogammaglobulinemia, suggesting protein-losing enteropathy that resulted in the leakage of proteins from the gastrointestinal tract (Table 1).



Imaging examinationsAbdominal radiography did not show niveau or free air. Computed tomography (CT) revealed wall thickening of the stomach, small intestine, and colon with contrast enhancement on the lumen side (Figure 1A). The contrast enhancement of the outer layer of the intestine was poor, indicating edematous change. Upper gastrointestinal endoscopy showed edematous thickening of the stomach and diffusely distributed erosion throughout the descending duodenum (Figure 1B). Colonoscopy showed generalized edema and depression with erythema, mainly at the end of the ileum. The depression had no ulcerated surface.

Further diagnostic work-upBiopsy showed the absence of malignancy and vasculitis, as well as neoplastic changes including hyperplasia. Mild inflammatory cellular infiltration and some granulation tissues and ulcers were present. Cytomegalovirus staining was negative. Because the differential diagnoses included infectious enteritis, GPA-associated gastroenteritis, and drug-induced enteritis, the patient was initially treated with antibiotics including meropenem. However, hypoalbuminemia progressed. Therefore, we replenished albumin and gamma globulin and performed PE. Her condition did not improve, and her serum albumin levels remained approximately 1.5 mg/dL for > 1 mo. This clinical course ruled out the presence of an infectious etiology. Therefore, we decided to augment immunosuppression, and increased the PSL dose to 60 mg/d and added tacrolimus. Nevertheless, the enteritis did not respond to these treatments, which was contradictory to the autoimmune phenomenon, including GPA-associated enteritis. At this point, infectious and vasculitic etiologies were unlikely.

FINAL DIAGNOSISThe CPA-associated enteropathy was considered, as previously reported by Yang et al[4].

TREATMENTBecause her condition became seriously exhausted, we tapered the steroid dose and continued supportive therapy including intravenous hyperalimentation, albumin supplementation, PE, and infection control (Figure 2).

OUTCOME AND FOLLOW-UPThe supportive therapies improved the patient’s condition, and her enteritis gradually regressed (Figure 3). She was finally discharged 7 mo after the second admission and underwent rehabilitation. At 2 years from discharge, she is now in complete remission and has returned to her previous work.

DISCUSSIONThe present case was complicated with severe enteritis, which was considered to be an adverse effect of CPA. Although the patient had severe protein-losing enteropathy, her condition was improved by aggressive supportive therapies with repeated adminis-trations of albumin and globulin for several months.

CPA is one of the frequently used immunosuppressant drugs for vasculitidis. Gastrointestinal involvement of GPA is one of the conditions requiring CPA[5]. However, CPA-associated enteritis is very rare, and there exists only one case describing a lethal course in a patient with GPA[4].

When the present patient manifested abdominal symptoms, her vasculitis was in remission, which was supported by the absence of ear symptoms, normal urinalysis, normal inflammatory marker levels, and negative MPO-ANCA. In addition, there were no histologic signs of vasculitis, granulomatous colitis, or gastritis in biopsy tissues. Further, treatment with immunosuppressants did not improve her condition, which was similar to the previously described case[4,6-8]. In addition, the blood



Figure 1 Clinical images on admission. A and B: Computed tomography scans of the abdomen showing diffuse wall thickening of the stomach, small intestine, and colon; C and D: Upper gastrointestinal endoscopy images showing diffuse erosion throughout the descending duodenum; E and F: Colonoscopy images showing generalized edema and depression with erythema mainly at the end of the ileum.

culture, fecal culture, Clostridium difficile toxin, and cytomegalovirus tests were all negative. Because it was possible that these test results were false negative, antibiotics were administered, which did not improve the enteritis.

The mechanisms by which CPA causes enteritis are not well defined. In the field of oncology, CPA is likely to be associated with gastrointestinal involvement and mucositis, which may be related to the gut microbiota[9]. In a mouse model, Viaud et al[10] reported that CPA altered the microbiota composition in the small intestine and induced the translocation of selected species of gram-positive bacteria into secondary lymphoid organs.

We compared the clinical characteristics between the present case and the previous case. In the present case, symptoms appeared 38 d after the first intravenous CPA (500 mg/mo), whereas the previous case showed symptoms 2 wk after the start of oral CPA (100 mg/d). Therefore, the cumulative dose of CPA at the onset of symptoms was 1 g in the present case and 1.4 g in the previous case. Gastrointestinal lesions were distributed in the stomach, duodenum, small bowel, and colon in the present case, whereas they were limited to the small bowel and colon in the previous case. In the present case, CT showed massive wall thickening with prominent contrast enhancement in the mucosa. The outer layer showed edematous thickening. Upper gastrointestinal endoscopy showed edematous thickening of the stomach and diffusely distributed erosion throughout the descending duodenum. Colonoscopy showed generalized edema from the transverse colon to the rectum. The clinical



Figure 2 Clinical course. Alb: Albumin; IV CPA: Intravenous cyclophosphamide; mPSL: Methylprednisolone; PE: Plasma exchange; PSL: Prednisolone; TAC: Tacrolimus.

Figure 3 Improvement of computed tomography findings. The computed tomography images were taken at different time points after the second hospitalization, as indicated.

imaging findings in the previous case were similar to the diffuse mural thickening on CT and denuded and erythematous mucosa on endoscopy in the present case.

The difference from the previous report was that the present patient survived with intensive supportive therapy. In the previous case, CPA was continued for 1 mo with reduction to 50 mg/d after the complication of enteritis appeared, and then stopped. Therefore, the patient received 2.1 g CPA in total. In the present case, intravenous CPA was immediately stopped after the development of abdominal symptoms and the patient received 1 g CPA in total. Whether the cumulative dose of CPA influences the length of enteritis is uncertain, but the lower total CPA dose may be one of the reasons for the survival of the present patient. Nonetheless, this case confirmed that enteritis could develop not only with oral administration but also with intravenous injection of CPA.

With respect to treatment, no specific therapy exists for CPA-induced enteritis. However, the associated tissue injury is severe and can last for months, resulting in the loss of serum proteins including albumin and gamma globulin. Therefore, aggressive



supportive therapies including hyperalimentation, replenishment of albumin and gamma globulin, and PE in severe cases are important. Because continuation of CPA is difficult in patients who experienced CPA-induced enteritis, rituximab or azathioprine can be the alternative treatment for GPA.

CONCLUSIONIn conclusion, severe enteritis is a rare but life-threatening adverse effect of CPA. Immediate discontinuation of CPA and persistent supportive treatment are crucial for survival.

ACKNOWLEDGEMENTSThe authors thank the staff of the department of hematology and rheumatology, Tohoku University, for helpful discussions.

REFERENCESGuillevin L, Pagnoux C, Seror R, Mahr A, Mouthon L, Toumelin PL; French Vasculitis Study Group (FVSG). The Five-Factor Score revisited: assessment of prognoses of systemic necrotizing vasculitides based on the French Vasculitis Study Group (FVSG) cohort. Medicine (Baltimore) 2011; 90: 19-27 [PMID: 21200183 DOI: 10.1097/MD.0b013e318205a4c6]

1

de Jonge ME, Huitema AD, Rodenhuis S, Beijnen JH. Clinical pharmacokinetics of cyclophosphamide. Clin Pharmacokinet 2005; 44: 1135-1164 [PMID: 16231966 DOI: 10.2165/00003088-200544110-00003]

2

Ahmed AR, Hombal SM. Cyclophosphamide (Cytoxan). A review on relevant pharmacology and clinical uses. J Am Acad Dermatol 1984; 11: 1115-1126 [PMID: 6392368 DOI: 10.1016/s0190-9622(84)80193-0]

3

Yang LS, Cameron K, Papaluca T, Basnayake C, Jackett L, McKelvie P, Goodman D, Demediuk B, Bell SJ, Thompson AJ. Cyclophosphamide-associated enteritis: A rare association with severe enteritis. World J Gastroenterol 2016; 22: 8844-8848 [PMID: 27818600 DOI: 10.3748/wjg.v22.i39.8844]

4

Haworth SJ, Pusey CD. Severe intestinal involvement in Wegener's granulomatosis. Gut 1984; 25: 1296-1300 [PMID: 6500368 DOI: 10.1136/gut.25.11.1296]

5

Eriksson P, Segelmark M, Hallböök O. Frequency, Diagnosis, Treatment, and Outcome of Gastrointestinal Disease in Granulomatosis with Polyangiitis and Microscopic Polyangiitis. J Rheumatol 2018; 45: 529-537 [PMID: 29419474 DOI: 10.3899/jrheum.170249]

6

Chow FY, Hooke D, Kerr PG. Severe intestinal involvement in Wegener's granulomatosis. J Gastroenterol Hepatol 2003; 18: 749-750 [PMID: 12753164 DOI: 10.1046/j.1440-1746.2003.30472.x]

7

Müller-Ladner U. Vasculitides of the gastrointestinal tract. Best Pract Res Clin Gastroenterol 2001; 15: 59-82 [PMID: 11355901 DOI: 10.1053/bega.2000.0156]

8

Liu T, Wu Y, Wang L, Pang X, Zhao L, Yuan H, Zhang C. A More Robust Gut Microbiota in Calorie-Restricted Mice Is Associated with Attenuated Intestinal Injury Caused by the Chemotherapy Drug Cyclophosphamide. mBio 2019; 10: e02903-18 [PMID: 30862756 DOI: 10.1128/mBio.02903-18]

9

Viaud S, Saccheri F, Mignot G, Yamazaki T, Daillère R, Hannani D, Enot DP, Pfirschke C, Engblom C, Pittet MJ, Schlitzer A, Ginhoux F, Apetoh L, Chachaty E, Woerther PL, Eberl G, Bérard M, Ecobichon C, Clermont D, Bizet C, Gaboriau-Routhiau V, Cerf-Bensussan N, Opolon P, Yessaad N, Vivier E, Ryffel B, Elson CO, Doré J, Kroemer G, Lepage P, Boneca IG, Ghiringhelli F, Zitvogel L. The intestinal microbiota modulates the anticancer immune effects of cyclophosphamide. Science 2013; 342: 971-976 [PMID: 24264990 DOI: 10.1126/science.1240537]

10


https://dx.doi.org/10.1097/MD.0b013e318205a4c6


https://dx.doi.org/10.2165/00003088-200544110-00003


https://dx.doi.org/10.1016/s0190-9622(84)80193-0






https://dx.doi.org/10.3899/jrheum.170249




https://dx.doi.org/10.1053/bega.2000.0156


https://dx.doi.org/10.1128/mBio.02903-18


https://dx.doi.org/10.1126/science.1240537

Published by Baishideng Publishing Group Inc

7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA

Telephone: +1-925-3991568

E-mail: [email protected]

Help Desk: https://www.f6publishing.com/helpdesk


© 2021 Baishideng Publishing Group Inc. All rights reserved.


https://www.f6publishing.com/helpdesk


20 - World Journal of Gastroenterology

Documents