Therapies for Advanced Stage Hepatocellular Carcinoma With Macrovascular Invasion or Metastatic Disease: A Systematic Review and Meta-analysis Richard S. Finn, 1 Andrew X. Zhu, 2 Wigdan Farah, 3 Jehad Almasri, 3 Feras Zaiem, 3 Larry J. Prokop, 3 Mohammad Hassan Murad, 3 and Khaled Mohammed 3 Hepatocellular carcinoma (HCC) is a complex disease most commonly arising in the background of chronic liver disease. In the past two decades, there has been a significant increase in our understanding of both the clinical and molecular het- erogeneity of HCC. There has been a robust increase in clinical trial activity in patients with poor prognostic factors, such as macrovascular invasion and extrahepatic spread (EHS). We aimed to synthesize the evidence for the treatment of patients with advanced HCC based on these baseline characteristics, including patients with both Child-Pugh (CP) scores of A and B. A comprehensive search of several databases from each database inception to February 15, 2016 any language was conducted. We included 14 studies (three randomized controlled studies [RCTs] and 11 observational studies). We included studies that compared sorafenib, transarterial bland embolization/transarterial chemoembolization, yttrium-90/ radiation therapy, ablation (or combination), and no therapy. Two RCTs comparing sorafenib to best supportive care demonstrated a consistent improvement in overall survival (OS) for patients with advanced HCC and metastatic vascular invasion (MVI) and/or EHS and CP A liver disease (hazard ratio, 0.66 [95% confidence interval, 0.51-0.87]; I 2 5 0%). Several observational studies evaluated locoregional therapies alone or in combination with other treatments and were lim- ited by very-low-quality of evidence. This was true for both patients with EHS and MVI. Conclusion: In patients with advanced HCC and CP A liver function, sorafenib is the only treatment that has been shown to improve OS in random- ized studies. High-quality data supporting the use of other treatment modalities in this setting, or in the setting of patients with less compensated (CP B) liver disease, are lacking. (HEPATOLOGY 2018;67:422-435) T he optimal management of hepatocellular car- cinoma (HCC) requires a multidisciplinary approach that brings together expertise in liver surgery, hepatology, interventional radiology, and medical oncology. Current recommendations for screening aim to identify smaller tumors that can be treated with curative intent with surgical resection, radiofrequency ablation (RFA), and/or liver trans- plant. (1) However, a large number of patients present with disease beyond criteria that would be considered for curative approaches. Historically there has been a large unmet need for systemic treatments in HCC. Only in 2008 were data with the multikinase inhibitor, sorafenib, shown to Abbreviations: AASLD, American Association for the Study of Liver Diseases; AFP, alpha-fetoprotein; CI, confidence interval; CP, Child-Pugh; cry- oRx, cryotherapy; EHS, extrahepatic spread; GRADE, Grading of Recommendations Assessment, Development, and Evaluation; HAIC, hepatic arte- rial infusion chemotherapy; HCC, hepatocellular carcinoma; HR, hazard ratio; MVI, metastatic vascular invasion; OR, odds ratio; OS, overall survival; PICO, population, intervention, comparison, and outcomes; PVE, portal vein embolization; PVTT, portal vein tumor thrombosis; RCTs, randomized controlled studies; RFA, radiofrequency ablation; RR, risk ratio; RT, radiation therapy (external beam); TABE, transarterial bland embo- lization; TACE, transarterial chemoembolization; TAE, transarterial embolization; Y90, yttrium-90. Received January 10, 2017; accepted June 16, 2017. Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.29486/suppinfo. Supported by a contract from the American Association for the Study of Liver Diseases (AASLD) to M. Murad (Mayo Clinic). 422 HEPATOLOGY, VOL. 67, NO. 1, 2018 A HE STUDY OF LIVER D I SE ASES T MERICAN ASSOCIATION FOR
14
Embed
Therapies for Advanced Stage Hepatocellular Carcinoma With ... · Hepatocellular carcinoma (HCC) is a complex disease most commonly arising in the background of chronic liver disease.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Therapies for Advanced StageHepatocellular Carcinoma WithMacrovascular Invasion or MetastaticDisease: A Systematic Reviewand Meta-analysisRichard S. Finn,1 Andrew X. Zhu,2 Wigdan Farah,3 Jehad Almasri,3 Feras Zaiem,3 Larry J. Prokop,3
Mohammad Hassan Murad,3 and Khaled Mohammed3
Hepatocellular carcinoma (HCC) is a complex disease most commonly arising in the background of chronic liver disease.
In the past two decades, there has been a significant increase in our understanding of both the clinical and molecular het-
erogeneity of HCC. There has been a robust increase in clinical trial activity in patients with poor prognostic factors, such
as macrovascular invasion and extrahepatic spread (EHS). We aimed to synthesize the evidence for the treatment of
patients with advanced HCC based on these baseline characteristics, including patients with both Child-Pugh (CP) scores
of A and B. A comprehensive search of several databases from each database inception to February 15, 2016 any language
was conducted. We included 14 studies (three randomized controlled studies [RCTs] and 11 observational studies). We
included studies that compared sorafenib, transarterial bland embolization/transarterial chemoembolization, yttrium-90/
radiation therapy, ablation (or combination), and no therapy. Two RCTs comparing sorafenib to best supportive care
demonstrated a consistent improvement in overall survival (OS) for patients with advanced HCC and metastatic vascular
invasion (MVI) and/or EHS and CP A liver disease (hazard ratio, 0.66 [95% confidence interval, 0.51-0.87]; I2 5 0%).
Several observational studies evaluated locoregional therapies alone or in combination with other treatments and were lim-
ited by very-low-quality of evidence. This was true for both patients with EHS and MVI. Conclusion: In patients with
advanced HCC and CP A liver function, sorafenib is the only treatment that has been shown to improve OS in random-
ized studies. High-quality data supporting the use of other treatment modalities in this setting, or in the setting of patients
with less compensated (CP B) liver disease, are lacking. (HEPATOLOGY 2018;67:422-435)
The optimal management of hepatocellular car-cinoma (HCC) requires a multidisciplinaryapproach that brings together expertise in
liver surgery, hepatology, interventional radiology, andmedical oncology. Current recommendations forscreening aim to identify smaller tumors that can betreated with curative intent with surgical resection,
radiofrequency ablation (RFA), and/or liver trans-plant.(1) However, a large number of patients presentwith disease beyond criteria that would be consideredfor curative approaches.Historically there has been a large unmet need for
systemic treatments in HCC. Only in 2008 were datawith the multikinase inhibitor, sorafenib, shown to
Abbreviations: AASLD, American Association for the Study of Liver Diseases; AFP, alpha-fetoprotein; CI, confidence interval; CP, Child-Pugh; cry-
oRx, cryotherapy; EHS, extrahepatic spread; GRADE, Grading of Recommendations Assessment, Development, and Evaluation; HAIC, hepatic arte-
Received January 10, 2017; accepted June 16, 2017.Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.29486/suppinfo.
Supported by a contract from the American Association for the Study of Liver Diseases (AASLD) to M. Murad (Mayo Clinic).
422
HEPATOLOGY, VOL. 67, NO. 1, 2018
AHE STUDY OF LIVER D I S E ASESTMERICAN ASSOCIATION FOR
improve survival greater than placebo.(2) Since thattime, despite numerous attempts to improve uponthese results, no study has demonstrated a survival ben-efit over sorafenib alone in a randomized study.(3)
The unique dual blood supply to the liver, dependenceof HCC on an arterial blood supply, and hypervascularnature of HCC have made it an attractive target forcatheter-based locoregional therapies. Although there isno standardized technique for transarterial chemoembo-lization (TACE) or universal application in terms ofnumber and frequency, the procedure generally involvesthe selective catheterization of tumor feeding arterialvessels with the subsequent infusion of cytotoxic agentsfollowed by embolizing the vessels. The definitive ran-domized controlled studies and meta-analyses that haveestablished a role for TACE in HCC were selective interms of tumor size and characteristics excluding patientswith metastatic vascular invasion (MVI) and/or extrahe-patic spread (EHS).(4,5) Since that time, there hasbeen broad application of locoregional therapies inpractice often beyond the criteria of those used inclinical studies. Radioembolization is a newer techniquethat is a catheter-based approach of delivering radiola-beled (yttrium-90; Y90) beads into the tumor bed.(6)
Studies with locoregional therapies such as TACE andradioembolization with Y90 have been performed toestablish the efficacy and safety of these approaches inpatients with MVI.We conducted this systematic review and meta-
analysis to synthesize existing evidence about effective-ness of systemic and locoregional approaches to treat-ing advanced HCC with MVI or EHS.
Materials and MethodsWe followed a predefined protocol developed by
HCC guideline and systematic review writing commit-tees of the American Association of for Liver Disease(AASLD). We reported this systematic review accord-ing to the Preferred Reporting Items for SystematicReviews and Meta-Analyses statement.(7)
ELIGIBILITY CRITERIA
We included comparative studies that enrolled adultswith Child-Pugh (CP) grade A or B cirrhosis andadvanced stage HCC with macrovascular invasion and/ormetastatic disease. We included studies that comparedsorafenib, transarterial bland embolization (TABE)/TACE, Y90/radiation therapy, ablation (or combination),and no therapy. Outcome of interest was mortality or sur-vival. We excluded studies that included CP grade C cir-rhosis, studies with advanced HCC but did not reportseparate outcomes for macrovascular invasion and/or met-astatic disease, noncomparative studies, no mortality orsurvival outcomes reported, case reports, cohorts with lessthan 5 patients, reviews, letters, errata, commentaries, andstudies published only as abstracts. Table 1 describes thepopulation, intervention, comparison, and outcomes(PICO) criteria of the two questions of interest.
SEARCH STRATEGY
A comprehensive search of several databases fromeach database inception to February 15, 2016 any
CopyrightVC 2017 by the American Association for the Study of Liver Diseases.
View this article online at wileyonlinelibrary.com.
DOI 10.1002/hep.29486
Potential conflict of interest: Dr. Zhu consults for Bristol-Myers Squibb, Eisai, Merck, Novartis, Sanofi, and Bayer. Dr. Finn consults for Bayer,
Pfizer, Novartis, Eisai, Merck, and Bristol-Myers Squibb.
ARTICLE INFORMATION:
From the 1Department of Medicine, Division of Hematology/ Oncology, Geffen School of Medicine at UCLA, Los Angeles, CA; 2Massa-
chusetts General Hospital Cancer Center and Harvard Medical School, Boston, MA; and 3Evidence-based Practice Center, Mayo Clinic
Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, MN.
language was conducted. The databases included OvidMedline In-Process & Other Non-Indexed Citations,Ovid MEDLINE, Ovid EMBASE, Ovid CochraneCentral Register of Controlled Trials, and Scopus.The search strategy was designed and conducted by anexperienced librarian with input from investigators.Supporting Table S1 shows the detailed searchstrategy.
STUDY SELECTION
Using an online reference management system(DistillerSR; Evidence Partners, Inc., Ottawa, Ontario,Canada), two reviewers were independently screenedthe titles and abstracts for potential eligibility. Full-textversions of the included abstracts were retrieved andscreened in duplicate. Disagreements were harmonizedby consensus and, if not possible by consensus, througharbitration by a third reviewer.
DATA EXTRACTION
We extracted the following variables from eachstudy: study characteristics, including primary authorand time period of study/year of publication; patientbaseline characteristics, including number of patients,age, number of lesions, size of hepatic lesions, level ofalpha-fetoprotein (AFP), stage of HCC, CP score,and cause of cirrhosis; previous treatment; interventiondetails; and outcomes of interest.
METHODOLOGICAL QUALITYAND RISK OF BIAS ASSESSMENT
We used a Cochrane risk of bias tools to assess themethodological quality of randomized controlled stud-ies (RCTs) and modified Newcastle-Ottawa Scaleto assess the methodological quality of observationalstudies. Quality of evidence was evaluated using theGrading of Recommendations Assessment, Develop-ment, and Evaluation (GRADE) approach.(8) Weused the following criteria to evaluate quality of
evidence: risk of bias; indirectness (i.e., surrogate out-comes); imprecision (i.e., wide confidence intervals[CIs]); inconsistency or heterogeneity; and publicationbias.
STATISTICAL ANALYSIS
We extracted hazard ratios (HRs) with 95% CIsand calculated risk ratios (RRs), and 95% CI usingbimanual distribution. We then pooled the log-transformed HRs/RRs using the fixed-effect modelbecause of the small number of included studies. I2
was used to assess heterogeneity, with values over 50%suggesting high heterogeneity. Statistical analyses wereconducted using Stata software (version 13; StataCorpLP, College Station, TX). Because of the small num-ber of studies, we could not assess publication bias byexamining funnel plot asymmetry or Egger’s regressiontest.
ResultsThe initial search resulted in 2,779 citations for
both questions. We eventually included 14 studies(three RCTs(2,9,10) and 11 observational studies(11-21)).Figure 1 shows the study selection process. Detailedbaseline characteristics of the studies are described inTable 2.
METHODOLOGICAL QUALITY OFTHE INCLUDED STUDIES
For RCTs (Fig. 2), all the studies reported completeoutcome data and no selective reporting. Two studiesreported random sequence generation and blinding ofthe participants and personnel. One study reportedblinding of the outcome assessment, and no studiesreported allocation concealment. For observationalstudies (Fig. 3), the overall risk of bias is high attrib-uted to unclear or high risk of bias in selection ofcohorts, outcome assessment, adequacy of follow-up,
TABLE 1. PICO of the Proposed Questions
Q1 Q2
Population Adults with CP class A or B cirrhosis andadvanced-stage HCC with macrovascular involvement
Adults with CP class A or B cirrhosis andadvanced-stage HCC with metastatic disease
Intervention vs.comparison
Sorafenib versus TABE/TACE/Y90/RT vs. ablation(or combination) vs. no therapy
Sorafenib versus TABE/TACE/Y90/RT vs. ablation(or combination) vs. no therapy
Outcomes Mortality Mortality
Study design Comparative studies Comparative studies
and lack of source of funding reporting. Assessment ofthe methodological quality for the studies included isreported in Tables 3 and 4.
Q1: INTERVENTIONS TO TREATADULTS WITH CHILD CLASS AOR B CIRRHOSIS ANDADVANCED STAGE HCC WITHMACROVASCULAR INVASION
Fourteen comparative studies compared severalinterventions to treat adults with CP class A or B cir-rhosis and advanced HCC with macrovascular involve-ment. Eleven studies(9,11,13-21) included advancedHCC patients with portal vein tumor thrombosis(PVTT). Summary of the evidence is presented inTable 5.Two RCTs(2,10) compared sorafenib versus placebo
and reported overall survival (OS) as outcome ofinterest. The majority of patients enrolled in the two
studies were CP class A cirrhosis (96.6%). Comparedto placebo, sorafenib improved OS with moderatequality of evidence (HR, 0.66 [95% CI, 0.51-0.87];I2 5 0%).One RCT(9) compared sorafenib plus cryotherapy
(cryoRx) versus sorafenib alone and enrolled 104 HCCpatients. A total of 80.9% of the patients were CP classA cirrhosis. Compared to sorafenib alone, sorafenibplus cryoRx showed no statistical significant improve-ment in 1-year survival rate with moderate quality ofevidence (RR, 1.7 [95% CI, 0.99-2.78]).One observational study(19) compared percutaneous
RFA versus control. The study enrolled 57 advancedHCC patients with PVTT. A total of 78.9% ofpatients were CP class A cirrhosis. Compared to con-trol, percutaneous RFA reduced mortality (RR, 0.81[95% CI, 0.67-0.97]). Quality of evidence was verylow downgraded attributed to imprecision and seriousrisk of bias.One observational study(20) compared TACE versus
patients with PVTT. In both CP class A (HR, 0.34[95% CI, 0.24-0.48]) and CP class B (HR, 0.26 [95%CI, 0.16-0.43]) HCC patients with cirrhosis, chemo-embolization with/without RT improved OS com-pared to sorafenib. Quality of evidence was very lowdowngraded attributed to imprecision and serious riskof bias.One observational study(13) compared TACE plus
portal vein embolization (PVE) versus TACE aloneand enrolled 116 patients. TACE plus PVE improved1-year survival compared to TACE alone (RR, 1.3[95% CI, 1.05-1.7]). No significant improvement wasnoticed in 3- and 5-year survival rate. Quality of evi-dence was very low attributed to serious risk of biasand imprecision.Several observational studies compared different inter-
vention, including: Iodine-131-lipiodol versus TACE/transarterial embolization (TAE)(17); cytotoxic chemo-therapy versus sorafenib(12); transhepatic arterial chemo-therapy versus control(21); hepatic arterial infusionchemotherapy (HAIC) plus sorafenib versus HAIC(16);sorafenib versus sorafenib plus TACE(11); radiotherapyversus sorafenib(15); and HAIC versus sorafenib.(14) Nostatistical significance improvement was noticed in sur-vival with very low quality of evidence (Table 5).
Q2: INTERVENTIONS TO TREATADULTS WITH CP CLASS A OR BCIRRHOSIS AND ADVANCED-STAGE HCC WITH METASTATICDISEASE
Four studies reported enrolled CP class A or Bpatients with cirrhosis and advanced HCC with meta-static disease and reported mortality and /or survivaloutcomes. Summary of evidence is presented inTable 6.Two RCTs(2,10) compared sorafenib versus placebo
and enrolled 311 patients. CP class A cirrhosis was96.6% of patients enrolled. Compared to placebo, sora-fenib showed no statistically significant improvementof OS with moderate quality of evidence (HR, 0.84[95% CI, 0.67-1.1]; I2 5 0%).Two observation studies(12,18) enrolled 167 patients
with advanced HCC and distant metastasis; one studycompared cytotoxic chemotherapy versus sorafenib andthe other compared chemoembolization with/withoutRT versus sorafinib. No statistically significant improve-ment in OS was noticed in both studies with very lowquality of evidence.
In this systemic review evaluating the effectivenessof systemic and locoregional treatments in patientswith advanced HCC with macrovascular invasion ormetastatic/EHS, we identified 14 studies (three RCTsand 11 observational studies). The current evidencesuggests that systemic treatment with sorafenibimproves OS when compared to no treatment. Theconfidence in this statement is supported by two ran-domized placebo-controlled studies. The use of othertreatment modalities, including combinations withsorafenib, TACE, and radioembolization, are not sup-ported with high level of evidence.
STRENGTHS AND LIMITATIONS
The data supporting systemic therapy in advancedHCC comes from large, randomized studies. In addi-tion, numerous prospective phase 3 randomized studieshave been completed and are ongoing of new systemictreatments for HCC. The results of these types ofstudies are required to establish and change standardsof care. In practice, the largest limitation has been thelack of evidence of this approach to improve survival inpatients with CP B liver disease and advanced HCC.To date, the studies with locoregional therapies havegenerally not been randomized and are observational.Whereas these studies can generate hypothesis and evi-dence for clinical decision making, the current studiesperformed to date have significant bias, limiting broadapplicability. Furthermore, since the initial literaturesearch, there have been additional studies evaluatingselective internal radiation therapy/Y90. Although mostare relatively small, retrospective/case control studies,(24-26)
recently the only prospective, randomized study comparingselective internal radiation therapy versus sorafenib (sys-temic therapy) in patients with advanced liver cancer with-out EHS that has progressed after TACE did not meet itsprimary endpoint of improving OS over sorafenib.(27)
Also, studies were not excluded based on the appropriate-ness of the control arms; for example, Yang et al. andGiorgio et al. evaluated cryotherapy and RFA, respectively,in patients with advanced HCC. Together, these datahighlight the need for prospective, randomized studies toprovide high levels of evidence on how best to integratenew approaches into the management of HCC.(28)
One aspect not captured in this analysis is tolerabil-ity and cost. Clearly, any intervention that hasimproved efficacy must be balanced by the toxicity.The side-effect profile for sorafenib is well establishedand often is accompanied by dose reductions anddelays. The decision to initiate treatment needs to bemade after a balanced review of the evidence weighedagainst the side-effect profile. In addition, the cost ofan intervention that is not associated with cure, but areduction in the risk of death, is often an issue of dis-cussion for patients and the health system.
CLINICAL AND RESEARCHIMPLICATIONS
There is clearly a need for an increased number ther-apeutic options with higher levels of evidence. Movingforward, there needs to be an emphasis on generatinghigh-quality data not only with systemic therapies, butwith locoregional therapies as well. This is feasible andlarge, randomized, prospective studies have been per-formed(22) and are ongoing.(23) Though the safety ofcatheter-based approaches has been established forpatients with CP A liver disease, more efficacy data areawaited before these approaches become routine in the
TABLE 6. Summary of Evidence for Outcomes Reported in Studies With Metastatic Disease
Intervention vs.comparison Design
Studies(n) Child-Pugh Outcome
Patients(n) ES (95% CI) GRADE
Sorafenib vs. Placebo RCTs 2 Class A (96.6%)Class B (0.4%)
Overall Survival 309 HR 0.84(0.67-1.1),
I2 5 0%
����MODERATE†
Cytotoxic chemotherapy vs.sorafenib
Observational study 1 Class A (76.1%)Class B (23.9%)
Overall Survival 66 HR 0.7(0.2-1.9)
����VERY LOW*†
Chemoembolization with/without RT vs. sorafinib
Observational study 1 Class A (64.4%)Class B (35.6%)
Overall Survival 101 HR 0.66(0.43-1.02)
����VERY LOW*†
*Methodological limitations.†Imprecision.Abbreviation: ES, effect size; RCT, randomized control trial.
HEPATOLOGY, Vol. 67, No. 1, 2018 FINN ET AL.
433
management of advanced HCC. The impact of thedegree of liver dysfunction in contributing to outcomesin advanced HCC must always be kept in mind. Thesafety of sorafenib has been established in patients withCP B liver disease and HCC, but whether it or anyintervention can improve the survival of this group ofpatients is lacking. Ongoing studies need to explorethis question for this large group of patients that areoften heterogenous in terms of survival. Finally, theclinical complexity of HCC requires a multidisciplin-ary approach and the recognition that clinical decisionscannot always be made based on available studies, butmust be individualized for any given patient takinginto account several factors, including the anatomicalstage of their disease, liver function, performance sta-tus, comorbidities, the treating centers’ level of exper-tise, and patient preferences.
REFERENCES
1) Bruix J, Sherman M, American Association for the Study of
Liver D. Management of hepatocellular carcinoma: an update.