Comparison of the survival and tolerability of radioembolization in elderly vs. younger patients with unresectable hepatocellular carcinoma

Research Article

Comparison of the survival and tolerability of radioembolizationin elderly vs. younger patients with unresectable

hepatocellular carcinoma

Rita Golfieri1,⇑, Josè Ignacio Bilbao2, Livio Carpanese3, Roberto Cianni4, Daniele Gasparini5,Samer Ezziddin6, Philipp Marius Paprottka7, Francesco Fiore8, Alberta Cappelli1,

Macarena Rodriguez2, Giuseppe Maria Ettorre9, Adelchi Saltarelli4, Onelio Geatti5,Hojjat Ahmadzadehfar6, Alexander R. Haug7, Francesco Izzo8, Emanuela Giampalma1,

Bruno Sangro2,10, Giuseppe Pizzi3, Ermanno Notarianni4, Alessandro Vit5, Kai Wilhelm6,Tobias F. Jakobs7, Secondo Lastoria8, on behalf of the European Network on Radioembolization

with Yttrium-90 Microspheres (ENRY) study collaborators�

1Azienda Ospedaliero-Universitaria, Policlinico S. Orsola-Malpighi, Bologna, Italy; 2Clinica Universidad de Navarra, Pamplona, Spain; 3IFO ReginaElena National Cancer Institute, Rome, Italy; 4Ospedale Santa Maria Goretti, Latina, Italy; 5Azienda Ospedaliera Santa Maria della Misericordia,Udine, Italy; 6Uniklinik Bonn, Bonn, Germany; 7Ludwig-Maximilians Klinikum der Universität München, Munich, Germany; 8Istituto NazionaleDei Tumori G. Pascale, Naples, Italy; 9San Camillo Hospital, Rome, Italy; 10Centro de Investigacion Biomedica en Red de Enfermedades

Hepaticasy Digestivas (CIBEREHD), Spain

See Focus, pages 643–645

Background & Aims: The European Network on Radioemboliza-tion with Yttrium-90 resin microspheres study group (ENRY)conducted a retrospective study to evaluate the outcomes amongelderly (P70 years) and younger patients (<70 years) withunresectable hepatocellular carcinoma (HCC) who receivedradioembolization at 8 European centers.

Journal of Hepatology 20

Keywords: Elderly patients; Hepatocellular carcinoma; HCC; Radioembolization;SIRT; Safety; Tolerability; Survival.Received 20 December 2012; received in revised form 19 April 2013; accepted 10 May2013; available online 23 May 2013q DOI of original article: http://dx.doi.org/10.1016/j.jhep.2013.07.007.⇑ Corresponding author. Address: Department of Digestive Diseases and InternalMedicine, Radiology Unit, Sant’Orsola-Malpighi Hospital, Azienda Ospedaliero-Universitaria di Bologna, Via Albertoni 15, 40138 Bologna, Italy. Tel.: +39 0516362 311; fax: +39 051 6362 699.E-mail address: [email protected] (R. Golfieri).� See Addendum for collaborators.Abbreviations: HCC, hepatocellular carcinoma; RFA, radiofrequency ablation; PEI,percutaneous ethanol injection; TACE, transarterial chemoembolization; BCLC,Barcelona Clinic Liver Cancer; AEs, adverse events; INR, International NormalizedRatio; MELD, Model for End-Stage Liver Disease; SIRT, selective internal radiationtherapy; 90Y, Yttrium-90; ENRY, European Network on Radioembolization withYttrium-90; EASL, European Association for the Study of the Liver; CT, computedtomography; AFP, alpha fetoprotein; ECOG, European Cooperative OncologyGroup; 99mTc-MAA, Technetium-99m macroaggregated albumin; CTCAE,common toxicity criteria adverse events; GI, gastrointestinal; ANOVA, analysisof variance; SD, standard deviation; ALT, alanine transaminase; GBq, gigabecqu-erel; HBV, hepatitis B virus; HCV, hepatitis C virus; NASH, non-alcoholicsteatohepatitis; REILD, radioembolization-induced liver disease; GGTP, gamma-glutamyl transpeptidase.

Methods: Patients with confirmed diagnosis of unresectableHCC who either progressed following resection or locore-gional treatment and/or who were considered poor candi-dates for chemoembolization were evaluated by amultidisciplinary team for radioembolization with 90Y-resinmicrospheres (SIR-Spheres; Sirtex Medical). The survivaloutcome and all adverse events were compared betweenthe two age groups.Results: Between 2003 and 2009, 128 elderly and 197 youn-ger patients received radioembolization. Patients in bothgroups had similar demographic characteristics. Many elderlyand younger patients alike had multinodular, BCLC stage Cdisease, invading both lobes (p = 0.648). Elderly patients hada lower tumor burden, a smaller median target liver volume(p = 0.016) and appeared more likely to receive segmentaltreatment (p = 0.054). Radioembolization was equally well tol-erated in both cohorts and common procedure-relatedadverse events were predominantly grade 1–2 and of shortduration. No significant differences in survival between thegroups were found (p = 0.942) with similar median survivalin patients with early, intermediate or advanced BCLC stagedisease.Conclusions: Radioembolization appears to be as well-toleratedand effective for the elderly as it is for younger patients withunresectable HCC. Age alone should not be a discriminating factorfor the management of HCC patients.� 2013 European Association for the Study of the Liver. Publishedby Elsevier B.V. All rights reserved.

13 vol. 59 j 753–761

Research Article

Hepatocellular carcinoma (HCC) is the fifth most common can-cer in the world and its incidence is increasing [1,2] particu-larly in the elderly population, defined in this paper as thoseover the age of 70 years [3–5]. As the life expectancy improveswithin the general population, discussions on the best way tomanage ageing HCC patients have become increasingly rele-vant. The elderly tend to be considered clinically ‘fragile’ dueto comorbidities and a poorer performance status, which makethem less amenable and/or tolerant to resection, transarterialor systemic treatment [6–8]. In the past, the elderly have beenconsidered poor candidates for major surgery and non-surgicaltreatments such as radiofrequency ablation (RFA), percutaneousethanol injection (PEI), and transarterial chemoembolization(TACE). This assumption has been challenged by recent evi-dence, which suggests that the outcome of radical and/or othereffective HCC treatments is not influenced by age, when thecorrect selection of patients is adopted [5,6]. However, sincethe majority of these data come from undifferentiated groups,not distinguished by prognostic factors [1,6], these resultsremain controversial and the impact of old age per se, as anindependent factor affecting outcome, has yet to be clarified.After major hepatectomy for HCC, there is a trend towardshigher morbidity and mortality rates in the elderly comparedwith the young [9], but these differences tend not to be statis-tically significant [9–11].

TACE is widely used as a non-surgical treatment and is consid-ered to be effective in prolonging survival in patients with HCCand may be an acceptable alternative to surgery for high-riskelderly patients. The literature, however, reflects the divergentexperience with TACE with equivalent outcomes in the youngand old subjects in some studies [6,9,12,13], and poorer out-comes in the elderly in other studies [14,15].

RFA and PEI are radical therapies, which are recommended forvery early stage HCC by the most recent amendment to the Bar-celona Clinic Liver Cancer (BCLC) staging system [16]. Althoughthe published data in the elderly are limited, a large series fromJapan [17,18] has recently suggested that RFA might be as safeand as effective in elderly and non-elderly patients alike, and thatboth should be treated in the same manner.

There is, however, some evidence from the US NationalOrgan Procurement and Transplantation Network which sug-gests that age may be a key factor determining prognosisamongst the few elderly transplant recipients [19]. Overall, sur-vival for septuagenarians with liver transplants (compared withyounger patients) declined more rapidly with time when theyhave undergone transplantation (even though elderly trans-plant recipients tended to be healthier than younger transplantrecipients with a lower incidence of diabetes, lower Body MassIndices, lower International Normalized Ratios [INR], higherserum albumin levels, and a lower Model for End-Stage LiverDisease [MELD] score) [19]. Due to the greater incidence ofconfounding factors with increasing age, elderly patients areless likely to be eligible for treatment with resection and/orloco-regional therapies, regardless of disease stage, and insteadtend to be managed with systemic therapies such as sorafenib[8]. The limited published data on sorafenib in the elderlyindicated that increasing age does not appear to impact onthe tolerability of sorafenib with a similar frequency ofsorafenib-associated adverse events (AEs) and median

754 Journal of Hepatology 201

treatment duration across the age groups [20,21]. However,Morimoto and colleagues observed that those patients olderthan 75 years tend to experience more frequent side effectswith standard doses of sorafenib [20] and for those atincreased risk for thromboembolic and/or bleeding events,therapy interruptions may increase the risk of a rapid diseaseprogression [7].

Radioembolization (also known as selective internal radiationtherapy [SIRT]) has been recently confirmed as an effective andwell-tolerated therapy in intermediate- and advanced-stageHCC patients [22–26], but the effects of advancing age on the tol-erance and clinical outcomes following radioembolization inelderly patients are largely unknown with only one previous pub-lished report in a cohort with either primary or metastatic livertumors [27].

Therefore, a retrospective analysis was conducted by theEuropean Network on Radioembolization with Yttrium-90(90Y) resin microspheres (ENRY) study group to evaluate theclinical outcomes among elderly compared with youngerpatients based on the database generated by the radioemboli-zation treatment of 325 patients with unresectable HCC per-formed at eight European centers. This analysis supplementsthe data, published in Hepatology [25], from the primary anal-yses of this cohort.

Materials and methods

Patient enrollment

Local Review Board authorization was received to conduct a retrospectiveanalysis of consecutive elderly and younger patients with unresectable HCCwho received radioembolization between 25 September, 2003 and 17 Decem-ber, 2009.

Prior to treatment, patients were evaluated by multidisciplinary teams fortheir suitability for radioembolization with 90Y resin microspheres (SIR-Spheres�;Sirtex Medical Limited, Sydney, Australia). All patients in these analyses had aconfirmed diagnosis of HCC with liver-only or liver-dominant tumors, whichhad either progressed following surgical resection or loco-regional treatmentand/or who were considered poor candidates for TACE because of presence ofportal vein invasion or thrombosis or extensive tumor burden. Diagnosis ofHCC was either histologically proven or based on the European Association forthe Study of the Liver (EASL) criteria [16,28].

Baseline computed tomography (CT) scans of the abdomen and chest wereperformed in order to evaluate tumor burden, location, the volume of both thetarget tumor and liver. Laboratory blood tests, including a complete blood count,prothrombin time, liver function tests, creatinine, and alpha-fetoprotein level(AFP) measurements were obtained. Baseline functional performance status ofeach patient was determined according to the European Cooperative OncologyGroup (ECOG) criteria.

The appropriateness of radioembolization was considered by multidisci-plinary teams consisting of hepatologists, oncologists, radiotherapists, physi-cians, and radiologists. Only patients who met the following inclusioncriteria were considered for radioembolization [25]: ECOG performance statusof 0–2; an untreated life expectancy of >12 weeks; not amenable to curativetherapy (surgical resection, ablation or liver transplantation); uncompromisedpulmonary function; adequate hematologic parameters (i.e., granulocyte count1.5 � 109/L, platelets 50 � 109/L), renal function (creatinine <2.0 mg/dL), andliver function (i.e. bilirubin 62.0 mg/dL). Patients were excluded from radio-embolization if there was: evidence of any uncorrectable flow to the gastroin-testinal (GI) tract observed on angiography or Technetium-99mmacroaggregated albumin (99mTc-MAA) scan; estimated radiation dose greaterthan 30 Gy (16.5 mCi) delivered to the lungs in a single administration or50 Gy on multiple administrations; abnormal organ or bone marrow function(total bilirubin level >2.0 mg/dL in the absence of a reversible cause; serumalbumin <3.0 g/dL); limited hepatic reserve; or ascites or other clinical signsof liver failure on physical examination. The radioembolization procedure haspreviously been described [25].

3 vol. 59 j 753–761

JOURNAL OF HEPATOLOGY

All adverse events were graded using the National Cancer Institute Common Tox-icity Criteria Adverse Events Version (CTCAE) 3.0. The nature and severity of allprocedure-related events (fatigue, nausea and vomiting, abdominal pain, andfever) were evaluated from day 1 to day 7, laboratory changes from day 1 to month3 and radiation-related events (long-term fatigue, GI ulceration, and pneumonitis)from day 8 to month 3 post-radioembolization.

Statistical analysis

All statistical analyses were conducted using SAS (SAS Institute Inc., Cary NC) ver-sion 9.2 XP Pro statistical analyses software. The Kaplan-Meier product-limitmethod was used to compute non-parametric estimates of survival. The p-valuesfor continuous baseline variables were assessed by one-way ANOVA; the p-valuesfor dichotomous variables by the Fisher’s exact test, and p-values for nominalcategorical variables by the Chi-Square general association test. The Cochran-Mantel-Haenszel was used to compare the CTCAE distribution between cohorts.

Results

Patient characteristics

One hundred and twenty-eight elderly patients (39.4%; meanage: 74; range: 70–87 years) and 197 patients <70 years (60.6%;

Table 1. Baseline patient, disease and treatment characteristics among elderly (P70

Characteristic Parameter

Sex MaleFemale

Age, yr Mean ± SDRange

ECOG performance statusi 0123

Prior procedures Surgical (resection, transplant)Vascular (TACE/TAE)Ablation (PEI, RFA)Any prior procedure

Cirrhosis YesEtiology Hepatitis B

Hepatitis CChild-Pugh class A

BTumor burden (nodules) 1

2-5>5

Bilobar YesExtra-hepatic metastases Yes (lymph, bone, adrenal, pulmonary)Portal vein occlusion Patent

BranchMain

Ascites YesEncephalopathy Yes

BCLC stage ABCD

Journal of Hepatology 201

mean age: 58; range: 22–69.8 years) were included in this anal-ysis (Table 1); details of a further sub-analysis of a small cohortof 49 very elderly patients who were at least 75 years old (meanage: 78.4; range 75–87 years) are provided in SupplementaryTable 1.

All patients (whether elderly or younger) chosen forradioembolization had similar baseline liver function tests (totalbilirubin >1.5 mg/dL: 13.3% vs. 19.5%), Child-Pugh class A (85.2%vs. 80.7%), underlying cirrhosis (81.3% vs. 76.6%) and performancestatus (ECOG 0–1: 89.8% vs. 86.2%), respectively. A few patientsbeyond the inclusion criteria for radioembolization were treatedat the discretion of the physician and evaluated, including 3patients from one center with an ECOG performance status of3. A greater proportion of younger patients presented with hepa-titis B virus (HBV) infection, as compared to the elderly (7.8% vs.16.3%: p = 0.028), whereas the rates of hepatitis C virus (HCV)infection were similar between the two groups (43.1% vs.46.1%: p = 0.648). Prior procedures, such as surgery, ablation orTACE/TAE, were performed in similar proportions between thetwo groups (any prior procedure: 45.3% vs. 39.1%).

In addition, both age groups had similar proportions of BCLCstage A and B patients. BCLC stage A patients who elected toreceive radioembolization were either on the transplant waitinglist or had tumors not amenable to resection or ablation; while

years) and younger patients (<70 years).

Age ≥70 yr(n = 128)

Age <70 yr(n = 197)

p value between sub-groups

102 (79.7%)26 (20.3%)

163 (82.7%)34 (17.3%)

0.559

74.3 ± 3.9770.0-87.0

58.1 ± 8.8622.0-69.8

<0.001

70 (54.7%)45 (35.2%)13 (10.2%) 0

106 (54.1%)63 (32.1%) 24 (12.2%)3 (1.5%)i

0.486

25 (19.5%) 31 (15.7%) 0.45239 (30.5%) 50 (25.4%) 0.37315 (11.7%) 14 (7.1%) 0.16758 (45.3%) 77 (39.1%) 0.300104 (81.3%) 151 (76.6%) 0.33810 (7.8%) 32 (16.3%)i 0.02859 (46.1%) 85 (43.1%) 0.648109 (85.2%)19 (14.8%)

159 (80.7%)38 (19.3%)

0.371

34 (26.6%)50 (39.1%) 44 (34.4%)

44 (22.4%)71 (36.2%)81 (41.3%)i

0.212

65 (51.2%)i 107 (54.3%) 0.6488 (6.3%) 22 (11.2%) 0.170104 (81.3%)16 (12.5%)8 (6.3%)

145 (73.6%)28 (14.2%)24 (12.2%)

0.172

12 (10.5%)ix 25 (14.1%)xi 0.4712 (1.8%)ix 5 (2.8%)xi 0.708

21 (16.4%)35 (27.3%)72 (56.3%)0

31 (15.7%)52 (26.4%)111 (56.3%)3 (1.5%)

0.660

3 vol. 59 j 753–761 755

Table 1 (continued)

Characteristic Parameter Age ≥70 yr(n = 128)

Age <70 yr(n = 197)

p value between sub-groups

α-fetoprotein, ng/ml >400 38 (30.6%)iv 71 (37.8%)vi 0.225

Total bilirubin, mg/dl Mean ± SD>1.5 mg/dl

1.1 ± 0.4517 (13.3%)

1.1 ± 0.65ii

38 (19.5%)ii0.4190.174

Albumin, g/dl Mean ± SD<3.5 g/dl

3.6 ± 0.5151 (43.6%)viii

3.6 ± 1.1975 (41.9%)x

0.9730.810

INR Mean ± SD>1.2

1.2 ± 0.25ii

27 (21.3%)i1.2 ± 0.25iii

48 (24.7%)iii0.6660.502

ALT, IU/L Mean ± SD 57.0 ± 46.36 64.3 ± 50.78v 0.194

Creatinine, mg/dl Mean ± SD 1.0 ± 0.25i 0.9 ± 0.37vii 0.033

Occlusion of non-target arteries (GDA, etc) 76 (59.4%) 122 (61.9%) 0.644

Activity, GBq, administered MedianRange

1.50.5-4.0

1.70.3-3.9

0.078

Target treatment Whole liverRight lobe Left lobeSegmental

62 (48.4%)44 (34.4%)9 (7.0%)13 (10.2%)

94 (47.7%)75 (38.1%)19 (9.6%)9 (4.6%)

0.214

Treatment approach Single session Sequential lobar

127 (99.2%)1 (0.8%)

195 (99.0%)2 (1.0%)

1.000

Target tumor volume, ml MedianRange

167.23.0-1908

250.02.2-4000

0.006

Target liver volume, ml MedianRange

134098-3816

1470240-5566

0.008

Whole liver volume, ml MedianRange

1610 (898-3816)

1874 (859-5566)

<0.001

Number of treatments 1 23

119 (93.0%)7 (5.5%)2 (1.6%)

180 (91.4%)16 (8.1%)1 (0.5%)

0.880

Percentages calculated on available data.Missing baseline data on i1 patient; ii2 patients; iii3 patients; iv4 patients; v5 patients; vi9 patients; vii10 patients; viii11 patients; ix14 patients; x18 patients; xi20 patients.p Value for continuous variables by one-way ANOVA, p values for dichotomous variables by Fisher’s exact test, and p value for nominal categorical variables by Chi-Squaregeneral association test.

Research Article

BCLC stage B patients received radioembolization if they wereconsidered either poor candidates for radical therapy or TACE(due to bilobar and/or multiple [>5] tumors) or had disease pro-gression following TACE.

Assessment prior to radioembolization found that manyelderly and younger patients alike had multinodular (73.4% vs.77.6%), advanced BCLC stage C disease (56.3% vs. 56.3%) invadingboth lobes (51.2% vs. 54.3%). Elderly patients had a lower tumorburden in the liver (median target tumor volume: 167.2 vs.250.0 mL; p = 0.006), a smaller median target liver volume(1340 vs. 1470 mL; p = 0.008), a smaller whole liver volume(1610 vs. 1874 mL; p <0.001) and may have been more likely toreceive segmental treatment (10.2% vs. 4.6%; p = 0.069 for seg-mental compared with other treatment approaches), and thesefindings were also reflected in the very elderly cohort (Supple-mentary Table 1). Radioembolization was mostly performed asa single procedure in both elderly and younger cohorts (93.0%vs. 91.4%, respectively).

Safety and tolerability

Radioembolization was equally well tolerated in all cohorts(Table 2 and Supplementary Table 2). Common procedure-

756 Journal of Hepatology 201

related adverse events (fatigue, nausea and/or vomiting,abdominal pain, fever, and raised bilirubin) were predomi-nantly of mild-to-moderate intensity and of short duration.Of the common procedure-related events which occurred inthe elderly, none were grade P3 in the very elderly cohort(P75 years, n = 49) except one patient with grade 3 fatigueand 2 patients with grade 4 changes in bilirubin. Gastrointesti-nal (GI) ulceration was predominantly of mild or moderateseverity in both the younger and elderly patients (p = 0.320).Severe GI ulcers (grade P3) were less common in elderly thanyounger patients (0.8% vs. 2.7%).

Severe increases in total bilirubin (to grade P3) at 3 monthscompared to baseline were observed in 4.3% and 6.9% of theelderly and younger populations, respectively (p = 0.432) (Table3) and 4.2% of the very elderly (Supplementary Table 3). Agreater number of elderly patients experienced hypoalbumine-mia (p = 0.018) and elevated alanine transaminase (ALT)(p = 0.015) at 3 months, although these changes were restrictedto grade 1–2.

Of the 201 deaths recorded in the overall cohort of 325patients during a median follow-up of 10.0 months (range 0.2–48.0), 3 (0.9%) were considered to be definitely related and 11(3.4%) were considered to be probably related to the procedure.

3 vol. 59 j 753–761

Table 2. Main procedure-related clinical adverse events in the elderly (P70 years) and younger patients (<70 years) in the first 3 months post-treatment by severity(CTCAE v3).

CTCAEStudy sub-group N

CTCAE v3: number (%) of patients p value between sub-groups†All Grades Grade 1 Grade 2 Grade 3 Grade 4 Grade 5

Fatigue0.658Age ≥70 yr 128 68 (53.1%) 58 (45.3%) 8 (6.3%) 2 (1.6%) 0 0

Age <70 yr 197 109 (55.3%) 94 (47.7%) 9 (4.6%) 6 (3.0%) 0 0Nausea and/or vomiting

0.806Age ≥70 yr 128 41 (32.0%) 36 (28.1%) 5 (3.9%) 0 0 0Age <70 yr 197 63 (32.0%) 53 (26.9%) 9 (4.6%) 1 (0.5%) 0 0

Abdominal pain0.165Age ≥70 yr 128 31 (24.2%) 26 (20.3%) 5 (3.9%) 0 0 0

Age <70 yr 197 57 (28.9%) 44 (22.3%) 8 (4.1%) 5 (2.5%) 0 0Fever

0.269Age ≥70 yr 128 19 (14.8%) 17 (13.3%) 2 (1.6%) 0 0 0Age <70 yr 197 21 (10.7%) 19 (9.6%) 2 (1.0%) 0 0 0

GI ulceration0.320Age ≥70 yr 128 3 (2.3%) 0 2 (1.6%) 1 (0.8%) 0 0

Age <70 yr 197 9 (4.6%) 3 (1.5%) 1 (0.5%) 4 (2.0%) 0 1 (0.7%)CTCAE v3: Common Terminology Criteria for Adverse Events version 3.0.�p Value for CTCAE distribution comparison between cohorts by Cochran-Mantel-Haenszel row mean score test statistic.

Table 3. Comparison of laboratory toxicities in the elderly (P70 years) and younger patients (<70 years) by severity (CTCAE v3) between baseline and month 3.

CTCAEStudy sub-group

N Pre-radioembolization Month 3 Change of CTCAE grade at month 3 p value between sub-groups†All grades Grade ≥3‡ All grades Grade ≥3‡ Decreased Same Increased

Total bilirubin0.432Age ≥70 yr 117 20.5% 0 50.4% 4.3% 2.6% 59.0% 38.5%

Age <70 yr 175 24.0% 0 47.4% 6.9% 6.3% 59.4% 34.3%Albumin

0.018Age ≥70 yr 97 38.1% 0 45.4% 1.0% 10.3% 62.9% 26.8%Age <70 yr 140 37.9% 0 35.7% 0.7% 13.6% 72.9% 13.6%

ALT0.015Age ≥70 yr 109 53.2% 1.8% 57.8% 2.8% 11.0% 67.9% 21.1%

Age <70 yr 163 63.8% 1.8% 57.1% 3.7% 18.4% 70.6% 11.0%INR

0.911Age ≥70 yr 113 23.0% 0 33.6% 0 3.5% 82.3% 14.2%Age <70 yr 164 22.0% 0 29.9% 3.0% 4.3% 80.5% 15.2%

Creatinine0.906Age ≥70 yr 115 8.7% 0 13.0% 0 2.6% 89.6% 7.8%

Age <70 yr 161 8.1% 0.6% 10.6% 2.5% 1.2% 91.9% 6.8%Platelets

0.408Age ≥70 yr 102 47.1% 1.0% 52.0% 1.0% 9.8% 74.5% 15.7%Age <70 yr 166 42.8% 3.0% 53.0% 4.8% 9.0% 71.1% 19.9%

CTCAE: Common Terminology Criteria for Adverse Events version 3.0;�Differences in CTCAE grade from baseline to month 3 (month 3 minus baseline) between cohorts were assessed by the Wilcoxon rank sum test.�Differences in laboratory values between baseline and month 3 were also assessed by McNemar test regarding Grade 3–4 CTCAE (Yes/No) at month 3 vs. Grade 3–4 CTCAEat month 0, and were statistically significant (p <0.05) for total bilirubin in the age <70 year cohort (p <0.001), and showed a trend for total bilirubin in the age P70 yearcohort (p = 0.063).

JOURNAL OF HEPATOLOGY

All-cause mortality did not differ significantly between the youngand elderly on day 30 (2 ([1.0%]) vs. 0 (0.0%); p = 0.521), day 60 (8[4.1%] vs. 5 [3.9%]; p = 1.000) or day 90 after the procedure (13[6.6%] vs. 9 [7.0%]; p = 1.000).

Journal of Hepatology 201

Overall survival

Kaplan-Meier analysis revealed no significant difference in sur-vival following radioembolization between elderly and younger

3 vol. 59 j 753–761 757

Table 4. Comparison of survival by age in the elderly (P70 years) and younger patients (<70 years).

Characteristic Parameter Median overall survival (mOS), months (95% CI)†

Age ≥70 yr (n = 128) Age <70 yr (n = 197)N mOS (95% CI) p value N mOS (95% CI) p value

All patients 128 14.5 (10.6-16.8) n.a. 197 12.8 (10.8-17.9) n.a.ECOG performance status 0 70 16.6 (12.4-33.7) 0.070 106 18.4 (12.8-22.8) <0.001

1-2 58 10.7 (8.2-15.3) 87 8.3 (6.6-11.4)3 0 n.a. 3 5.2 (2.2-n.r.)

Child-Pugh class A 109 15.3 (10.9-18.4) 0.015 159 13.6 (11.4-18.8) 0.088B 19 7.4 (4.4-15.7) 38 10.3 (5.5-19.4)

Tumor burden (nodules) 1-5 84 16.8 (10.9-24.5) 0.012 115 15.9 (12.4-22.1) 0.001>5 44 10.7 (6.0-13.1) 81 9.5 (7.2-11.8)

Bilobar No 62 15.8 (10.0-19.8) 0.427 90 17.9 (11.9-29.5) 0.002Yes 65 12.4 (8.3-16.6) 107 11.2 (8.6-13.8)

Extra-hepatic metastases No 120 15.1 (10.7-17.1) 0.061 175 13.6 (11.2-18.8) 0.008Yes 8 8.3 (1.1-13.1) 22 7.2 (4.3-17.9)

Portal vein occlusion Patent 104 15.7 (10.9-18.6) 0.008 145 13.6 (10.9-19.4) 0.076Branch/main 24 8.3 (5.3-10.9) 52 10.8 (7.7-13.8)

Ascites No 102 16.8 (10.9-19.8) <0.001 152 13.6 (11.2-18.4) <0.001Yes 12 6.6 (3.9-14.5) 25 6.1 (3.4-8.6)

BCLC stage A 21 23.7 (15.1-38.1) 0.002 31 27.4 (19.4-46.8) <0.001B 35 16.9 (10.6-n.r.) 52 18.4 (12.8-22.8)C 72 10.3 (7.4-13.1) 111 9.7 (7.5-11.7)D 0 n.a. 3 5.2 (2.2-n.r.)

α-fetoprotein ≤400 ng/ml 86 18.4 (12.6-24.5) 0.001 117 18.8 (12.8-22.1) 0.017>400 ng/ml 38 8.2 (6.8-13.1) 71 10.3 (7.2-11.9)

Total bilirubin ≤1.5 mg/dl 111 15.7 (10.7-18.4) 0.048 157 14.1 (11.2-19.1) 0.023>1.5 mg/dl 17 8.3 (3.1-15.3) 38 9.5 (5.3-13.8)

INR ≤1.2 100 12.6 (10.0-16.8) 0.466 146 18.4 (13.6-22.1) <0.001>1.2 27 18.6 (7.4-31.7) 48 7.7 (5.6-9.5)

ALT ≤Median 64 18.6 (10.7-31.7) 0.073 98 15.4 (11.9-19.4) 0.106>Median 64 10.9 (7.4-14.9) 94 10.8 (8.6-15.5)

�Median survival calculated by Kaplan-Meier analysis.95% CI, 95% confidence interval; n.a., not applicable; n.r., not reached.

Research Article

patients (median 14.5 [95% CI 10.6–16.8] months vs. 12.8 [95%CI 10.8–17.9] months, respectively; p = 0.942) (Table 4 andFig. 1). Further analysis by patient age revealed that mediansurvival in the very old (P75 years; n = 49) and in patients<75 years (n = 276) was 14.9 (95% CI 8.3–23.7) months and12.8 (95% CI 10.9–15.8) months, respectively (SupplementaryTable 4); and in patients P65 years (n = 183) and <65 years(n = 142), median survival was 13.6 (95% CI 10.9–16.8) monthsand 12.8 (95% CI 10.4–17.9) months, respectively.

Median overall survival of elderly and younger patientswas similar in patients with early, BCLC stage A disease(23.7 [15.1–38.1] vs. 27.4 [19.4–46.8] months), intermediate,BCLC stage B disease (16.9 [10.6–not reached] vs. 18.4[12.8–22.8] months) or advanced BCLC stage C disease (10.3[7.4–13.1] months vs. 9.7 [7.5–11.7] months), respectively.Broadly equivalent trends were observed in the very elderly,elderly, and the young for each prognostic variable, with nosignificant differences between the two cohorts (Table 4 andSupplementary Table 4).

758 Journal of Hepatology 201

Discussion

A worldwide trend towards increased age in patients who arediagnosed with HCC has been observed [2,4,5,29] in developedcountries, where HCV and alcohol account for most cases ofHCC. The median age of patients at diagnosis is now over 60 years[30], and the proportion of elderly patients is expected toincrease, due to a number of epidemiological phenomena: (1)the spread of HCV, which is more common in the elderly patients,counterbalanced by the decreasing incidence of HBV infectiondue to vaccination campaigns [4]; (2) the rising incidence of cir-rhosis due to alcohol abuse and metabolic disorders such as‘cryptogenic’ and non-alcoholic steatohepatitis (NASH)-relatedcirrhosis [3,4,6,29]; and (3) the ‘delaying’ effect of anti-viral ther-apy on the primary as well as secondary occurrence of HCC inHBV and HCV infected cases is well documented [31]. Therefore,compared with younger patients, HCC in elderly patients is asso-ciated with different underlying liver disease with lower rates ofHBV [11,32], but a higher incidence of HCV-infection, alcohol

3 vol. 59 j 753–761

Time from radioembolization (mo)

Surv

ival

dis

tribu

tion

func

tion

0.00

0.25

0.50

0.75

1.00

0 6 12 18

24 30 36 42

48

54

n = 32

5

n = 11

6

n = 38

n = 13

p = 0.942 om 8.21 791 ry 07< egAAge ≥70 yr om 5.41 821

(10.8-7.9)(10.6-6.8)

)IC %59( lavivrus naideMn retemaraP

Fig. 1. Kaplan-Meier analysis of overall survival by age.

JOURNAL OF HEPATOLOGY

abuse and/or genetic susceptibility to the development of HCC[4,6,32,33]. Our results confirm a significantly lower proportionof elderly patients with hepatitis B whereas the rate of HCV-asso-ciated HCC was similar in both groups. Regarding gender, con-trary to the literature indicating a gradual increase in theproportion of the female population affected by HCC withincreasing age [9,32,34,35], our results did not show any genderprevalence between the elderly and non-elderly groups. Theexact reason for this discordance of the sex ratio in our findings,compared with previous studies, is unclear.

Evaluating the distribution of prior treatments (received by>40% of patients in this study), no statistical difference was foundin the number of vascular procedures (TACE/TAE), ablations andresections previously performed in the elderly and non-elderlygroups. This suggests that, although the elderly group was pre-sumed to poorly tolerate more intensive or invasive treatments,in reality, other factors such as performance status may comeinto play in the decision making. The Italian survey by Mirici Cap-pa and colleagues [6] indicated a trend towards the less frequentuse of aggressive therapies in the elderly, but for those patientswho were treated with TACE and RFA, the treatment was well-tolerated regardless of age. These finding are in agreement withother studies [5,33] reporting that an age beyond 70 years doesnot influence the outcome of radical and/or effective HCC treat-ments, providing that the correct selection of patients is adopted.As expected, a patient’s underlying clinical state was the majordeterminant when choosing the therapeutic modality.

Of interest, we found that elderly patients had a significantlylower median tumor burden in the liver, which more frequentlyallowed segmental treatment as compared to the youngerpatients. This difference could be related to a significantly smallermedian target liver volume, which reflected the difference inwhole liver volume. A negative correlation between age and livervolume has been previously reported and confirms that elderlypatients have smaller livers, as a consequence of ageing [36]. Inaddition, delayed tumor growth (less aggressive tumors) inelderly patients may have also been a factor, since we found thatyounger patients tended to have more advanced disease, withmore frequent extrahepatic metastasis, portal occlusion, anincreased number of nodules and higher tumor burden; althoughthere was no significant difference in survival by BCLC stagebetween the elderly and younger cohorts. In our series, the num-ber of radioembolization treatments was similar in both groupsand was equally well tolerated.

Considering that the study we performed was not a clinicaltrial and therefore the policies may be different across centers,we considered only patients with liver-only or liver-dominant

Journal of Hepatology 201

tumors, which had either progressed following surgical resec-tion or loco-regional treatment and/or patients who were con-sidered poor candidates for TACE because of presence ofportal vein invasion or thrombosis or extensive tumor burden.These data are described in Table 1, which clearly demonstratedthat the distribution of prior treatments is similar in both agegroups.

Common procedure-related AEs (fatigue, nausea and/or vom-iting, abdominal pain and fever) were predominantly grade 1–2and of short duration. A significantly greater number of elderlypatients experienced grade 1–2 hypoalbuminemia and elevatedALT at 3 months; although these changes were clinicallyirrelevant.

Safety and tolerability in our study population were similar toprevious experiences with radioembolization [23,24], where theincidence of fatigue was reported to be 54% to 61% together withabdominal pain (23–56%), nausea and/or vomiting (20–30%), andlow-grade fever (3–12%). Overall, these adverse events appear tobe significantly milder than the post-embolization syndromeobserved after conventional TACE.

In our series, an increase in total bilirubin (to grade P3) at3 months compared to baseline was observed in a similar num-ber of elderly and younger patients. Nearly 6% of the wholepatient population (almost half of them treated with whole-liver approach) had grade 3 or higher CTCAE bilirubin levelsat 3 months after therapy [25]. This is lower than the rate of14% reported up to 3 months after radioembolization with90Y-glass microspheres (mostly treated in a lobar approach)in the other large study from Chicago [23]. The current opinionis that this early increase in bilirubin levels reflects somedegree of radioembolization-induced liver disease (REILD).REILD has been described in some non-cirrhotic patients as aform of sinusoidal obstruction syndrome appearing 4–8 weeksafter radioembolization, with signs of jaundice, mild ascitesand a moderate increase in gamma-glutamyl transpeptidase(GGTP) and alkaline phosphatase, and as such is not true liverdecompensation [37,38].

Cirrhotic patients have been shown to develop a similarsyndrome for which the histological background is stillunknown [39]. This opinion is further supported by the factthat the increase in bilirubin observed after radioembolizationis distinct from other changes reflecting impaired liver functionsuch as decreased albumin levels or prothrombin activity [25].In fact, a worsening of CTCAE grade (mostly 1 or 2) of albuminafter radioembolization was observed in 27% and 14% of elderlyand young patients (p = 0.018), but relevant changes (grade 3or higher) were observed in a similar proportion of elderlyand young patients alike (1.0% and 0.7% of patients, respec-tively). In order to assess further whether advancing age wasa factor in the development of clinically significant liver func-tion changes after radioembolization, additional analyses wereconducted in patients 75 years or older. These data reveal thatthere was no significant increase from baseline in the incidenceof grade P3 ALT 3 months after radioembolization (from 2[4.5%] patients to 3 [6.8%] patients; p = 1.00) and bilirubin(from 0 to 2 [4.2%] patients; p = 0.500), and there were nopatients P75 years with grade P3 changes in albumin. Simi-larly, increases in INR CTCAE grade occurred in 14% and 15%of the elderly (P70 years) and young patients, without anygrade 3 or higher increases in prothrombin adverse eventseven in patients P75 years old. These findings emphasize the

3 vol. 59 j 753–761 759

Research Article

Concerning efficacy, our study demonstrated similar mediansurvival, regardless of age following radioembolization in theelderly and non-elderly groups. Our data confirm that, similarto other treatments [6,12,34], prognosis after radioembolizationlargely depends on pre-treatment liver function and tumorburden but not on patients’ age. Assessment of patients bydisease stage found that median overall survival did not signifi-cantly differ between the elderly (including very elderly) andyounger cohorts with advancing BCLC stage disease (Table 4and Supplementary Table 4). Assessment of each prognostic var-iable found no significant differences in overall survival betweenthe two cohorts (Table 4 and Supplementary Table 4) and con-firmed that disease stage, rather than age, was the main driverof overall survival.

A large European retrospective analysis of HCC patients hasrecently found that life expectancy of patients with HCC is unaf-fected by age, despite a higher prevalence of comorbidities and amean difference in age between the elderly and young group of14 years; thereby indicating that the occurrence of HCC (with alow survival [<20%] at 5 years) outweighs the impact of bothcomorbidity and age per se on life expectancy [6]. Only diseasestage was found to be an independent predictor of survival. Whenelderly and younger groups were matched for the main confound-ing factors, the prognosis for each type of treatment (RFA, PEI orTACE) was similar in elderly and younger patients, indicating thatthe treatment did not adversely impact outcome; although elderlyindividuals were more likely to receive percutaneous proceduresand less likely to receive hepatic resections or TACE [6].

When compared to TACE and sorafenib, radioembolization inthe present series provides similar survival rates across tumorstages [24,25,40,41]. However, the mildness of procedure-relatedevents after radioembolization compared with TACE, togetherwith a longer time-to-progression and similar survival times,suggests that radioembolization could replace TACE in the elderlyor in more fragile HCC patients [41]. In our opinion, an effectivesingle procedure such as radioembolization may be more accept-able to elderly patients than multiple procedures with TACE.While sorafenib also represents a good treatment option forelderly patients, the increased frequency of adverse events withage (beyond 75 years), without dose-modification, may impactthe outcome of very old patients [8].

The main limitation of our study was that analyses wereretrospective, although many patients were followed prospec-tively. The study therefore only gives information on the patientstreated. While it is difficult to provide a retrospective assessmentof the number of patients who did not meet the strict patientselection criteria for radioembolization, we have estimated thatof those referred to our tertiary care centers as potential candi-dates for radioembolization, approximately 12–15% could notbe treated with radioembolization after detailed pretreatmentwork-up due to factors including excessive hepato-pulmonaryshunting or uncorrectable vascular abnormalities, and thisproportion of patients did not differ by age. The findings of ouranalyses are consistent across the age groups suggesting thateven in this retrospective analysis, the recommendations fromthe manufacturer and expert consensus [42] were largelyfollowed (with only a few treated patients with compromisedliver function or ECOG performance status >2).

760 Journal of Hepatology 201

Our data clearly support the need for further studies using aprospective randomized design to better delineate the survivalbenefits in younger and older HCC patients, eventually by usinga propensity score analysis to determine prognostic predictorsof survival in both groups. In conclusion, for patients withunresectable HCC that meet the eligibility criteria and do not haveconcomitant disorders that would otherwise preclude treatment,radioembolization appears to be a well-tolerated and effectivetreatment option for the elderly patients, for whom tolerabilityand time in hospital may be important considerations.

Conflict of interest

RG, JIB, FTK, TFJ, and RTH have received honoraria for scientificpresentations from Sirtex Medical Ltd, Sydney, Australia.

Addendum

The European Network on Radioembolization with Y90 Micro-spheres (ENRY) Study Collaborators:

Cinzia Pettinato, Bruna Angelelli, Fabio Monari, MatteoRenzulli, Cristina Mosconi, Maria Cristina Galaverni, RenzoMazzarotto, Gilberto Gavaruzzi, Stefano Fanti (Azienda Ospedali-ero-Universitaria, Policlinico S. Orsola-Malpighi, Bologna, Italy);Giovanni Vennarecci, Roberto Santoro (General Surgery andTransplantation San Camillo Hospital, Rome, Italy); GiuseppePelle, Luca Filippi (Ospedale Santa Maria Goretti, Latina, Italy);Giudo Ferretti (Azienda Ospedaliera Santa Maria della Misericor-dia, Udine, Italy); Christiane Kuhl (Uniklinik Bonn, Germany);Peter Bartenstein, Maximilian F. Reiser, Frank Thomas Kolligs,Ralf Thorsten Hoffmann (LMU Klinikum der Universität Mün-chen, Munich, Germany).

Acknowledgements

The authors wish to thank their colleagues within the EuropeanNetwork on Radioembolization with Yttrium-90 resin micro-spheres (ENRY) group who gave their support to the realizationof this study:

Pamplona: Javier Arbizu, Alberto Benito, Jose I. Bilbao, DeliaD’Avola, Mercedes Iñarrairaegui, Macarena Rodriguez, BrunoSangro; Rome: Livio Carpanese, Giuseppe M. Ettorre, Carlo L.Maini, Michele Milella, Giuseppe Pizzi, Rosa Sciuto, GiovanniVennarecci; Bologna: Bruna Angelelli, Alberta Cappelli, EmanuelaGiampalma, Rita Golfieri, Cristina Mosconi, Matteo Renzulli, Cin-zia Pettinato, Fabio Monari, Renzo Mazzarotto; Udine: Guido Fer-retti, Daniele Gasparini, Onelio Geatti, Orfea Manazzone, GiorgioSoardo, Pierluigi Toniutto, Alessandro Vit; Latina: Oreste Bagni,Roberto Cianni, Antonio D’Agostini, Ermanno Notarianni, AdelchiSaltarelli, Rita Salvatori, Carlo Urigo; Napoli: Vittorio Albino, LuigiAloy, Cecilia Arrichiello, Roberto D’Angelo, Francesco Fiore,Francesco Izzo, Secondo Lastoria; Bonn: Hojjat Ahmadzadehfar,Samer Ezziddin, Carsten Meyer, Holger Palmedo, Hans HeinzSchild, Volker Schmitz, Kai Wilhelm; Munich: Peter Bartenstein,A Haug, Ralf T. Hoffmann, Tobias F. Jakobs, Frank T. Kolligs, Phi-lipp M. Paprottka, Christoph Trumm.

3 vol. 59 j 753–761

JOURNAL OF HEPATOLOGY

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.jhep.2013.05.025.

References

[1] El-Serag HB, Rudolph KL. Hepatocellular carcinoma: epidemiology andmolecular carcinogenesis. Gastroenterology 2007;132:2557–2576.

[2] Ozenne V, Bouattour M, Gouttè N, Vullierme MP, Ripault MP, Castelnau C,et al. Prospective evaluation of the management of hepatocellular carcinomain the elderly. Dig Liver Dis 2011;43:1001–1005.

[3] Stroffolini T, Trevisani F, Pinzello G, Brunello F, Tommasini MA, Iavarone M,et al. Changing aetiological factors of hepatocellular carcinoma and theirpotential impact on the effectiveness of surveillance. Dig Liver Dis2011;43:875–880.

[4] Teratani T, Ishikawa T, Shiratori Y, Shiina S, Yoshida H, Imamura M, et al.Hepatocellular carcinoma in elderly patients. Cancer 2002;95:816–823.

[5] Nanashima A, Abo T, Nonaka T, Fukuoka H, Hidaka S, Takeshita H, et al.Prognosis of patients with hepatocellular carcinoma after hepatic resection:are elderly patients suitable for surgery? J Surg Onc 2011;104:284–291.

[6] Mirici-Cappa F, Gramenzi A, Santi V, Zambruni A, Di Micoli A, Frigerio M,et al. Treatments for hepatocellular carcinoma in elderly patients are aseffective as in younger patients: a 20-year multicentre experience. Gut2010;59:387–396.

[7] Boehm S, Rothermundt C, Hess D, Joerger M. Antiangiogenic drugs inoncology: a focus on drug safety and the elderly – a mini-review.Gerontology 2010;56:303–309.

[8] Morimoto M, Numata K, Kondo M, Hidaka H, Takada J, Shibuya A, et al.Higher discontinuation and lower survival rates are likely in elderly Japanesepatients with advanced hepatocellular carcinoma receiving sorafenib. Hep-atol Res 2011;41:296–302.

[9] Poon RT, Fan ST, Lo CM, Liu CH, Ngan H, Ng IO, et al. Hepatocellularcarcinoma in the elderly: results of surgical and non-surgical management.Am J Gastroenterol 1999;94:2460–2466.

[10] Huang J, Li BK, Chen GH, Li JQ, Zhang YQ, Li GH, et al. Long-term outcomesand prognostic factors of elderly patients with hepatocellular carcinomaundergoing hepatectomy. J Gastrointest Surg 2009;13:1627–1635.

[11] Oishi K, Itamoto T, Kobayashi T, Oshita A, Amano H, Ohdan H, et al.Hepatectomy for hepatocellular carcinoma in elderly patients aged 75 yearsor more. J Gastrointest Surg 2009;13:695–701.

[12] Hoshida Y, Ikeda K, Kobayashi M, Suzuki Y, Tsubota A, Saitoh S, et al. Chronicliver disease in the extremely elderly of 80 years or more: clinicalcharacteristics, prognosis and patient survival analysis. J Hepatol1999;31:860–866.

[13] Yau T, Yao TJ, Chan P, Epstein RJ, Ng KK, Chok SH, et al. The outcomes ofelderly patients with hepatocellular carcinoma treated with transarterialchemoembolization. Cancer 2009;1:5507–5515.

[14] Biselli M, Forti P, Mucci F, Foschi FG, Marsigli L, Caputo F, et al.Chemoembolization versus chemotherapy in elderly patients with unresec-table hepatocellular carcinoma and contrast uptake as prognostic factor. JGerontol A Biol Sci Med Sci 1997;52:M305–M309.

[15] Mondazzi L, Bottelli R, Brambilla G, Ramboldi A, Rezakovic I, Zavaglia C, et al.Transarterial oily chemoembolization for hepatocellular carcinoma: amultivariate analysis of prognostic factors. Hepatology 1994;19:1115–1123.

[16] Llovet JM, Ducreux M, Lencioni R, Di Bisceglie AM, Galle PR, Dufour JF, et al.EASL-EORTC clinical practice guidelines: management of hepatocellularcarcinoma. J Hepatol 2012;56:908–943.

[17] Hiraoka A, Michitaka K, Horiike N, Hidaka S, Uehara T, Ichikawa S, et al.Radiofrequency ablation therapy for hepatocellular carcinoma in elderlypatients. J Gastroenterol Hepatol 2010;25:403–407.

[18] Takahashi H, Mizuta T, Kawazoe S, Eguchi Y, Kawaguchi Y, Otuka T, et al.Efficacy and safety of radiofrequency ablation for elderly hepatocellularcarcinoma patients. Hepatol Res 2010;40:997–1005.

[19] Schwartz J. Liver transplantation in septuagenarians receiving MELD excep-tion points for HCC: the national experience. Liver Transpl2012;18:1395–1396.

[20] Lencioni R, Kudo M, Ye SL, Bronowicki JP, Chen XP, Dagher L, et al. Firstinterim analysis of the GIDEON (Global Investigation of therapeutic

Journal of Hepatology 201

decisions in hepatocellular carcinoma and of its treatment with sorafeNib)non-interventional study. Int J Clin Pract 2012;66:675–683.

[21] Wong H, Tang YF, Yao TJ, Chiu J, Leung R, Chan P, et al. The outcomes andsafety of single-agent sorafenib in the treatment of elderly patients withadvanced hepatocellular carcinoma (HCC). Oncologist 2011;16:1721–1728.

[22] Carr BJ, Kondragunta V, Buch SC, Branch RA. Therapeutic equivalence insurvival for hepatic arterial chemoembolization and Yttrium 90 microspheretreatments in unresectable hepatocellular carcinoma. A two-cohort study.Cancer 2010:1–10.

[23] Salem R, Lewandowski RJ, Mulcahy MF, Riaz A, Ryu RK, Ibrahim S, et al.Radioembolization for hepatocellular carcinoma using yttrium-90 micro-spheres: a comprehensive report of long-term outcomes. Gastroenterology2010;138:52–64.

[24] Hilgard P, Hamami M, Fouly AE, Scherag A, Muller S, Ertle J, et al.Radioembolization with yttrium-90 glass microspheres in hepatocellularcarcinoma: European experience on safety and long-term survival. Hepa-tology 2010;52:1741–1749.

[25] Sangro B, Carpanese L, Cianni R, Golfieri R, Gasparini D, Ezziddin S, et al.Survival after 90Y resin microsphere radioembolization of hepatocellularcarcinoma across BCLC stages: a European evaluation. Hepatology2011;54:868–878.

[26] Mazzaferro V, Sposito C, Bhoori S, Romito R, Chiesa C, Morosi C, et al.Yttrium(90) radioembolization for intermediate-advanced hepatocarci-noma: a phase II study. Hepatology 2013;57:1826–1837.

[27] Iñarrairaegui M, Bilbao JI, Rodríguez M, Benito A, Sangro B. Liver radioemb-olization using 90Y resin microspheres in elderly patients: tolerance andoutcome. Hosp Pract 2010;38:103–109.

[28] Bruix J, Sherman M. Practice Guidelines Committee, American Associationfor the Study of Liver Diseases. Management of hepatocellular carcinoma.Hepatology 2005;42:1208–1236.

[29] Santi V, Buccione D, Di Micoli A, Fatti G, Frigerio M, Farinati F, et al. Thechanging scenario of hepatocellular carcinoma over the last two decades inItaly. J Hepatol 2012;56:397–405.

[30] McGlynn KA, London WT. Epidemiology and natural history of hepatocel-lular carcinoma. Best Pract Clin Gastroenterol 2005;19:3–23.

[31] Yang T, Lu JH, Zhai J, Lin C, Yang GS, Zhao RH, et al. High viral load isassociated with poor overall and recurrence-free survival of hepatitis Bvirus-related hepatocellular carcinoma after curative resection: a prospec-tive cohort study. Eur J Surg Oncol 2012;38:683–691.

[32] Kozyreva ON, Chi D, Clark JW, Wang H, Theall KP, Ryan DP, et al. Amulticenter retrospective study on clinical characteristics, treatment pat-terns, and outcome in elderly patients with hepatocellular carcinoma.Oncologist 2011;16:310–318.

[33] Fujii H, Itoh Y, Ohnishi N, Sakamoto M, Ohkawara T, Sawa Y, et al. Factorsassociated with the overall survival of elderly patients with hepatocellularcarcinoma. World J Gastroenterol 2012;18:1926–1932.

[34] Nomura F, Ohnishi K, Honda M, Satomura Y, Nakai T, Okuda K. Clinicalfeatures of hepatocellular carcinoma in the elderly: a study of 91 patientsolder than 70 years. Br J Cancer 1994;70:690–693.

[35] Dohmen K, Shirahama M, Miyamoto Y, Torii Y, Irie K, Ishibashi H. Differencesin survival based on the type of follow-up for the detection of hepatocellularcarcinoma: an analysis of 547 patients. Hepatol Res 2000;18:110–121.

[36] Vauthey JN, Abdalla EK, Doherty DA, Gertsch P, Fenstermacher MJ, Loyer EM,et al. Body surface area and body weight predict total liver volume inwestern adults. Liver Transpl 2002;8:233–240.

[37] Sangro B, Gil-Alzugaray B, Rodriguez J, Sola I, Martinez-Cuesta A, Viudez A,et al. Liver disease induced by radioembolization of liver tumors: descriptionand possible risk factors. Cancer 2008;112:1538–1546.

[38] Sangro B, Inarrairaegui M, Bilbao JI. Radioembolization for hepatocellularcarcinoma. J Hepatol 2012;56:464–473.

[39] Gil-Alzugaray B, Chopitea A, Iñarrairaegui M, Bilbao JI, Rodriguez-Fraile M,Rodriguez J, et al. Prognostic factors and prevention of radioembolization-induced liver disease. Hepatology 2013;57:1078–1087.

[40] Bruix J, Raoul JL, Sherman M, Mazzaferro V, Bolondi L, Craxi A, et al. Efficacyand safety of sorafenib in patients with advanced hepatocellular carcinoma:subanalyses of a phase III trial. J Hepatol 2012;57:821–829.

[41] Salem R, Lewandowski RJ, Kulik L, Wang E, Riaz A, Sato KT, et al.Radioembolization results in longer time-to-progression and reduced tox-icity compared with chemoembolization in patients with hepatocellularcarcinoma. Gastroenterology 2011;140:497–507.

[42] Coldwell D, Sangro B, Wasan H, Salem R, Kennedy A. General selectioncriteria of patients for radioembolization of liver tumors: an internationalworking group report. Am J Clin Oncol 2011;34:337–341.

3 vol. 59 j 753–761 761