AF

M

Ohn

TTU

T

0d

3

Malignancy

ntiviral Prophylaxis and Recurrence of Hepatocellular Carcinomaollowing Liver Transplantation in Patients With Hepatitis B

.A. Zimmerman, R.M. Ghobrial, M.J. Tong, J.R. Hiatt, A.M. Cameron, and R.W. Busuttil

ABSTRACT

Background. Orthotopic liver transplantation (OLT) is a viable treatment option forpatients with hepatitis B (HBV) and concomitant hepatocellular carcinoma (HCC).However, cancer recurrence following transplantation approaches 20%. This study soughtto identify the clinical and pathological factors associated with post-OLT survival.Methods. Univariate and multivariate analyses considered the following variables:combination viral prophylaxis, HBV recurrence, tumor stage, vascular invasion, distribu-tion, nodularity, pre- and post-OLT tumor size, pre-OLT alpha-fetoprotein (AFP), Milanand UCSF criteria, and Asian race.Results. Cumulatively, HCC recurrence-free survival was 77%, 62%, and 53% at 1, 3,and 5 years, respectively, and was significantly better in patients who were free of viralrecurrence post-OLT. Similarly, patients treated with combination prophylaxis had asignificantly lower mortality than those who were not.Conclusions. Multivariate analysis revealed that AFP � 500 ng/mL, presence of vascularinvasion by explant, HBV recurrence, and combination prophylaxis were independent

predictors of HCC recurrence-free survival.

ffaa

D1r

RTHOTOPIC LIVER TRANSPLANTATION (OLT)is the preferred treatment option for patients with

epatitis B (HBV)-mediated end-stage liver disease. Origi-ally a contraindication to transplantation, 5-year survival

From the Department of Surgery, Division of Liver and Pancreasransplantation, The Pfleger Liver Institute, The Dumont-UCLAransplant Center, The David Geffen School of Medicine at theniversity of California Los Angeles, Los Angeles, California, USA.This work was supported in part by The Dumont Foundation,

he Pfleger Liver Institute, and The Joanne Barr Foundation.

041-1345/07/$–see front matteroi:10.1016/j.transproceed.2007.07.085

276

ollowing OLT for HBV-related disease is now similar to thator other indications.1 Before effective prophylaxis was avail-ble, HBV reinfection often led to early graft cirrhosis with

2-year mortality rate of 50%.2,3 The current antiviral

Address reprint requests to Ronald W. Busuttil, MD, PhD, Theumont-UCLA Transplant Center, 77-120 CHS, Box 957054,0833 Le Conte Ave, Los Angeles, CA 90095-7054. E-mail:busuttil@mednet.ucla.edu

© 2007 by Elsevier Inc. All rights reserved.360 Park Avenue South, New York, NY 10010-1710

Transplantation Proceedings, 39, 3276–3280 (2007)

p(dH

hcScUttaHiusatufdd

P

Fdrldr7pca

H

HBmI1mOftp4Lrae

D

Hdfa

aoiepiesoH(

dotatmsn

S

SGnas

RS

Tcamasr37

H

HTpssc(ptgthHOc

ANTIVIRAL PROPHYLAXIS AND HCC RECURRENCE 3277

rophylactic regimen of hepatitis B immune globulinHBIG) and lamivudine has decreased viral recurrence andramatically improved the clinical outcome after OLT forBV over the last 2 decades.4

Pathologic progression of HBV-induced liver injury toepatocellular carcinoma (HCC) is well documented.5 Withhanges in the organ allocation system in 2002 in the Unitedtates, patients with HBV and concomitant HCC are moreommonly being treated with cadaveric OLT for cure.6

nfortunately, up to 40% of patients develop recurrentumor following transplantation.7 Lamivudine mono-herapy was recently shown to delay the progression ofdvanced liver disease and subsequent development ofCC in nontransplant patients.8 Moreover, survival was

mproved with combination antiviral therapy in patientsndergoing OLT for HBV-associated disease.9 We ob-erved at our own institution that combination therapyttenuate HBV recurrence after transplantation.1 However,he effect of viral prophylaxis on tumor recurrence isnknown. The current study was undertaken to identifyactors associated with prolonged patient survival and toetermine the influence of combination HBIG and lamivu-ine therapy on HCC recurrence following OLT.

ATIENTS AND METHODS

rom 1988 to 2005, 101 patients underwent OLT for end-stage liverisease secondary to HBV with concomitant HCC. A retrospectiveeview was conducted to identify clinical variables associated withong-term overall and HCC recurrence-free survival. Patients wereivided into 3 treatment groups based on the HBV prophylaxisegimen received: HBIG and lamivudine combination therapy (n �7), HBIG monotherapy (n � 8), and no therapy (n � 15). Oneatient was treated with lamivudine monotherapy and was ex-luded from the study. Importantly, patients coinfected with hep-titis C were excluded from this study.

BIG and Lamivudine Dosing

BIG was dosed as described previously.1 HBIG (North Americaniologicals, Boca Raton, Fla, United States) was administered toost patients prior to 1998 using a fixed dosing protocol: 10,000 U

V during the anhepatic phase, then 10,000 U/d IV for 7 days, then0,000 U every month thereafter. After 1998, HBIG dosing wasodified to provide 10,000 U IV during the anhepatic phase ofLT and 2000 U/d for 6 days, then 1560 U IM with the dosing

requency titrated to trough serum titers of HbsAb IgG. Serumiters were maintained at �500 IU/L during the first 6 monthsost-OLT. From 6 to 12 months, titers were kept between 250 and99 IU/L. After 1 year, titers were held to 100 to 250 IU/L.amivudine was given orally after OLT at 150 mg/d. Patients

emain on this regimen indefinitely. Preoperative lamivudine wasdministered to patients with decompensated liver disease and/orvidence of active viral replication.

iagnosis and Evaluation

BV was diagnosed prior to OLT by HbsAg seropositivity and/oretection of HBV DNA. Recurrent HBV was diagnosed in similarashion. Preoperative evaluation of all patients included a history

nd physical exam, laboratory studies, alpha-fetoprotein (AFP), r

nd duplex ultrasound. Preoperative diagnosis of HCC was basedn abdominal computed tomography (CT) or magnetic resonance

maging (MRI) of the abdomen. Extrahepatic metastasis wasxcluded based on abdominal/chest CT, MRI, and bone scintigra-hy prior to OLT. Final diagnosis of HCC was made by patholog-

cal examination of the explanted liver. Patients who demonstratedxtrahepatic metastasis, lymph node involvement, or vascular inva-ion of the portal vein, hepatic vein, or vena cava identified by CTr MRI were not considered for transplantation. Recurrence ofCC was identified using CT, MRI, positron emission tomography

PET), and routine chest X ray.Maintenance immunosuppression consisted of either a triple-

rug regimen including cyclosporine, azathioprine, and prednisone,r dual therapy with tacrolimus and prednisone. Steroids wereapered per standard protocol, with 1 g of methylprednisolonedministered IV on the day of transplantation and tapered rapidlyo 20 mg/d over 1 week; oral prednisone was started on day 8 (20g/d) and tapered over 2 months to 5 mg/d. Beginning in 1995,

teroids were discontinued at 6 months in HBV patients who didot exhibit rejection.

tatistical Analysis

urvival analysis was performed using the Kaplan-Meier method.roup survival curves were compared using the log-rank test for

onparametric data. In addition, 12 variables were simultaneouslynalyzed for independent significance using multivariate backwardtep-down Cox proportional hazard regression.

ESULTSurvival Analysis

he study included 81 males and 19 females with HBV andoncomitant HCC who underwent OLT. Mean recipientge was 56 � 8.7 years (range, 33–73 years) with 33.5onths follow-up. Overall survival was 82%, 65%, and 57%

t 1, 3, and 5 years, respectively. Tumor recurrence-freeurvival rates were 77%, 62%, and 53% at 1, 3, and 5 years,espectively. In the combination therapy group, 1- and-year HCC recurrence-free survival rates were 88% and8% (Fig 1).

CC and HBV Recurrence After OLT

Tumor recurrence. Radiographic evidence of recurrentCC post-OLT was found in 15 of 100 patients (15%;able 1). Recurrent tumors accounted for 11 deaths. All 7atients who did not receive combination therapy, and whouffered tumor recurrence, are dead. The most commonites of tumor recurrence were liver and lung. In theombination HBIG/lamivudine group, 8 of 77 patients10.4%) experienced tumor recurrence. Conversely, 7 of 23atients (30.4%) who received no therapy or HBIG mono-herapy demonstrated HCC recurrence. In the prophylaxisroup, 19 of 77 patients (24.6%) had a stage III or IVumor. Seven of 23 patients (30%) not receiving prophylaxisad a stage III or IV tumor. Nine of the 15 patients withCC recurrence met the Milan criteria preoperatively.verall, the majority of patients (n � 41) were Child’s

lassification A. Interestingly, 8 patients did not have cir-

hosis. The average MELD score for this cohort was 32.

twptrcmvtowfrt9trrr1ra

F

naohT(vtwsp

av

tiCaM.

FrwcH(

3278 ZIMMERMAN, GHOBRIAL, TONG ET AL

HBV reinfection post-OLT. Overall, the majority of pa-ients receiving combination therapy (95%) were treatedith lamivudine pre-OLT. The average duration of therapyrior to OLT was 9.6 months. Twelve of 77 patients (15%)reated with combination therapy exhibited post-OLT viraleinfection. However, 6 of the 8 patients (75%) in theombination therapy group who developed post-OLT tu-or recurrence demonstrated HBV reemergence. Con-

ersely, 15 of 23 patients (65%) receiving HBIG mono-herapy or no therapy demonstrated HBV reinfection. Fourf the 7 patients (57%) not receiving combination therapyho developed tumor recurrence suffered HBV reinfection

ollowing transplantation. Cumulatively, mean time to viralecurrence was 11 months (n � 27). Conversely, mean timeo tumor recurrence (n � 15) was 23.7 months. Specifically,

patients had both viral and tumor recurrence followingransplantation. In this subgroup, the mean time to HBVeemergence was 11.6 months, with subsequent tumorecurrence at 27.5 months. In 8 of these 9 patients, HBVelapse preceded identification of recurrent tumor. Overall,4 of 32 patients died from causes other than HBV or HCCecurrence, including sepsis, pneumonia, pancreatic cancer,nd stroke.

actors Affecting Survival

Overall survival. Univariate analysis revealed that combi-ation antiviral therapy was associated with improved over-ll survival compared with either no treatment (P � .0001)r HBIG monotherapy (P � .0001). Recurrence of HBVad a negative impact on overall survival (P � .0061).umor stage, presence of vascular invasion, nodularity

unifocal versus multifocal), tumor distribution (unilobarersus bilobar), pre-OLT AFP level greater than or lesshan 500 ng/mL, pre- and postoperative tumor size, tumorsithin Milan or UCSF criteria, and Asian race were not

ignificant. Multivariate analysis showed that combination

ig 1. HCC recurrence-free survival by antivi-al treatment. Tumor recurrence-free survivalas dramatically better in patients treated withombination prophylaxis compared with eitherBIG monotherapy (P � .0001) or no treatment

P � .0001).

rophylaxis, presence of vascular invasion, elevated AFP,

nd bilobar distribution significantly influenced overall sur-ival (P � .05).

HCC recurrence-free survival. By univariate analysis, an-iviral combination therapy (HBIG/lamivudine) positivelynfluenced HCC recurrence-free survival (P � .0001; Table 2).onversely, HBV viral reinfection post-OLT negativelyffected HCC recurrence-free survival (P � .0004; Fig 2).ultivariate analysis revealed combination therapy (P �

0001), HBV recurrence (P � .0275), the presence of

Table 1. HCC Recurrence and Patient Characteristics

Total no. of recurrences 15/100No. of deaths 11Average MELD score 32Tumor stage

I 2II 5III 7IV 1

HBV prophylaxisNo treatment 5HBIG only 2Combination therapy 8

HBV recurrencePositive 10Negative 5

HBV recurrence (�)No treatment 2HBIG only 2Combination therapy 6

Site of recurrenceLiver 9Lung 8Other 1

Child’s classificationA 41B 30C 21

None 8

v(f

D

Sttcotitrir6iHfv

awcw(isf

HHrret(

ptahterepedadtrr

OoH

CH

CVPH

ANTIVIRAL PROPHYLAXIS AND HCC RECURRENCE 3279

ascular invasion (P � .0386), and elevated pre-OLT AFPP � .0193) had a significant impact on tumor recurrence-ree survival (Table 3).

ISCUSSION

everal risk factors have been identified which may predictumor recurrence and decreased patient survival. Data forhe International Tumor Registry suggests tumor size �5m and positive lymph nodes are associated with poorverall and tumor recurrence-free survival.10 A report fromhe United Network for Organ Sharing (UNOS) databasencluded over 48,000 transplant recipients.11 Nearly 1000 ofhese patients had HCC with a cumulative 5-year survivalate post-OLT of 48%. Not surprisingly, long-term survivals substantially better in patients who do not suffer aecurrence (64% versus 22%).6 Hemming et al12 observed a5% rate of tumor recurrence in patients with vascularnvasion. This study included 31 patients with concomitant

BV and HCC. Importantly, the overall 5-year survival rateor patients treated with antiviral prophylaxis (HBIG/lami-udine) was 72% compared with 29% in those untreated.

Previously, we reported a survival benefit for prophylacticntiviral therapy with HBIG and lamivudine in patientsith HBV after OLT.1 In that series the most commonause of death among 43 patients with concomitant HCCas viral recurrence (64%), followed by HCC recurrence

23%). The current study suggests that combination therapys associated with a reduction in viral reinfection andignificant improvement in overall and tumor recurrence-ree survival following OLT for HBV and HCC. Moreover,

Table 2. Univariate Analysis and HCC Recurrence-FreeSurvival

Hazard Ratio P

ombination therapy 0.11 �.0001BV recurrence (yes vs no) 3.28 .0004

BV reinfection may be a critical component favoringCC recurrence. Cumulatively, the incidence of tumor

ecurrence following transplantation is lower in patientseceiving combination therapy (10.4% versus 30.4%). Inter-stingly, approximately one third of the patients in eachreatment group had advanced stage tumors in the explant24.6% versus 30%).

While this is among the largest single-center series ofatients with HBV and HCC, we acknowledge its limita-ions. This is a retrospective, single-center experience over17-year time period. As such, our comparison cohort thatistorically received either no treatment or HBIG mono-herapy is small and was transplanted early in our experi-nce. Furthermore, many patients transplanted prior tooutine use of antiviral prophylaxis died postoperatively ofarly HBV reinfection. Thus, long-term follow-up is notossible in these earlier cases. However, despite the in-quality of patients in each group, multivariate analysisemonstrated a clear relationship between antiviral therapynd HCC recurrence. Finally, as our experience spans 2ecades, we do not have complete data sets for the patientsransplanted early in the mid-1980s and early 1990s prior tooutine implementation of viral DNA for diagnosis. As aesult, our ability to evaluate specific viral factors is limited.

While the specific mechanisms of tumor recurrence afterLT are controversial and beyond the scope of this study,

ur observations suggest a link between reemergence ofBV and HCC recurrence in the posttransplantation set-

Fig 2. HCC recurrence-free survival by HBVviral recurrence. HBV reinfection post-OLT sig-nificantly reduced tumor recurrence-free sur-

Table 3. Multivariate Analysis and HCC Recurrence-FreeSurvival

Hazard Ratio P

ombination therapy 0.09 �.0001ascular invasion (yes vs no) 2.58 .0386reoperative AFP (�500 vs �500 ng/mL) 2.75 .0193BV recurrence (yes vs no) 2.20 .0275

vival (P � .0004).

tmopopthtitpspsmoAtcr

lHnekw

R

tA

cl

ealm

t1

r

hn

rno

w3

lpB

ct

tUs

t

hs

sa

ov

3280 ZIMMERMAN, GHOBRIAL, TONG ET AL

ing. Tumor recurrence is the result of clinically occultetastatic disease at the time of operation or dissemination

f tumor during the recipient hepatectomy.13 The tumor-romoting effect of HBV reinfection after transplantationn these microscopic metastatic deposits has not beenroved, and the role of reemergence of active HBV infec-ion is not obligatory for HCC recurrence. On the otherand, our data suggest that these 2 factors play an impor-ant role. Despite successful attenuation of viral reinfectionn the majority of patients with combination prophylaxis, 8reated patients still developed recurrent tumor. Of these 8atients, only 2 did not develop recurrent HBV, furtherupporting causality. Conversely, 3 patients not receivingrophylaxis also experienced HCC recurrence in the ab-ence of viral reinfection. Clearly HCC recurrence is aultifactorial process involving the pro-oncogenic influence

f the virus as well as characteristics specific to the tumor.s such, the actual tumor stage is prognostically impor-

ant.14 Patients with large, poorly differentiated tumors withoncomitant vascular invasion have a much higher rate ofecurrence.15

In conclusion, these data suggest that antiviral prophy-axis is associated with improved survival and reduction in

CC recurrence after transplantation. Whether the combi-ation of HBIG and lamivudine has a direct antineoplasticffect, distinct from suppression of viral replication, is notnown. A multicenter study with a larger cohort of patientsill be necessary to validate these findings.

EFERENCES

1. Anselmo DM, Ghobrial RM, Jung LC, et al: New era of liverransplantation for hepatitis B: a 17-year single-center experience.nn Surg 235:611, 20022. Steinmuller T, Seehofer D, Rayes N, et al: Increasing appli-

ability of liver transplantation for patients with hepatitis B-related

iver disease. Hepatology 35:1528, 2002 2

3. Han SH, Ofman J, Holt C, et al: An efficacy and cost-ffectiveness analysis of combination hepatitis B immune globulinnd lamivudine to prevent recurrent hepatitis B after orthotopiciver transplantation compared with hepatitis B immune globulin

onotherapy. Liver Transpl 6:741, 20004. Kim WR, Poterucha JJ, Kremers WK, et al: Outcome of liver

ransplantation for hepatitis B in the United States. Liver Transpl0:968, 20045. Yang HI, Lu SN, Liaw YF, et al: Hepatitis B e antigen and the

isk of hepatocellular carcinoma. N Engl J Med 347:168, 20026. Roayaie S, Schwartz JD, Sung MW, et al: Recurrence of

epatocellular carcinoma after liver transplant: patterns and prog-osis. Liver Transpl 10:534, 20047. Marsh JW, Dvorchik I, Subotin M, et al: The prediction of

isk of recurrence and time to recurrence of hepatocellular carci-oma after orthotopic liver transplantation: a pilot study. Hepatol-gy 26:444, 19978. Liaw YF, Sung JJ, Chow WC, et al: Lamivudine for patients

ith chronic hepatitis B and advanced liver disease. N Engl J Med51:1521, 20049. Angus PW, McCaughan GW, Gane EJ, et al: Combination

ow-dose hepatitis B immune globulin and lamivudine therapyrovides effective prophylaxis against posttransplantation hepatitis. Liver Transpl 6:429, 200010. Molmenti EP, Klintmalm GB: Liver transplantation in asso-

iation with hepatocellular carcinoma: an update of the Interna-ional Tumor Registry. Liver Transpl 8:736, 2002

11. Yoo HY, Patt CH, Geschwind JF, et al: The outcome of liverransplantation in patients with hepatocellular carcinoma in thenited States between 1988 and 2001: 5-year survival has improved

ignificantly with time. J Clin Oncol 21:4329, 200312. Hemming AW, Cattral MS, Reed AI, et al: Liver transplan-

ation for hepatocellular carcinoma. Ann Surg 233:652, 200113. Schwartz M, Konstadoulakis M, Roayaie S: Recurrence of

epatocellular carcinoma after liver transplantation: is immuno-uppression a factor? Liver Transpl 11:494, 2005

14. Shimoda M, Ghobrial RM, Carmody IC, et al: Predictors ofurvival after liver transplantation for hepatocellular carcinomassociated with hepatitis C. Liver Transpl 10:1478, 2004

15. Roayaie S, Haim MB, Emre S, et al: Comparison of surgicalutcomes for hepatocellular carcinoma in patients with hepatitis Bersus hepatitis C: a western experience. Ann Surg Oncol 7:764,

000