Recurrence of hepatocellular carcinoma and hepatitis B reinfection in hepatitis B surface antigen–positive patients after liver transplantation

ORIGINAL ARTICLE

Recurrence of Hepatocellular Carcinoma andHepatitis B Reinfection in Hepatitis B SurfaceAntigen–Positive Patients After LiverTransplantationSammy Saab,1,2 Melina Yeganeh,1 Kelvin Nguyen,1 Francisco Durazo,1,2 Steven Han,1,2 Hasan Yersiz,2

Douglas G. Farmer,2 Leonard I. Goldstein,2 Myron J. Tong,1-3 and Ronald W. Busuttil2

Departments of 1Medicine and 2Surgery, University of California, Los Angeles, CA; and 3Huntington MedicalResearch Institutes, Pasadena, CA

Hepatitis B virus (HBV) reinfection and recurrence of hepatocellular carcinoma (HCC) after orthotopic liver transplantation(OLT) are associated with increased graft failure and reduced patient survival. We evaluated the effects of both HCCrecurrence and HBV reinfection on the long-term survival of these patients after OLT. One hundred seventy-five patientsunderwent OLT for HBV-related liver diseases and were the subjects of this retrospective study. We assessed risk factors forHBV reinfection, HCC recurrence, and survival post-OLT using univariate and multivariate analyses. During a mean follow-upof 43.0 � 42.0 months, 88 of 175 (50.3%) patients transplanted for HBV-related liver disease had HCC prior to OLT. Thirteen(14.8%) of these patients had HCC recurrence after OLT. The mean time for recurrence of HCC was 26.1 � 31.9 months.Twelve of 175 (6.9%) patients developed HBV reinfection after liver transplantation. The mean time for HBV reinfection was28.7 � 26.4 months. Ten of these 12 (83.3%) patients had HCC prior to OLT, and 5 (50%) developed recurrence of HCC. Onmultivariate analyses, pre-OLT HCC and recurrence of HCC post-OLT were significantly associated with HBV reinfection aftertransplantation (P � 0.031 and P � 0.001, respectively). HCC recurrence after OLT was associated with lymphovascularinvasion (P � 0.001) and post-OLT chemotherapy (P � 0.001). The 3- and 5-year survival rates were significantly decreasedin patients with HBV reinfection (P � 0.007) and in patients with HCC recurrence after OLT (P � 0.03). In conclusion, pre-OLTHCC and HCC recurrence after transplantation were associated with HBV reinfection and with decreased patient survival.Hepatitis B immunoglobulin and antiviral therapy was only partially effective in preventing HBV reinfection in patients with HCCrecurrence. Liver Transpl 15:1525-1534, 2009. © 2009 AASLD.

Received February 11, 2009; accepted July 14, 2009.

Hepatitis B virus (HBV) is the principal cause of cirrho-sis and hepatocellular carcinoma (HCC) worldwide andis responsible for at least 500,000 deaths per year.1

Prior to the availability of immunoprophylaxis and an-tiviral agents, the outcome of liver transplantation forhepatitis B–related liver disease was poor, and this of-ten led to HBV reinfection rates greater than 80% andmortality rates of 50% at 2 years.2 Prophylactic use ofhepatitis B immunoglobulin (HBIG) along with the nu-cleoside analogue lamivudine has markedly decreased

the reinfection rate of HBV by suppressing HBV repli-cation through their synergistic effect.3

However, despite reports demonstrating the reducedrisk of hepatitis B reinfection and improved survival ofpatients treated with dual prophylaxis after orthotopicliver transplantation (OLT) for HBV, approximately 10%of transplanted patients develop HBV reinfection.4 Fac-tors associated with HBV reinfection include a highpre-OLT HBV DNA level,5 hepatitis B e antigen positiv-ity,4 immunosuppression from steroids and from sys-

Abbreviations: ADV, adefovir; AFP, alpha-fetoprotein; CI, confidence interval; HBeAg, hepatitis B e antigen; HBIG, hepatitis Bimmunoglobulin; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; IFN, interferon; LAM, lamivudine;NA, not applicable; OLT, orthotopic liver transplantation; SD, standard deviation; UCSF, University of California, San Francisco.Address reprint requests to Sammy Saab, M.D., M.P.H., A.G.A.F., Pfleger Liver Institute, UCLA Medical Center, 200 Medical Plaza, Suite 214, LosAngeles, CA 90095. Telephone: 310-206-6705; FAX: 310-206-4197; E-mail: [email protected]

DOI 10.1002/lt.21882Published online in Wiley InterScience (www.interscience.wiley.com).

LIVER TRANSPLANTATION 15:1525-1534, 2009

© 2009 American Association for the Study of Liver Diseases.

temic chemotherapy,6 and pre-OLT HCC and post-OLTHCC recurrence.7

OLT remains the treatment of choice for HCC. Prior tothe use of the Milan criteria, survival was poor for pa-tients undergoing OLT for HCC and was secondary tocancer recurrence.8 With the Milan criteria (a singletumor � 5 cm or a maximum of 3 total tumors withnone � 3 cm), patients transplanted for HCC haveshown survival rates similar to those of HCC-negativepatients undergoing transplantation.9 The United Net-work for Organ Sharing (UNOS) has used the Milancriteria to select patients for OLT. Expansion of thetumor size limits for OLT was introduced by Yao et al.10

at the University of California, San Francisco (UCSF)when they noted similar posttransplantation survivalrates when less strict criteria were used (a single tu-mor � 6.5 cm or a maximum of 3 total tumors withnone � 4.5 cm and a cumulative tumor size � 8 cm).10

Although a recent study showed prolonged survival af-ter liver transplantation for tumors beyond the Milancriteria but within the UCSF criteria,11 expansion ofOLT criteria remains controversial because 10% to 60%of patients have HCC recurrence after transplanta-tion.12 Factors associated with HCC recurrence afterOLT include a tumor size �5 cm, lymphovascular inva-sion, and male gender.13 Recently, a high HBV viralload and elevated alpha-fetoprotein levels at the time ofhepatic resection for HCC were associated with subse-quent HCC recurrence.14

At our institution, we noted that a group of patientsstill developed HBV reinfection after OLT despite dualprophylaxis. We analyzed a retrospective series of 175patients who underwent OLT for hepatitis B–relatedliver disease, and we evaluated their virological andbiochemical data, tumor burden, comorbidities, antivi-ral therapy, and prolonged immunosuppressive ther-apy. The aim of this study was to determine the riskfactors associated with HBV reinfection after OLT, toexplore the relationship between HBV reinfection andHCC recurrence after OLT, and to evaluate the long-term survival outcome in these patients.

PATIENTS AND METHODS

Immunoprophylaxis

All patients were treated with a combination HBIG(North American Biologicals, Boca Raton, FL) and atleast 1 nucleos(t)ide agent (either lamivudine, adefovir,entecavir, tenofovir, or a combination thereof) for HBVprophylaxis after transplantation. Before 1998, mostpatients were given 10,000 units of HBIG intravenouslyduring the anhepatic phase, and this was then followedby a 7-day course of 10,000 units per day intravenouslyand then 10,000 units intravenously every month. After1998, patients were given 10,000 units intravenouslyduring the anhepatic phase, and they were given 2000units intravenously every day during postoperativedays 2 to 7 and 2000 to 10,000 units intravenously onpostoperative day 20 to keep antibody to hepatitis Bsurface antigen titers greater than 500 IU/mL. There-

after, they were given 1000 units intramuscularly every1 to 4 weeks to maintain antibody to hepatitis B surfaceantigen titer levels at �500 IU/mL at 0 to 6 monthspost-OLT, at �250 IU/mL at 6 to 12 months post-OLT,and at �100 IU/mL 12 months post-OLT. The nu-cleos(t) ide agent(s) were continued indefinitely.

Immunosuppression

A combination of cyclosporine-, tacrolimus-, azathio-prine-, and mycophenolate-based immunosuppressionwas used in the patients. Immunosuppressive dosingwas adjusted according to the therapeutic drug levelsand renal function. Corticosteroid dosing followed aroutine protocol at our institution. High-dose intrave-nous methylprednisolone was rapidly tapered per thestandard protocol from 1 g given intravenously on thetransplant day to 20 mg/day over 1 week after OLT.Oral prednisone was then started at an initial dose of 20mg daily and generally reduced by 2.5 mg monthly untilits discontinuation.

Serologic Monitoring

Patients were followed weekly after hospital dischargeuntil they were stable and then biweekly or monthly asclinically indicated. Standard biochemical tests of liverfunction were monitored at each follow-up visit. Serumhepatitis B surface antigen and HBV DNA were testedmonthly. Hepatitis B reinfection was diagnosed by pos-itive hepatitis B surface antigen in serum and by detec-tion of a positive HBV DNA level. From 1996 until 2004,HBV DNA levels were quantified with the Digene HBVtest (Digene Corp., Gaithersburg, MD) and Hybrid Cap-ture II technology (Digene) via a signal amplificationhybridization microplate assay. Thereafter, all HBVDNA levels were tested with the COBAS TaqMan HBVtest using real-time polymerase chain reaction amplifi-cation (Roche Molecular Systems, Inc., Branchburg,NJ).

Pretransplant and Posttransplant HCCTherapy

Pre-OLT HCC therapy was administered to 62 of the 88patients (70.4%) with HCC. Eleven patients (12.5%) re-ceived pre-OLT systemic chemotherapy, which con-sisted of a combination of antimetabolites (5-fluoroura-cil), platinum-based agents (cisplatin), anthracyclines(doxorubicin), and monoclonal antibodies (bevaci-zumab). Fifty-three patients (60.2%) received a combi-nation of local ablative therapy in the form of arterialchemoembolization, radiofrequency ablation, and/oran ethanol injection prior to transplantation. Five pa-tients (5.7%) underwent surgical resection of their tu-mor(s) prior to transplantation, and 8 patients (9.1%)had multimodality combination therapy. Of the 13 pa-tients with HCC recurrence, 10 patients (76.9%) re-ceived post-OLT systemic chemotherapy. Post-OLTchemotherapy was also administered to 5 patients(5.7%) with HCC but without HCC recurrence after a

1526 SAAB ET AL.

LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases

pathological examination of the explanted liver showedlymphovascular invasion. Four patients (4.5%) receivedpost-OLT chemotherapy for treatment of non-HCC tu-mors. Post-OLT chemotherapy regimens included acombination of antimetabolites (5-fluorouracil, capecit-abine, and gemcitabine), platinum-based agents (cis-platin and oxaliplatin), anthracyclines (doxorubicin),monoclonal antibodies (bevacizumab), multikinase in-hibitors (sorafenib), taxanes (taxotere), alkylatingagents (carboplatin), plant alkaloids (navelbine and eto-poside), and antitumor antibiotics (mitomycin).

Surveillance for HCC Recurrence

After transplantation, patients with known HCC werefollowed regularly in our outpatient hepatology andliver cancer clinics. Surveillance with computed tomog-raphy images was obtained at 3 months, 6 months, andannually thereafter. If there was concern for HCC re-currence, whole-body computed tomography or mag-netic resonance imaging was ordered at the discretionof the physician. All recurrent HCCs were confirmedwith tissue diagnosis.

Compliance

Posttransplant patient compliance was assessed bytrained transplant coordinators who monitored keptappointments and immunosuppressive medicine levelsand evaluated prescription use/refill requests. Data oncompliance were obtained from record reviews for all175 patients.

Statistical Analysis

Continuous variables were presented as means andstandard deviations. Cox regression analysis was usedto determine predictors of time to HBV reinfection. Can-didate predictors associated with a P value less than0.05 by univariate analysis were included in a stepwiseCox regression analysis to build a multivariable modelcontaining independent factors associated with HBVreinfection. The Kaplan-Meier method was used to es-timate survival curves and rates for patient survivalafter OLT, HBV-free reinfection, and HCC recurrence.The log-rank test was used to compare survival curvesbetween patients stratified on various characteristics ofHBV reinfection and HCC recurrence. All analyses wereperformed with SAS version 9.1 (SAS Institute, Inc.,Cary, NC). The data collection process followed institu-tional review board requirements.

RESULTS

Demographics

A retrospective review of our liver transplantation data-base was performed for 175 patients (137 males and 38females; mean age � 52.8 � 10.2 years) who underwentOLT for HBV-related liver disease between January1997 and January 2008 at the Dumont-UCLA Liver

Transplant Center. The mean follow-up time after OLTwas 43.0 � 42.0 months. Table 1 summarizes the pa-tients’ data at the time of transplantation. Data regard-ing HBV DNA and/or hepatitis B e antigen status beforetransplantation were not available in all cases. How-ever, 49 patients (35%) and 52 patients (37.7%) hadpositive pretransplantation HBV DNA and hepatitis B eantigen values, respectively. Hepatitis C virus coinfec-tion was present in 20 patients (11.4%). Prior to OLT,134 patients (76.6%) were treated with antiviral therapyconsisting of lamivudine, adefovir, entecavir, or a com-bination thereof (Table 1). HCC was present in 88 pa-tients (50.3%). Seventy-four patients (84.1%) with HCCwere diagnosed by preoperative computed tomographyor magnetic resonance imaging, whereas the remaining14 patients (15.9%) were diagnosed incidentally by apathological examination of the explant. Eighteen pa-tients (20.5%) had evidence of lymphovascular inva-sion. Thirteen patients (7.4%) in this study receivedsome type of systemic chemotherapy prior to trans-plant, and 19 patients (10.9%) received systemic che-motherapy after transplantation. Also, 75 patients(42.9%) received corticosteroid therapy for greater than6 months after transplantation. One hundred sixty-onepatients (92%) were compliant with their posttrans-plant medication regimen.

HBV Reinfection

Twelve of the 175 patients (6.9%) had post-OLT HBVreinfection. The mean interval time for developing HBVreinfection was 28.7 � 26.4 months. Ten of the 12patients (83.3%) with HBV reinfection had HCC prior totransplantation, and 5 of these 10 patients (50%) hadHCC recurrence after OLT. Table 2 shows the univari-ate analysis of risk factors associated with HBV rein-fection after liver transplantation. Factors associatedwith HBV reinfection were age greater than 60 years atOLT (P � 0.01), pre-OLT HCC (P � 0.004), HCC recur-rence after OLT (P � 0.02), and pre-OLT and post-OLTsystemic chemotherapy (P � 0.03 and P � 0.001, re-spectively). On multivariate analysis, only pre-OLTHCC and post-OLT recurrent HCC were significantlyassociated with hepatitis B reinfection after transplan-tation (P � 0.03 and P � 0.001, respectively).

In the 12 patients who experienced HBV reinfection,8 patients were treated with combination antiviral ther-apy (lamivudine, adefovir, entecavir, and/or tenofovir)after transplantation, and 4 were treated with a singlenucleos(t)ide agent (either lamivudine or adefovir; Table3).

HCC Recurrence

Of the 175 patients transplanted for HBV-related liverdisease, 88 patients (50.3%) had HCC at the time oftransplant (Table 4). Seventy-four of these patients(84.1%) were diagnosed with HCC on preoperative im-aging, whereas 14 (15.9%) were diagnosed incidentallyon pathological examination of the explant. ThirteenHCC patients (14.8%) developed radiographic evidence

RECURRENCE OF HEPATOCELLULAR CARCINOMA 1527


of recurrent HCC after transplantation. The mean HCCrecurrence time was 26.1 � 31.9 months. Among the 13patients, 5 (38%) also had HBV reinfection. By univar-

iate analysis, lymphovascular invasion (P � 0.001) andpost-OLT chemotherapy (P � 0.001) were associatedwith HCC recurrence (Table 5). By multivariate analy-

TABLE 1. Characteristics of the 175 Hepatitis B Surface Antigen–Positive Patients at the Time of Liver

Transplantation

Mean age (year) � SD 52.8 � 10.2

Mean follow-up (months) � SD 43.0 � 42.0Gender

Male 137 (78.3%)Female 38 (21.7%)

RaceAsian 100 (57%)Non-Asian 75 (43%)

HCV coinfection 20 (11.43%)HCC 88 (50.3%)

By preoperative imaging 74 (84.1%)Incidental on explant 14 (15.9%)

Lymphovascular invasion (n � 88) 18 (20.5%)HBV DNA (n � 140)

Positivity 49 (35.0%)Viral titer (IU/mL) � SD 5.17 � 1011 � 3.42 � 1012

HBeAg positivity (n � 138) 52 (37.7%)Pre-OLT antiviral therapy

None 41 (23.4%)ADV 12 (6.9%)ADV � entecavir 2 (1.1%)Entecavir 6 (3.4%)IFN 1 (0.6%)LAM 96 (54.9%)LAM � ADV 16 (9.1%)LAM � entecavir 1 (0.6%)

Medication compliance 161 (92%)

Abbreviations: ADV, adefovir; HBeAg, hepatitis B e antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV,hepatitis C virus; IFN, interferon; LAM, lamivudine; OLT, orthotopic liver transplantation; SD, standard deviation.

TABLE 2. Univariate Analysis of Factors Associated with HBV Reinfection After Liver Transplantation

Factor Hazard Ratio 95% CI P Value

Post-OLT systemic chemotherapy 9.44 2.98–29.90 �0.001Pre-OLT HCC 9.84 2.07–46.71 0.004Age � 60 4.34 1.37–13.70 0.012HCC recurrence 4.93 1.27–19.24 0.022Pre-OLT systemic chemotherapy 4.26 1.15–15.76 0.030Pre-OLT HBV DNA positivity 2.91 0.72–11.70 0.133HBV DNA levels at OLT (IU/mL)

200-99,999 4.14 0.45–38.40 0.211�100,000 3.50 0.77–15.86 0.104

Pre-OLT HBeAg positivity 1.72 0.43–6.93 0.444Lymphovascular invasion 1.56 0.40–6.11 0.527Male gender 1.52 0.33–6.92 0.591HCV coinfection 0.59 0.08–4.59 0.616Duration of steroids � 6 months 0.81 0.26–2.57 0.722Total tumor burden � 10 cm

Preoperative 0.81 0.10–6.37 0.837Explant 1.06 0.27–4.12 0.931

Abbreviations: CI, confidence interval; HBeAg, hepatitis B e antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma;HCV, hepatitis C virus; OLT, orthotopic liver transplantation.

1528 SAAB ET AL.


TA

BL

E3

.A

nti

vira

lT

her

apy

and

Ou

tco

mes

of

Pat

ien

tsw

ith

HB

VR

ein

fect

ion

Aft

erL

iver

Tra

nsp

lan

tati

on

Pati

ent

Yea

rof

Tra

nsp

lan

t

HB

VD

NA

Lev

el

Bef

ore

An

tivi

ral

Th

erapy

(IU

/m

L)

Pre

-OLT

An

tivi

ral

Th

erapy

Du

rati

on of

An

tivi

ral

Th

erapy

(Mon

ths)

HB

VD

NA

Lev

elat

OLT

(IU

/

mL)

Pos

t-O

LT

HB

V

Pro

ph

ylaxi

s

Tim

eto

HB

V

Rec

urr

ence

(Mon

ths)

HB

V

Tre

atm

ent

Fol

low

-Up

Sin

ceH

BV

Rec

urr

ence

(Mon

ths)

HC

C

Rec

urr

ence

Tim

eto

HC

C

Rec

urr

ence

(Mon

ths)

Com

plian

ceO

utc

ome

12002

Neg

ati

veLA

M12

Neg

ati

veH

BIG

�LA

M11

LA

M�

AD

V58

Yes

12

Yes

Exp

ired

*

21998

Neg

ati

veLA

M13

Neg

ati

veH

BIG

�LA

M88

LA

M�

AD

V28

NA

NA

Yes

Alive

32002

NA

LA

M11

4.0

5�

10

5H

BIG

�LA

M13

LA

M�

ten

ofov

ir53

No

NA

Yes

Alive

41997

NA

LA

M13

NA

HB

IG�

LA

M67

LA

M�

AD

V62

No

NA

Yes

Alive

52003

1.5

6�

10

6LA

M13

NA

HB

IG�

LA

M46

LA

M�

AD

V2

Yes

22

No

Exp

ired

*

62003

NA

LA

M6

NA

HB

IG�

LA

M41

LA

M�

AD

V2

Yes

23

Yes

Exp

ired

*

72005

Neg

ati

veLA

M17

Neg

ati

veH

BIG

�LA

M13

LA

M�

AD

V18

Yes

10

Yes

Exp

ired

*

82004

NA

LA

M9

2.1

8�

10

6H

BIG

�LA

M11

LA

M�

AD

V40

No

NA

Yes

Alive

91999

Neg

ati

veLA

M21

Neg

ati

veH

BIG

�LA

M3

LA

M14

NA

NA

Yes

Los

tto

follow

-up

10

1998

NA

LA

M20

NA

HB

IG�

LA

M20

LA

M1

No

NA

Yes

Exp

ired

†

11

1999

Neg

ati

veLA

M27

Neg

ati

veH

BIG

�LA

M20

LA

M12

No

NA

Yes

Exp

ired

‡

12

2007

NA

AD

V48

NA

HB

IG�

AD

V7

AD

V1

Yes

4Yes

Alive

Abbre

viat

ion

s:A

DV

,adef

ovir

;H

BIG

,h

epati

tis

Bim

mu

nog

lobu

lin

;H

BV

,h

epati

tis

Bvi

rus;

HC

C,

hep

ato

cellu

lar

carc

inom

a;

LA

M,

lam

ivu

din

e;N

A,

not

applica

ble

;O

LT,

orth

otop

iclive

rtr

an

spla

nta

tion

.*D

ue

toH

CC

recu

rren

ce.

†Du

eto

HB

Vre

curr

ence

.‡D

ue

tom

etast

ati

cpan

crea

tic

carc

inom

a.



sis, there was a trend toward both vascular invasion(P � 0.06) and post-OLT chemotherapy (P � 0.02) beingindependently associated with HCC recurrence. Onlypost-OLT chemotherapy (P � 0.001) was retained as anindependent predictor.

Five patients in this cohort had both HBV and HCC

recurrence. In 4 of the 5 patients (80%), HCC recur-rence preceded HBV recurrence. One patient presentedwith HBV recurrence first but was diagnosed with HCCrecurrence 1 month later. Correlation analysis con-firmed a strong correlation between HBV and HCC re-currence times (P � 0.01, r � 0.96; Fig. 1).

TABLE 4. Criteria Groups and Tumor Characteristics of Patients with Hepatocellular Carcinoma Based on

Preoperative Imaging and Pathological Examination of the Explant

Patients (n � 88)

Tumor criteria (pre-OLT/explant)Milan 53/38UCSF 9/12Beyond UCSF 12/30No gross mass 14/8

Number of tumors (pre-OLT/explant)1 50/342-3 17/28�4 0/7Multifocal* 6/11No gross mass 15/8

Total tumor burden (pre-OLT/explant)�5 cm 59/415-10 cm 20/28�10 cm or multifocal* 9/19

Abbreviations: OLT, orthotopic liver transplantation; UCSF, University of California, San Francisco.*Reported by a radiologist on preoperative imaging or by a pathologist on examination of the explanted liver.

TABLE 5. Univariate Analysis of Factors Associated with Hepatocellular Carcinoma Recurrence After Liver

Transplantation

Factor Hazard Ratio 95% CI P Value

Lymphovascular invasion 8.72 2.68–28.34 �0.001Post-OLT systemic chemotherapy 11.20 3.07–40.79 �0.001AFP � 100 ng/mL 3.11 0.91–10.63 0.071Tumor burden within UCSF criteria*

Preoperative 2.65 0.55–12.84 0.226Explant† NA NA NA

Tumor burden beyond UCSF criteriaPreoperative 2.68 0.78–9.18 0.117Explant 3.89 1.05–14.39 0.042

Total tumor burden � 10 cmPreoperative 2.78 0.76–10.14 0.121Explant 2.42 0.81–7.23 0.112

HBV DNA levels at OLT (IU/mL)200-99,999 2.47 0.47–12.87 0.282�100,000 0.74 0.09–6.32 0.781

Male gender 3.03 0.40–23.32 0.286Pre-OLT antiviral prophylaxis 2.11 0.47–9.54 0.334Pre-OLT HBeAg positivity 0.48 0.06–3.93 0.495Age � 60 0.71 0.23–2.18 0.551Pre-OLT HBV DNA positivity 1.38 0.33–5.81 0.658Pre-OLT systemic chemotherapy 0.91 0.20–4.13 0.904

Abbreviations: AFP, alpha-fetoprotein; CI, confidence interval; HBeAg, hepatitis B e antigen; HBV, hepatitis B virus; NA, notapplicable; OLT, orthotopic liver transplantation; UCSF, University of California, San Francisco.*Milan criteria were used for the reference group for both preoperative and explant analyses.†An analysis could not be performed because only 1 patient had hepatocellular carcinoma recurrence.

1530 SAAB ET AL.


Outcome and Survival

Overall survival after OLT was significantly reduced forpatients with HBV reinfection, with probabilities at 1, 3,and 5 years of 92.4%, 86.8%, and 84.3%, respectively,for patients without HBV reinfection versus 100%,70.0%, and 46.7%, respectively, for patients with viralreinfection (P � 0.007). Patients with HCC recurrenceafter OLT also showed poor 3- and 5-year survival ratesin comparison with patients without HCC recurrence(60.6% and 15.2% versus 70.5% and 70.5%, respec-tively; P � 0.03). Of the 12 patients with HBV reinfec-tion, 6 died, 5 are still alive, and 1 was lost to follow-up(Table 4). The cause of death for 4 of the 6 patients wasrecurrence of HCC. Two other patients died of compli-cations of hepatitis B reinfection, and 1 patient died ofmetastatic pancreatic carcinoma. Survival analysisshowed a reduced HBV reinfection rate in patients withno HCC prior to OLT (P � 0.001; Fig. 2) and no HCCrecurrence in comparison with HCC recurrence afterOLT (P � 0.001; Fig. 3). Cumulative survival after OLTwas associated with no HBV reinfection post-OLT (P �0.007; Fig. 4) and no HCC recurrence post-OLT (P �0.03, Fig. 5). Also, HBV reinfection–free survival wasassociated with the absence of HCC recurrence afterOLT (P � 0.01; Fig. 6).

DISCUSSION

HBV reinfection after liver transplantation results indecreased patient and graft survival. To identify riskfactors for HBV reinfection after OLT, the demographic,biochemical, virological, and tumor factors in our co-hort of patients were analyzed. The rate of HBV reinfec-tion in our study was similar to rates described in otherreports,15,16 and we showed that both pre-OLT HCCand HCC recurrence were independent risk factors forHBV reinfection after transplantation. A recent studyshowed that covalently closed circular DNA, the repli-cative form of DNA, was the main form of HBV DNAisolated from tissues with HCC.17 The replicative ca-

pacity of HBV DNA in HCC cells has been furtherstrengthened by the detection of HBV transcriptionalactivity in tumor cells by polymerase chain reactionanalysis.18 Recently, Faria et al.7 reported an associa-tion between HCC recurrence and HBV reinfection.These authors demonstrated the presence of covalentlyclosed circular DNA in both HCC cells and in nontumorcells in explanted livers from patients with HCC recur-rence, suggesting that HBV replication also may occurin tumor cells. Furthermore, the identification of HBVDNA in extrahepatic sites in the absence of HBV DNA inthe liver and serum indicates that viral replication mayoccur in occult extrahepatic metastatic sites and lead toHBV reinfection.

Figure 1. Correlation between HBV infection and HCC recur-rence times in liver transplant recipients. Abbreviations: HBV,hepatitis B virus; HCC, hepatocellular carcinoma. Figure 2. Relationship between the presence of HCC pre–liver

transplantation and HBV reinfection post–liver transplanta-tion. Abbreviations: HBV, hepatitis B virus; HCC, hepatocel-lular carcinoma; N, number of patients; OLT, orthotopic livertransplantation.

Figure 3. Relationship between the presence of recurrentHCC post–liver transplantation and HBV reinfection post–liver transplantation. Abbreviations: HBV, hepatitis B virus;HCC, hepatocellular carcinoma; N, number of patients; OLT,orthotopic liver transplantation.



Immunosuppression and systemic chemotherapyalso may lead to HBV reactivation. The reappearance ofHBV in patients receiving cytotoxic or immunosuppres-sive therapy has been well documented,19,20 and thesepatients have an increased risk of hepatic damage.21

The incidence of HBV reactivation in cancer patientsundergoing chemotherapy has been reported to rangefrom 19% to 48%.22 Although significant on univariateanalysis, pre-OLT and post-OLT systemic chemother-apy did not reach statistical significance on Cox regres-sion hazard analysis as a risk factor for HBV reinfectionin this study. The majority of our patients who receivedchemotherapy prior to liver transplantation had multi-focal tumors on preoperative imaging, and systemicchemotherapy was given in hopes of minimizing thetumor burden. Our HCC patients received chemother-

apy post-transplantation for 2 reasons: (1) as adjuvanttherapy after a pathological examination of the explantshowed lymphovascular invasion and (2) for treatmentof HCC recurrence. Although the higher rates of HBVreinfection in these patients may be explained by abreach in the immune system by cytotoxic chemother-apy, the higher risk of HBV reinfection may partly berelated to HCC recurrence itself. That is, actively repli-cating HCC cells may facilitate viral reproduction ofHBV DNA within the tumor cells. A large tumor burdenprior to transplant as well as lymphovascular invasionon the explant may indicate the presence of extrahe-patic, micrometastatic sites, which may serve as asource for HBV replication. It is therefore prudent toclosely monitor HBV DNA levels and liver tests in pa-tients who receive chemotherapy after transplantationin order to allow early detection and treatment of hep-atitis B reinfection. Lamivudine prophylaxis in patientsundergoing chemotherapy has been shown to be a cost-effective method for decreasing rates of HBV reactiva-tion and improving reactivation-related outcomes.23,24

Antiviral therapy with nucleos(t)ide agents is indicatedas a prophylactic measure before OLT in all hepatitis Bpatients who undergo chemotherapy, regardless ofHBV DNA levels.

Our data demonstrated that the presence of lympho-vascular invasion and post-OLT chemotherapy wereboth risk factors for tumor recurrence. In this study, weexamined the outcome of patients with respect to viralreinfection as well as tumor recurrence. Previous stud-ies had shown similar overall survival rates in post-OLTpatients with HBV reinfection with or without HCC.25

Kiyici et al.26 recently found no significant effect of HBVreinfection or HCC recurrence on overall survival inOLT patients. However, our analysis herein showed de-creased cumulative survival for patients with HBV re-infection and with HCC recurrence. Another reportfrom our institution demonstrated poor survival for pa-

Figure 4. Cumulative survival in liver transplant recipientswith and without HBV reinfection. Abbreviations: HBV, hepa-titis B virus; N, number of patients; OLT, orthotopic livertransplantation.

Figure 5. Cumulative posttransplant survival for patientswith and without HCC recurrence. Abbreviations: HCC, hepa-tocellular carcinoma; N, number of patients; OLT, orthotopicliver transplantation.

Figure 6. HBV reinfection–free survival in liver transplantrecipients with and without HCC recurrence. Abbreviations:HBV, hepatitis B virus; HCC, hepatocellular carcinoma; N,number of patients.

1532 SAAB ET AL.


tients transplanted with tumors beyond the UCSF cri-teria.11 Although their results did not show pretrans-plant hepatitis B to be a statistically significant riskfactor for mortality after transplantation for HCC, theeffect of hepatitis B reinfection was not assessed. De-creased long-term survival in patients who suffer HCCrecurrence has been well demonstrated,12 but the effectof HCC recurrence on viral reinfection and subsequentpatient survival has not been evaluated. HBV reinfec-tion has been shown to significantly affect overall andHCC recurrence–free survival, and a reduction in viralreinfection with combination prophylaxis had a signif-icant effect on decreasing mortality in these patients.27

In our study, all patients received combined prophy-laxis after transplantation, and pretransplant antiviraltherapy was not associated with a decreased risk ofHCC recurrence or HBV reinfection. Similarly to ourfindings, Wong et al.28 showed higher rates of HBVreinfection, graft loss, and mortality in patients withHBV infection complicated by HCC versus patients withHBV-related liver disease without HCC. In their study,HBV reinfection was the primary cause of death,whereas the principal cause of death in our patientswith HBV reinfection was HCC recurrence. Althoughdifferences in pretransplant characteristics, antiviraltreatment, and chemotherapy regimens may accountfor these differences, the fact remains that both HBVreinfection and HCC recurrence led to both decreasedgraft and patient survival.

The mean time to HBV reinfection in our series was28.7 � 26.4 months. However, the time to reinfectionwas longer in some of the transplant recipients in whomHCC recurrence preceded HBV reinfection (patients 5and 6). The temporal association between HCC recur-rence and later HBV reinfection has been reported byothers and is believed to be related to viral replication inHCC cells.7 Other potential causes of higher HBV re-currence times include a lack of compliance and in-creased viral resistance over time to oral nucleos(t)ideagents. However, we do not believe noncompliance wasa major issue in this cohort because of regular assess-ments of hepatitis B antibody titers to gauge HBIGinfusion and regular assessments of medication lists bydedicated transplant coordinators.

Although a limitation of this study includes its retro-spective design, it is one of the largest of its kind, andour results are consistent with those of previous stud-ies demonstrating a relationship between HCC recur-rence and HBV reinfection. Although we did not find astatistically significant correlation with HBV DNA levelsand HBV reinfection, high pretransplant HBV DNA lev-els have been shown to be associated with an increasedrisk of HBV reinfection and graft loss.5,29 An inherentlimitation of this retrospective study is that not all of theHBV DNA levels were known prior to transplantationand therefore the effect of antiviral therapy on viral loadis not available for all patients. Nevertheless, all pa-tients had undetectable hepatitis B surface antigen andHBV DNA levels after liver transplantation. Severalother reports have assessed factors associated withHBV reinfection after transplantation. Our study not

only comprehensively evaluated the serologic, viral,and biochemical factors related to HBV reinfection butalso examined the effects of immunosuppression andtumor characteristics on viral reinfection, HCC recur-rence, and clinical outcomes in these patients.

In conclusion, we found that pre-OLT HCC and re-current HCC after transplantation were significantlyassociated with HBV reinfection in liver transplant pa-tients. Both HBV reinfection and HCC recurrence de-crease overall patient survival. Thus, patients withHBV-related HCC who undergo OLT require close viro-logical monitoring and may benefit from antiviral pro-phylaxis prior to and after liver transplantation.

ACKNOWLEDGMENTStatistical analyses were performed by W. Scott Comu-lada and Sung-Jae Lee.

REFERENCES

1. World Health Organization. The World Health Report. Ge-neva, Switzerland: World Health Organization; 1997.

2. Starzl TE, Demetris AJ, Van Thiel D. Liver transplantation(2). N Engl J Med 1989;321:1092-1099.

3. Dan YY, Wai CT, Yeoh KG, Lim SG. Prophylactic strategiesfor hepatitis B patients undergoing liver transplant: a cost-effectiveness analysis. Liver Transpl 2006;12:736-746.

4. Angelico M, Di Paolo D, Trinito MO, Petrolati A, Araco A,Zazza S, et al. Increasing applicability of liver transplan-tation for patients with hepatitis B-related liver disease.Hepatology 2002;35:176-181.

5. Ben-Ari Z, Daudi N, Klein A, Sulkes J, Papo O, Mor E, et al.Genotypic and phenotypic resistance: longitudinal and se-quential analysis of hepatitis B virus polymerase muta-tions in patients with lamivudine resistance after livertransplantation. Am J Gastroenterol 2003;98:151-159.

6. Yi NJ, Suh KS, Cho JY, Kwon CH, Lee KW, Joh JW, et al.Recurrence of hepatitis B is associated with cumulativecorticosteroid dose and chemotherapy against hepatocel-lular carcinoma recurrence after liver transplantation.Liver Transpl 2007;13:451-458.

7. Faria LC, Gigou M, Roque-Afonso AM, Sebagh M, Roche B,Fallot G, et al. Hepatocellular carcinoma is associated withan increased risk of hepatitis B virus recurrence after livertransplantation. Gastroenterology 2008;134:1890-1899.

8. Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A,Bozzetti F, et al. Liver transplantation for the treatment ofsmall hepatocellular carcinomas in patients with cirrho-sis. N Engl J Med 1996;334:693-699.

9. Figueras J, Jaurrieta E, Valls C, Benasco C, Rafecas A,Xiol X, et al. Survival after liver transplantation in cirrhoticpatients with and without hepatocellular carcinoma: acomparative study. Hepatology 1997;25:1485-1489.

10. Yao FY, Ferrell L, Bass NM, Watson JJ, Bacchetti P, Ve-nook A, et al. Liver transplantation for hepatocellular car-cinoma: expansion of the tumor size limits does not ad-versely impact survival. Hepatology 2001;33:1394-1403.

11. Duffy JP, Vardanian A, Benjamin E, Watson M, FarmerDG, Ghobrial RM, et al. Liver transplantation criteria forhepatocellular carcinoma should be expanded: a 22-yearexperience with 467 patients at UCLA. Ann Surg 2007;246:502-511.

12. Roayaie S, Schwartz JD, Sung MW, Emre SH, Miller CM,Gondolesi GE, et al. Recurrence of hepatocellular carci-noma after liver transplant: patterns and prognosis. LiverTranspl 2004;10:534-540.



13. Schlitt HJ, Neipp M, Weimann A, Oldhafer KJ, Schmoll E,Boeker K, et al. Recurrence patterns of hepatocellular andfibrolamellar carcinoma after liver transplantation. J ClinOncol 1999;17:324-321.

14. Hung IF, Poon RT, Lai CL, Fung J, Fan ST, Yuen MF.Recurrence of hepatitis B-related hepatocellular carci-noma is associated with high viral load at the time ofresection. Am J Gastroenterol 2008;103:1663-1673.

15. Marzano A, Salizzoni M, Debernardi-Venon W, Smedile A,Franchello A, Ciancio A, et al. Prevention of hepatitis Bvirus recurrence after liver transplantation in cirrhoticpatients treated with lamivudine and passive immunopro-phylaxis. J Hepatol 2001;34:903-910.

16. Markowitz JS, Martin P, Conrad AJ, Markmann JF, Seu P,Yersiz H, et al. Prophylaxis against hepatitis B recurrencefollowing liver transplantation using combination lamivu-dine and hepatitis B immune globulin. Hepatology 1998;28:585-589.

17. Wong DK, Yuen MF, Poon RT, Yuen JC, Fung J, Lai CL.Quantification of hepatitis B virus covalently closed circu-lar DNA in patients with hepatocellular carcinoma.J Hepatol 2006;45:553-559.

18. Pollicino T, Squadrito G, Cerenzia G, Cacciola I, Raffa G,Craxi A, et al. Hepatitis B virus maintains its pro-onco-genic properties in the case of occult HBV infection. Gas-troenterology 2004;126:102-110.

19. Wands JR, Chura CM, Roll FJ, Maddrey WC. Serialstudies of hepatitis associated antigen and antibody inpatients receiving antitumor chemotherapy for myelo-proliferative and lymphoproliferative disorders. Gastro-enterology 1975;68:105-112.

20. Scullard GH, Smith CI, Merigan TC, Robinson WS, Greg-ory PB. Effects of immunosuppressive therapy on viralmarkers in chronic active hepatitis B. Gastroenterology1981;81:987-991.

21. Dai MS, Lu JJ, Chen YC, Perng CL, Chao TY. Reactivation

of precore mutant hepatitis B virus in chemotherapy-treated patients. Cancer 2001;92:2927-2932.

22. Yeo W, Johnson PJ. Diagnosis, prevention and manage-ment of hepatitis B virus reactivation during anticancertherapy. Hepatology 2006;43:209-220.

23. Saab S, Dong MH, Joseph TA, Tong MJ. Hepatitis B pro-phylaxis in patients undergoing chemotherapy for lym-phoma: a decision analysis model. Hepatology 2007;46:1049-1056.

24. Martyak LA, Taqavi E, Saab S. Lamivudine prophylaxis iseffective in reducing hepatitis B reactivation and reactiva-tion-related mortality in chemotherapy patients: a meta-analysis. Liver Int 2008;28:28-38.

25. Wong SN, Reddy KR, Keeffe EB, Han SH, Gaglio PJ,Perrillo RP, et al. Comparison of clinical outcomes inchronic hepatitis B liver transplant candidates with andwithout hepatocellular carcinoma. Liver Transpl 2007;13:334-342.

26. Kiyici M, Yilmaz M, Akyildiz M, Arikan C, Aydin U, SigirliD, et al. Association between hepatitis B and hepatocellu-lar carcinoma recurrence in patients undergoing livertransplantation. Transplant Proc 2008;40:1511-1517.

27. Zimmerman MA, Ghobrial RM, Tong MJ, Hiatt JR, Cam-eron AM, Busuttil RW, et al. Antiviral prophylaxis andrecurrence of hepatocellular carcinoma following livertransplantation in patients with hepatitis B. TransplantProc 2007;39:3276-3280.

28. Wong PY, McPeake JR, Portmann B, Tan KC, NaoumovNV, Williams R. Clinical course and survival after livertransplantation for hepatitis B virus infection complicatedby hepatocellular carcinoma. Am J Gastroenterol 1995;90:29-34.

29. Marzano A, Lampertico P, Mazzaferro V, Carenzi S, ViganoM, Romito R, et al. Prophylaxis of hepatitis B virus recur-rence after liver transplantation in carriers of lamivudine-resistant mutants. Liver Transpl 2005;11:532-538.

1534 SAAB ET AL.


Documents

Recurrence of hepatocellular carcinoma and hepatitis B reinfection in hepatitis B surface antigen–positive patients after liver transplantation