The combination of lamivudine and hepatitis B immunoglobulin (HBIG) reduces the risk of hepatitis B virus (HBV) recurrence after liver transplantation (LT). However, the efficacy of this strategy and the need for combined therapy with adefovir dipivoxil (ADV) in patients who select lamivudine-resistant strains (YMDD) before surgery is still unknown. Twenty-two patients treated with lamivudine (LAM) who underwent LT after YMDD-mutant selection were studied. In 13 patients, YMDD mutants were associated with an HBV DNA breakthrough greater than 5 log10 (group A: phenotypic resistance), and 11 were treated with ADV to decrease viral load before LT. In the remaining 9 patients who did not experience the viral breakthrough, YMDD mutants were detected only retrospectively in sera stored at the time of LT (group B: genotypic resistance). During 35 months of post-LT follow-up, none of the 11 patients of group A treated with ADV before and after surgery (in addition to HBIG and LAM) had HBV recurrence, and neither did any of the 7 subjects of group B treated with LAM before and after transplantation (in addition to HBIG). HBV recurred in 2 patients of group A (untreated with ADV before surgery and transplanted with an HBV DNA exceeding 5 log10) and in 2 subjects of group B (who spontaneously stopped HBIG after surgery). In carriers of YMDD mutants, the risk of post-LT HBV recurrence is low, provided that preemptive and prophylactic ADV (in addition to LAM and HBIG) treatment is used in highly viremic patients and prophylactic LAM (or ADV) and HBIG therapy is continued in low viremic patients. (Liver Transpl 2005;11:532–538.)
In hepatitis B virus (HBV) DNA-positive patients, the rate of HBV recurrence after liver transplantation (LT) remains significant (50%–90%) despite monoprophylaxis with hepatitis B immunoglobulins (HBIG) or with lamivudine (LAM). In both cases, post-LT HBV recurrence may be associated with the appearance of viral mutants selected by either HBIG or LAM, whose occurrence appears correlated with the serum levels of HBV DNA at the time of LT.1–4
Resistance to LAM is characterized by the substitution of methionine with valine or isoleucine at residue 204 (formerly 552) within the tyrosine-methionine–aspartate-aspartate (YMDD) motif of the viral DNA polymerase. This mutation results in a more than 10,000-fold decrease in sensitivity to LAM treatment in vitro.5
The clinical lack of virologic response to LAM associated with the YMDD mutants is currently defined by a viral breakthrough with HBV DNA greater than 5 log10 copies/mL.6, 7 In the transplant setting, the appearance of the YMDD mutant during posttransplantation LAM prophylaxis is frequently associated with recurrence of the hepatitis B surface antigen (HBsAg) and liver disease in the graft.3
Although combined prophylaxis with LAM and HBIG is effective in preventing HBV recurrence after LT,8–14 whether it is equally effective in transplant candidates who have developed YMDD mutants before transplantation is not clear. Data for transplant recipients carrying the YMDD mutant at the time of surgery are few and contradictory. In two studies,13, 14 the presence of a LAM-resistant HBV DNA in serum at the time of transplantation was reported to predict post-LT viral recurrence despite HBIG and LAM prophylaxis. Conversely, combined LAM and HBIG prophylaxis prevented the clinical recurrence in a single highly viremic YMDD-positive patient for more than 32 months of follow-up.15 In another report, HBIG and LAM prophylaxis was effective in a YMDD carrier who experienced subacute liver failure, although data on HBV DNA load at the time of transplantation were unknown.16
In the process of the development of LAM resistance, genotypic resistance (YMDD detection without a significant HBV DNA breakthrough) frequently precedes the phenotypic resistance.17, 18 YMDD mutants can now be successfully treated with alternative drugs, such as adefovir dipivoxil (ADV), which appears to be effective against both wild-type and LAM-resistant HBV strains.19, 20 ADV has been shown to be able to control HBV recurrence after LT in a few case reports.21–23 However, the need for ADV therapy (before or after transplantation) as a prevention of hepatitis B recurrence is still debated, and data are lacking, especially when the differences of genotypic or phenotypic resistance to LAM are taken into account. In two recent studies,24, 25 pretransplantation ADV therapy improved the virologic, biochemical, and clinical parameters while patients were waiting for a new liver, even though data on post-LT HBV recurrence were not analyzed.
In this study, we analyzed a retrospective consecutive series of 22 HBsAg-positive patients with cirrhosis who developed YMDD mutation while on LAM therapy before LT, with particular reference to the additional use of ADV in combination with the current strategy for prevention of HBV recurrence after transplantation.
HBV, hepatitis B virus; LT, liver transplantation; LAM, lamivudine; ADV, adefovir; CsA, cyclosporine; HBIG, hepatitis B immunoglobulin; PCR, polymerase chain reaction; HBsAg, hepatitis B surface antigen.
Patients and Methods
Between 1997 and 2002, 99 patients with HBV-related cirrhosis underwent preemptive LAM treatment while waiting for LT in 2 Northern Italian liver transplantation centers: Molinette Hospital in Turin and National Cancer Institute in Milan. Patients were enlisted because of end-stage liver disease with or without hepatocellular carcinoma.
At listing, before any antiviral therapy, all patients had a viral load of more than 5 log10 copies/mL by a nonamplified assay7 and were treated with LAM (GlaxoWellcome, Greenford, UK) at an oral dose of 100 mg daily, to decrease the HBV DNA level before LT and to ameliorate liver function.
During pretransplantation LAM therapy, 13 of 99 patients had a rebound of serum HBV DNA to levels greater than 5 log10 copies/mL and alanine-aminotransferase (ALT) levels greater than 1.5 times the upper limit of normal (YMDD-positive group A, phenotypic resistance). Eleven patients (cases 3–13) were suspended from the active waiting list at the moment of the virologic breakthrough until the ADV 10 mg/daily was added to LAM as rescue compassionate-use protocol (Gilead Sciences, Foster City, CA, GS-98-435 Study). As soon as HBV DNA levels fell below 5 log10 copies/mL, patients were relisted to LT and treated with triple prophylaxis (ADV, LAM, HBIG) in the posttransplantation follow-up (Table 1). The remaining 2 patients (cases 1 and 2) continued only LAM before surgery, because of the unavailability of ADV while on the waiting list, and underwent LT. After liver replacement, both patients received only a double prophylaxis (LAM and HBIG).
Table 1. Demographic and Clinical Features in YMDD Carriers at the Time of Mutant Detection
Of the remaining 86 patients who did not experience the viral breakthrough and who underwent LT with HBV DNA levels less than 5 log10 copies/mL, a frozen serum sample collected at the time of surgery was retrospectively tested for HBV DNA by polymerase chain reaction (PCR) and INNO-LiPA assays. Seventy-seven patients had undetectable HBV DNA or had a viral load between 2.3 and 4 log10 copies/mL but wild-type sequence at the YMDD motif (YMDD-negative group), whereas 9 had detectable HBV DNA and LAM-resistant strains. These 9 patients (YMDD-positive group B, genotypic resistance) received LAM and HBIG after transplantation. During follow-up, 2 noncompliant patients (case numbers 14 and 15) stopped HBIG on their own initiative and decided to continue just LAM orally.
Overall, 22 patients with LAM resistance (YMDD-positive) were studied: 13 with phenotypic (group A) and 9 with genotypic (group B) resistance. In each patient of the series, detection of mutation in position 204 of the polymerase gene was confirmed, with various associations with mutations in positions 180 and 207 (Table 1).
In this group, 21 cadaveric and 1 living related liver transplantations were performed. There were 3 female patients and 19 male patients; the mean age was 49 years (range, 28–58 years). Seven patients were HBeAg positive (46% of group A subjects) and 15 had antibodies to the HBeAg (anti-HBe) (89% of group B subjects). All were negative for antibodies to HDV and only 1 (case 17) was positive for anti-hepatitis C antibodies; in 13 patients, hepatocellular carcinoma was confirmed by histology in the explanted cirrhotic liver. Before starting LAM, the mean viral load was comparable in both groups (7.8 log10 in group A and 7.4 log10 in group B, respectively) but phenotypic resistance developed more frequently in patients treated for a longer time with LAM (mean time of treatment, 25.5 months in group A and 15 months in group B, respectively) (Tables 1 and 2).
Table 2. HBV Recurrence in Lamivudine-Resistant Carriers (YMDD) in Relation to the Antiviral Therapy Before Surgery and to the Type of Prophylaxis After Liver Transplantation
In both LT centers, a mean total dose of 50,000 IU of intravenous HBIG (Hepatect, Biotest, Dreieich, Germany; VenBig, Kedrion, Barga, Italy) was administered during the first month, in an aim to produce anti-HBs titers in the serum greater than 500 IU/L. After hospital discharge and during the entire course of follow-up, each patient received a mean of 6,000 IU of intravenous HBIG every month, with the aim of keeping anti-HBs titers above 100 IU/L.
After transplantation, a double-drug induction therapy with cyclosporine (CsA) and steroids was used. Steroids were tapered from day 0 to day 5 at 25 mg/day and discontinued 3 to 6 months after transplantation. Tacrolimus was introduced as a rescue therapy for steroid-resistant rejection or in case of CsA toxicity.
Assays and Definition
Liver function and serologic tests were measured with conventional assays (AUSRIA II and HBe Kit; Abbott Laboratories, North Chicago, IL; AB AUK 3, Sorin Biomedica, Saluggia, VC, Italy; Ortho Diagnostic System, Milano, Italy). Before starting LAM or ADV, all patients tested positive for HBV DNA by a nonamplified assay (Digene Hybrid-Capture II; Digene Corp., Gaithersburg, MD LLQ 0.5 pg/mL or 141,500 copies/mL) or branched DNA signal amplification (Bayer Corp., Tarrytown, NY, LLQ 700,000 Eq/mL).
During therapy and at the time of transplantation HBV DNA was detected using a quantitative PCR commercial assay, whose lower limit of detection is 200 copies/mL (2.3 log10 copies/mL) (Amplicor HBV, COBAS Monitor, Roche, Basel, Switzerland). In patients HBV DNA–positive by PCR at the time of transplantation, LAM-resistant mutants were tested with INNO-LiPA HBV DR (Innogenetics Inc., Ghent, Belgium). Recurrence of HBV was defined as the reappearance of HBsAg in serum at any time interval after liver replacement.
After LT, the 77 patients in the YMDD-negative group were followed for a median of 39 months. Three patients (4%) died without evidence of HBV recurrence at 4, 12, and 30 months after surgery. Overall, HBV relapse occurred in 4 patients (5%) while on HBIG and LAM prophylaxis without evidence of YMDD selection at the time of hepatitis B recurrence.
All 22 patients in the YMDD-positive group survived transplant surgery and were discharged from the hospital after a median of 17 days. One patient in group B (case number 17) died 20 months after LT from a massive bleeding after a liver biopsy without evidence of HBV recurrence. His follow-up was censored at the time of death. The current study focuses on the outcome of HBV infection, with special reference to pretransplantation and posttransplantation treatment.
Group A (Phenotypic Resistance)
In the 11 patients with phenotypic resistance to LAM who were treated with ADV while awaiting transplantation, the antiviral therapy was well tolerated (Table 2). In particular, renal side effects and consequent dose reduction were not registered although most of the transplant candidates suffered liver insufficiency. During 114 days (range, 31–200) of ADV treatment the mean Child-Turcotte-Pugh score did not change significantly, and HBV DNA decreased on average from 6.9 (range, 5.7–9.0) to 4.2 log10 copies/mL (range, <2.3–6.0), with a median decrease of 2.6 log10 copies/mL (range, 0.5–5.1). At transplantation, all patients but 1 (case 9) achieved an HBV DNA below 5 log10 copies/mL. As reported previously, 2 cases with phenotypic resistance to LAM arrived at transplantation with HBV DNA levels greater than 5 log10 copies/mL, because ADV was not available during the pretransplantation period.
Group B (Genotypic Resistance)
In the 9 patients who were retrospectively identified as carriers of genotypic YMDD mutants, HBV DNA levels ranged from 2.4 to 4.0 log10 copies/mL (median, 2.9) at transplantation.
The median posttransplantation follow-up for the entire series was 35 months (range, 16–87) (Table 2 and Fig. 1). Overall, HBV recurrence was detected in 4 YMDD carriers: 2 subjects in the group A and 2 in the group B; differences are discussed in the following sections.
Group A (Phenotypic Resistance)
Patients were followed for a median of 29 months (range, 16–64) after LT. The 2 patients who suffered HBV recurrence during this period underwent transplantation in the presence of high serum levels of HBV DNA (9 log10 copies/mL) as a result of YMDD mutants and did not receive ADV before or after LT but received LAM and HBIG combination therapy after surgery. HBV recurred within the first 2 months after LT in both patients, featuring a fibrosing cholestatic hepatitis in 1.
Conversely, none of the 11 patients treated with ADV before and after LT showed any sign of HBV recurrence during follow-up. Nine patients received triple prophylactic therapy (ADV, LAM, HBIG) throughout the follow-up, whereas in 2 patients (case numbers 4 and 5), ADV was stopped 3 and 32 days after LT for primary liver malfunction and renal failure and psychiatric side effects, respectively. It is noteworthy that combined HBIG and LAM prophylaxis was effective in preventing HBV recurrence in these 2 patients.
Group B (Genotypic Resistance)
Patients were followed for a median of 43 months after LT (range, 18–87). HBV recurred in the 2 patients who decided to stop HBIG on their own initiative and remained on LAM monoprophylaxis 4 and 3 months after surgery. In these patients, HBV DNA at the time of surgery was detectable only by PCR with an HBV DNA level of 2.5 and 4.0 log10 copies/mL. The remaining 7 patients receiving a posttransplantation prophylaxis with LAM and HBIG did not show any clinical/biochemical sign of HBV recurrence throughout the entire follow-up.
Clinical Outcome After Liver Transplantation
Patients who did not experience the HBV recurrence after LT remained persistently HBsAg negative without any biochemical or clinical evidence of HBV, whereas all 4 subjects who experienced recurrence had a clinical and histologic profile of chronic HBV. After the HBV recurrence, 3 patients stopped HBIG, continued LAM, and added ADV.
The protective effect of LAM prophylaxis on post-LT HBV recurrence is not complete in viremic patients, and its loss of efficacy is frequently associated with the selection of YMDD resistant mutants.1–3 The use of post-LT combination therapy with passive immunoprophylaxis (HBIG) and LAM also reduces the rate of HBV recurrence in viremic subjects, because these drugs act synergistically on HBV replication.8–14
Pre-LT YMDD mutant selection can pose an increased risk of HBV recurrence, related to a possible inefficacy of the prophylactic combination therapy15, 16; however, the effects of the viral load at the time of LT and, consequently, of the genotypic versus the phenotypic resistance17, 18 on the rate of recurrence or the need for ADV therapy in addition either before or after LT, are still not clear.
In this study, a correlation between HBeAg positivity (indicating a higher viral fitness), duration of LAM therapy, and phenotypic resistance was confirmed, in accordance with previous reports.2, 3 However, a low risk of HBV recurrence was observed in patients with genotypic resistance (group B) who underwent transplantation with a viral load of less than 5 log10 copies/mL and who were treated, after surgery, with a combined prophylaxis of HBIG and LAM. By contrast, both patients with phenotypic resistance (group A) who selected YMDD mutants with a significant virologic breakthrough of greater than 5 log10 copies/mL before LT and not treated with ADV before LT developed HBV recurrence while on LAM and HBIG prophylaxis.
The lack of the protective effect of combined prophylaxis with HBIG and LAM in highly viremic patients confirms that in YMDD carriers, the synergistic efficacy of combination is directly related to the viral load at the time of LT, in agreement with reports from Germany.13, 14 By contrast, the level of viremia at the time of transplantation was not relevant in both patients with genotypic resistance who stopped HBIG after LT. In this setting, the HBV recurrence was related to the discontinuation of HBIG and to the maintenance of LAM monoprophylaxis in presence of the YMDD variant, as previously described.2
The characteristics of the study, as a retrospective uncontrolled clinical trial, do not allow definitive conclusions but give some indication of the association of antiviral(s) and immunoprophylaxis after LT in YMDD carriers. In this setting, further controlled studies are needed to define the most appropriate prophylaxis regimen (a double or a triple prophylaxis), but the prevalence of YMDD variants in these patients could favor the use of ADV in LT candidates and the addition of ADV to HBIG after LT both in genotypic and phenotypic YMDD resistance.
The very low frequency of escape mutants associated with ADV therapy and the risk of hepatitis flares leading to decompensation of cirrhosis after YMDD mutants selection could suggest that ADV should replace LAM for antiviral therapy before LT in all viremic patients listed. However, in this setting the role of LAM remains important, in our opinion, because the faster antiviral effect of LAM is effective in patients with cirrhosis (± hepatocellular carcinoma) who need an LT in an acceptable (6–12 months) timeframe, while a careful virologic monitoring and ADV addition in case of YMDD selection can prevent hepatitis flares, virologic breakthroughs, and a new risk of post-LT recurrence.
In conclusion, this study confirms the importance of a dynamic virologic analysis in HBsAg carriers treated with LAM both before and after LT. In these patients, the majority of HBV recurrences are highly preventable by using combined prophylaxis with HBIG and LAM, but preemptive therapy with LAM before LT can select resistant mutants.
Conventional combined prophylaxis with LAM and HBIG for prevention of HBV recurrence maintains its efficacy in YMDD genotypic resistance when HBV DNA levels remain below 5 log10 copies/mL. Thus, these patients should not be removed from the waiting list but could require ADV addition in relation to the risk of a significant viral relapse during further follow-up on the waiting list. The risk of recurrence becomes very high with LAM monoprophylaxis, and this regimen or HBIG discontinuation must be avoided in the presence of YMDD at the time of transplantation. In these patients, the use of ADV and HBIG after LT could be justified by the baseline presence of the LAM-resistant variant. However, when YMDD carriers develop a virologic breakthrough before LT (phenotypic resistance), the risk of recurrence during the conventional combined prophylaxis with HBIG and LAM becomes very high, but the use of ADV before LT aimed to reduce the viral load associated with HBIG after surgery permits the control of HBV recurrence.