Preventing the recurrence of hepatitis B virus (HBV) infection is essential after liver transplantation in patients infected with HBV. In Korea, where HBV infection is prevalent in the general population, patients who test positive for hepatitis B surface antigen (HBsAg) account for 80% of all adult liver transplant recipients; hepatitis C virus infection and alcohol-related cirrhosis account for only a small proportion of liver transplants.
Posttransplant HBV prophylaxis modalities are roughly divided into 3 types: hepatitis B immunoglobulin (HBIG) monotherapy, antiviral agent monotherapy, and combination therapy.1–3 Initial HBIG therapy and subsequent antiviral monotherapy are also used in practice. Posttransplant vaccination appears to be a promising strategy, but its effect on HBV-associated adult recipients requires further investigation.1, 4–8 HBIG therapy, although known to be very effective, is associated with several disadvantages, such as the high cost of therapy, the cumbersome method of administration, the adverse effects of HBIG, and the possible development of mutation.1, 9, 10 In particular, because of the high cost of HBIG therapy, antiviral therapy using lamivudine has been a subject of interest since the late 1990s. However, because of frequent mutations, the HBV breakthrough rate with that form of treatment was much higher than expected, and the use of combination therapy was a matter of concern. The introduction of new antiviral agents makes the HBV prophylaxis regimens more diverse and effective than ever before.3
There is a lack of consensus on the upper limit of the HBV infection recurrence rate below which a specific HBV prophylaxis regimen can be effective, although ideally this rate should be zero. In actual practice, it has been seen that cost considerations have scored over effectiveness of prophylaxis when a specific HBV prophylaxis regimen has been selected.
Because HBV-associated liver disease remains the primary indication for adult liver transplantation in Korea, the need for selecting a highly effective HBV prophylaxis regimen cannot be overemphasized. In our institution, high-dose HBIG monotherapy is used as the primary HBV prophylaxis regimen, treatment with antiviral agents being used a rescue therapy for HBV infection recurrence. To evaluate the efficacy of this regimen, we analyzed the results of treatment in 639 adult patients with HBV-associated end-stage liver diseases who underwent living donor liver transplantation (LDLT).
anti-HBc, hepatitis B core antibody; anti-HBs, antibody to hepatitis B surface antigen; CI, confidence interval; HBeAg, hepatitis B e antigen; HBIG, hepatitis B immunoglobulin; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HBV DNA, hepatitis B virus deoxyribonucleic acid; HCC, hepatocellular carcinoma; LDLT, living donor liver transplantation; PCR, polymerase chain reaction; SD, standard deviation.
PATIENTS AND METHODS
Between February 1997 and December 2004, 776 adult patients underwent LDLT in our institution, with the exclusion of retransplantation. After the exclusion of 5 patients with both HBV and hepatitis C virus infection and 1 patient who received an HBsAg-positive liver graft,11 639 (82.3%) of those 776 patients were diagnosed as having HBV-associated end-stage liver diseases. Forty-four (6.9%) of the 639 patients died within the first 3 months after LDLT. The surviving patients were monitored until April 2007 or their death, and their medical records were retrospectively reviewed with respect to prevention and surveillance for HBV recurrence.
Viral markers [HBsAg, antibody to hepatitis B surface antigen (anti-HBs), hepatitis B core antibody, hepatitis B e antigen (HBeAg), and hepatitis B virus deoxyribonucleic acid (HBV DNA) probe] were routinely checked as a part of the pretransplant workup for HBV in recipients and living donors. HBV DNA probe was measured by means of a nonamplified hybrid capture assay with a rough detection threshold of 3.3 pg/mL, which is equivalent to about 900,000 copies/mL. In this study, pretransplant replicative HBV infection was defined as serum HBeAg positivity and/or HBV DNA probe positivity. The HBV genotype was not investigated because nearly all HBV infections in Korea are known to be genotype C. HBV DNA polymerase chain reaction (PCR) assay, which has a detection threshold of 50 copies/mL, has also been performed routinely in a late phase of the study since 2003.
Following LDLT, the quantitative HBsAg and anti-HBs titers of the recipients were monitored at every follow-up visit to the outpatient clinic. The serum anti-HBs titers, determined only by immunoradiometric assay, were used in this study. When HBV recurrence was suspected, every viral marker was assessed again, and serial follow-up with HBV DNA PCR assay was performed. An HBV DNA PCR assay was also used to monitor the response to treatment in patients with HBV recurrence. Studies to determine viral mutation were also conducted to identify resistance to lamivudine or adefovir. In order to differentiate HBV recurrence from graft rejection, a percutaneous or transjugular liver biopsy was performed.
High-Dose HBIG as a Primary HBV Prophylaxis Protocol
The HBIG monotherapy protocol used during the study period was as follows. During the anhepatic phase, 10,000 IU of HBIG (Green Cross Corp., Korea) was infused; during the first week after liver transplantation, 10,000 IU of HBIG was infused daily; for the next month, 10,000 IU of HBIG was infused once each week; and thereafter, 10,000 IU of HBIG was infused once per month to maintain a trough serum level greater than 500 IU/L. During the study period, most patients received a slow intravenous infusion of an intramuscular preparation of HBIG in order to reduce adverse effects. Recently, a highly purified intravenous preparation of HBIG has been available in Korea, enabling the rapid infusion of 10,000 IU in 1 hour.
This protocol is a regimen typical of high-dose HBIG monotherapy, and it was uniformly applied even to HBeAg-positive or HBV DNA probe–positive recipients. The sequential changes of anti-HBs titer according to the time sequences were estimated as follows. During the first week, the trough level of anti-HBs was sustained at more than several thousand international units per liter; the peak values of anti-HBs reached over 5000 IU/L; at 1 month of therapy, the anti-HBs trough level had increased to more than 1000 IU/L; and thereafter, the anti-HBs trough level decreased to a value between 500 and 1000 IU/L. After the first 6 to 12 months of treatment, the interval of HBIG infusions was determined to be set at 30 to 60 days on a case-to-case basis because of great interindividual and intraindividual variations in HBIG requirements.12
In patients with unusually low anti-HBs titers, despite the regular infusion of HBIG (that is, the serum level of HBIG was lower than 500 IU/L on 2 consecutive measurements, even after the infusion of HBIG in shortened intervals of 20-25 days) but without any evidence of HBV recurrence, HBV DNA PCR assay was performed, and lamivudine (100 mg/day) was administered preemptively as a combination therapy regimen. Lamivudine treatment was carefully tapered off only after restoration in the anti-HBs titer of greater than 500 IU/L with an HBIG infusion interval of 30 days or more. For patients undergoing anticancer treatment for recurrent hepatocellular carcinoma (HCC), lamivudine was added during the latter part of the study period with the aim of enhancing the prophylactic effect.13
Rescue Antiviral Therapy for HBV Recurrence
HBV recurrence was defined as the patient becoming seropositive for either HBsAg or HBeAg. HBV recurrence diagnosed only at the terminal stage of recurrent disseminated HCC was not taken into account in assessing the overall incidence of HBV recurrence because such patients were often unable to receive regular HBIG infusion but was included only for multivariate analysis. When the serum anti-HBs titers decreased rapidly despite scheduled HBIG infusions, lamivudine was added as combination therapy. Once HBV recurrence was diagnosed in a patient, HBIG therapy was discontinued in principle and replaced with antiviral therapy using 1 or 2 antiviral agents. After its introduction in Korea, adefovir was used as a sole agent or in combination therapy when the patient's virologic or biochemical response to lamivudine was inadequate.14–17 Since 2004, lamivudine and adefovir have been administered concurrently as treatment for HBV recurrence, regardless of lamivudine resistance. Since 2006, when entecavir became available in Korea, it has been used as a first-line or second-line rescue agent for patients with HBV recurrence. During the acute phase of HBV recurrence, immunosuppressive therapy was either reduced or temporarily terminated according to the graft liver function. Steroid pulse therapy was banned when the results of liver biopsy showed concurrent HBV recurrence and acute graft rejection. Mutant studies on lamivudine and adefovir resistance have been performed routinely at our institution during the last few years, but no such study has been performed for HBIG resistance.
The peritransplant primary immunosuppression protocol used at our institution for adult patients undergoing LDLT for HBV-associated end-stage liver disease consists of interleukin-2 receptor inhibitor (basiliximab) on days 0 and 4 and an intraoperative steroid bolus (10 mg/kg), an intravenous or oral calcineurin inhibitor, and corticosteroid recycling since day 1; adjunctive mycofenolate mofetil is reserved for patients showing calcineurin inhibitor–associated side effects. Corticosteroids are rapidly tapered off within the first 3 months.
All numerical data are reported as the mean with the standard deviation or as the median with the range. Survival curves were estimated with the Kaplan-Meier method and compared with the log-rank test. The categorical variable was compared by the chi-square test. The variables reaching statistical significance by univariate analysis were then included for multivariate analysis with proportional hazard regression. A P value of less than 0.05 was considered statistically significant.
The clinical features of 639 patients undergoing LDLT for HBV-associated end-stage liver diseases are summarized in Table 1. The overall 5-year survival rate of these patients was 80.7% (Fig. 1). Forty-four patients (6.9%) died during the first 3 months after LDLT because of various surgical and medical complications not related to HBV recurrence. Another 76 patients died during the mean follow-up period of 53 months (range, 4-124 months). The main causes of those late deaths were HCC recurrence or de novo malignancy (n = 36), chronic rejection and other complications of graft function (n = 20), HBV recurrence (n = 6), and a few other causes such as biliary complications, cerebrovascular accident, trauma, and suicide (n = 14). During the follow-up period, HBV recurrence was diagnosed in 38 (6.4%) patients, leading to an overall 5-year recurrence-free survival rate of 75.3% (Fig. 1).
Table 1. Clinical Features of the Liver Transplant Recipients with HBV-Associated End-Stage Liver Diseases and Living Donors
Abbreviations: anti-HBc, hepatitis B core antibody; anti-HBs, antibody to hepatitis B surface antigen; HBeAg, hepatitis B e antigen; HBV, hepatitis B virus; HBV DNA, hepatitis B virus deoxyribonucleic acid; SD, standard deviation.
Patients with serum HBeAg positivity and/or HBV DNA probe positivity.
Dual graft means simultaneous implantation of 2 partial liver grafts from 2 donors.
Primary HBV Prophylaxis Protocols Using HBIG and Antivirals
All 595 patients who survived more than 3 months received high-dose HBIG therapy indefinitely until the diagnosis of HBV recurrence. The time interval between HBIG administrations was decided on a case-to-case basis after the first 6 to 12 months of liver transplantation. The mean interval was 30 days in 68 (11.4%) patients, 40 days in 96 (16.1%), 45 days in 178 (29.9%), 50 days in 155 (26.1%), and 60 days in 98 (16.5%). Of these, 71 patients also received lamivudine in addition to HBIG, for a duration varying from 3 months to 4 years, during the follow-up period. At the time of last follow-up for this study, 19 (3.9%) of the 486 survivors without HBV recurrence received combination therapy with an HBIG infusion interval of 30 days in 8 (26.7%) of 30 patients, 40 to 50 days in 9 (2.5%) of 360 patients, and 60 days in 2 (2.1%) of 96 patients.
Rates and Patterns of HBV Recurrence
Because no HBV recurrence was detected during the first 3 months of liver transplantation, 44 patients who died during the perioperative period (≤ 3 months) were excluded from this HBV recurrence study. Of 595 surviving patients, 38 (6.4%) experienced HBV recurrence during the mean follow-up period of 53 months, leading to a 5-year HBV recurrence rate of 7.3% (Fig. 2A). HBV recurrence occurred after a mean of 25.7 ± 16.4 months [median, 20 months (range, 5-74 months)] after LDLT.
When HBV recurrence was initially suspected, serum levels of both HBsAg and HBeAg were simultaneously checked in 19 of 38 patients. The results revealed HBeAg positivity before the appearance of HBsAg positivity in 12 (63.2%) patients, only HBsAg positivity in 4 (21.1%), and concomitant positivity for HBsAg and HBeAg in 3 (15.8%). Within 1 month after seroconversion to HBeAg positivity, seroconversion to HBsAg positivity occurred in all patients, even though their anti-HBs titer was in the range of tens to hundreds of international units per liter.
The results of liver biopsies performed in 18 of 38 HBV-recurrent patients demonstrated pathologic evidence of HBV infection in all patients. Three patients also demonstrated concomitant evidence of acute rejection, which was treated successfully by an increase in the immunosuppression without use of steroid pulse therapy.
Pretransplant replicative HBV infection (serum HBeAg positivity and/or HBV DNA probe positivity) resulted in a higher rate of HBV recurrence: The 5-year HBV recurrence rate was 10.2% in the replicator group (n = 364) versus the recurrence rate of 2.9% in the nonreplicator group (n = 231; P = 0.009; Fig. 2B). When the HBeAg status was considered alone, there was also a significant increase of HBV recurrence in HBeAg-positive patients (P = 0.001). A history of use of an antiviral agent during the pretransplant period did not influence the posttransplant HBV recurrence rate significantly (P = 0.682). In addition, anti-HBs positivity with or without hepatitis B core antibody positivity in living donors did not make any difference in the HBV recurrence rate (P = 0.873).
Without consideration of concomitant antiviral administration, the intervals for HBIG administration were inversely correlated with the HBV recurrence rates (P < 0.001; Fig. 3). When only patients with a 30-day HBIG infusion interval were considered, the concomitant use of an antiviral agent was associated with a significant reduction in HBV recurrence (P = 0.046; Fig. 4).
Malignancy and HBV Recurrence
During follow-up, 36 patients died from recurrent or de novo cancers (HCC in 33 patients, cholangiocarcinoma in 1, lung cancer in 1, and gastric cancer in 1), despite vigorous anticancer treatment. The HBV recurrence rate in patients with recurrent HCC was 24.2% versus 6.9% in the patients with no evidence of HCC recurrence (P = 0.002). Eight of 33 patients who had recurrent HCC demonstrated seroconversion to HBsAg positivity after having received HBIG monotherapy, and these patients were then treated with antiviral therapy. During the later part of the study period, combination therapy or antiviral therapy was used for patients with HCC recurrence.
Risk Factor Analysis for HBV Recurrence
In univariate analysis, the significant risk factors for HBV recurrence were active pretransplant HBV replication, need for frequent HBIG infusions (≤30 days), and HCC recurrence. In multivariate analysis, all these variables were also found to be independent predictive factors for HBV recurrence (Table 2).
Table 2. Multivariate Analysis for the Posttransplant Recurrence of HBV Infection
Abbreviations: CI, confidence interval; HBeAg, hepatitis B e antigen; HBIG hepatitis B immunoglobulin; HBV, hepatitis B virus; HBV DNA, hepatitis B virus deoxyribonucleic acid.
Patients with serum HBeAg positivity and/or HBV DNA probe positivity.
Need for frequent HBIG infusions (interval ≤ 30 days)
Hepatocellular carcinoma recurrence
Rescue Treatment for HBV Recurrence
Until late 2002, lamivudine with or without famciclovir was first administered to the patients diagnosed with recurrent HBV infection (n = 9). Of these patients, 5 died because of rapidly progressive loss of liver function 1 to 4 months after HBV recurrence. The results of percutaneous liver biopsy in these patients revealed pathological changes consistent with those of fibrosing cholestatic hepatitis.
After the introduction of adefovir in late 2002, use of adefovir with or without lamivudine was the first-line rescue therapy for HBV recurrence. Since 2004, the concurrent administration of adefovir and lamivudine has been routinely used for the treatment of HBV recurrence. Since the introduction of adefovir, only 1 of 29 consecutive patients with HBV recurrence have died from liver failure due to intractable HBV infection. Recently, entecavir has been regarded as an essential rescue agent for HBV recurrence.
HBV DNA PCR assays were performed in the 23 patients who demonstrated HBV recurrence but stable liver function after treatment with adefovir and lamivudine. The results of this assay showed that HBV DNA was undetected in all but 2 of these patients. The HBV DNA levels in these 21 patients were as follows: 2 log10 copies/mL in 1 patient, 3 log10 copies/mL in 8 patients, 4 log10 copies/mL in 5 patients, 5 log10 copies/mL in 2 patients, and greater than 5 log10 copies/mL in 5 patients. The conversion to HBsAg seronegativity was usually achieved when the viral load fell to about 3 log10 copies/mL. Entecavir and clevudine were administered to 2 patients who demonstrated a sustained HBV DNA PCR result of higher than 6 log10 copies/mL. One of these patients exhibited a significant decrease in viral load during the initial 6 months of therapy.
HBV prophylaxis regimens incorporating HBIG are usually of 2 types, that is, high-dose or low-dose therapy. Generally, during the early postoperative period, a very high serum anti-HBs titer is maintained with frequent HBIG infusions. Thereafter, the predose level of anti-HBs is determined according to a customized protocol, such as a titer greater than 500 IU/L in high-dose protocols and a titer greater than 100, 150, or 200 IU/L in low-dose protocols. A combination therapy regimen using lamivudine has usually been accompanied by low-dose HBIG therapy.18, 19
There have been a few large-volume, long-term follow-up studies on HBIG therapy. In a single-center series of 284 patients,20 most received HBIG monotherapy, combination therapy with lamivudine being reserved for some high-risk patients. In this study, the overall 10-year actuarial survival rate was 74.4%, and the 10-year actuarial HBV recurrence rate was 25.4%.20 Marzano et al.21 examined the outcome of treatment with low-dose HBIG monotherapy versus combination therapy in 177 patients. They found that the serum anti-HBs titer was greater than 500 IU/L during the first 3 months of therapy, and thereafter it remained above 100 IU/L for the rest of the patient's life. In both study groups, the overall 5-year HBV recurrence rate was 8%.21 It is known that low-dose HBIG monotherapy is not always effective when it is used for treating HBV DNA replicators. It has been reported that the HBV recurrence rate with low-dose HBIG monotherapy, although low in nonreplicators, is definitely high in replicators.21 In the single-center series by Roche et al.,20 the effects of high-dose and low-dose HBIG therapies were evaluated only in HBV DNA replicators. It was found that the risk of HBV recurrence was very high in the low-dose HBIG group compared to the high-dose HBIG group. These results suggest that a high titer of anti-HBs (that is, >500 IU/L) can prevent the failure of prophylactic therapy, even in HBV DNA replicators. The 5-year patient survival rate of 80.7% and 5-year HBV recurrence rate of 7.3% in our study are comparable to those in the previously mentioned studies.20, 21
There are 2 main causes of treatment failure during HBIG prophylaxis: a low titer of serum anti-HBs during the early posttransplant period and escape mutations in the pre-S region and “a” determinant later on. An inadequately low titer of serum anti-HBs during the early posttransplant period appears to be the main cause of early treatment failure. A large amount of HBIG is required for ensuring adequate anti-HBs titers during the early posttransplant period. In our study, HBV DNA became undetectable on PCR assay after the infusion of a total of 50,000 to 100,000 IU of HBIG over a period of 1 week, although HBsAg levels became undetectable only a few days after HBIG therapy. These observations provide the rationale for administering a high dose of HBIG daily for the first week of therapy. Escape mutations in the pre-S region and “a” determinant appear to be the main causes of late failure.1, 9, 22 Those mutations have been found to reduce the binding capability of HBIG, thereby diminishing the protective efficacy. Escape mutations often result in a paradoxical situation of concurrent HBsAg positivity and high anti-HBs titer, which can persist for several weeks. It has been reported that these mutations can revert to pretransplant sequences after the termination of treatment with HBIG, but the efficacy of HBIG in such cases has not been determined.9
HBV breakthrough during high-dose HBIG therapy resulted in the early appearance of HBeAg in 63.2% of patients in our study prior to their seroconversion to HBsAg positivity. This appears to be unusual because of the natural sequences of acute HBV infection in nontransplant patients. It may be suggested that reappearance of HBsAg due to a mutation may be partially hindered by the high anti-HBs titers. Differences in the sensitivity of HBeAg and HBsAg assays could be another reason for this early appearance of HBeAg positivity. Therefore, when HBV recurrence is suspected after HBIG therapy, a serologic study should include tests for both HBeAg and HBsAg at the same time.
It has been suggested that HBV prophylaxis regimens can be tailored according to the patient's pretransplant HBV status. The viral load at the time of transplantation seems to be associated with posttransplant HBV recurrence. However, the effect of intentional antiviral-induced reduction of the viral load prior to liver transplantation remains questionable when high-dose HBIG therapy is to be used after transplantation because HBV in the tissue and circulating blood will be eliminated soon by HBIG therapy.13, 21 The result of this study showed that pretransplant use of an antiviral did not significantly influence the posttransplant HBV recurrence.
We found that a 30-day HBIG infusion interval was used in 44 (12.1%) of the surviving 354 patients with pretransplant HBV replication and in 24 (10.4%) of 231 patients without pretransplant HBV replication (P = 0.504). This result suggests that the amount of neutralizing HBIG after seroconversion to HBsAg negativity does not vary significantly on the basis of the patient's status of pretransplant HBV replication. However, in our study, the need for more frequent HBIG infusions was found to be an independent risk factor for HBV recurrence. With the effect of concomitant antiviral administration disregarded, the interval established for HBIG infusions was related inversely to HBV recurrence (Fig. 3). Moreover, the concomitant use of an antiviral agent decreased the risk of HBV recurrence in patients who required a monthly infusion of HBIG (Fig. 4). These data suggest that there may be a close correlation between the posttransplant viral load and HBV recurrence rate. In our clinical practice, we do not consider pretransplant antiviral-induced reduction of the HBV load yet. Instead, we would suggest the use of combination therapy in patients requiring monthly HBIG infusions, particularly those patients who were found to be HBV replicators on pretransplant workup. As our results show, there was no HBV recurrence in patients who received the timely concurrent antiviral agent therapy (n = 6), whereas HBIG monotherapy conferred a 5-year HBV recurrence rate of 32.3% (n = 38; P = 0.110).
An important but often neglected aspect is the discrepancy in anti-HBs titers when they are measured by different methods such as microparticle enzyme immunoassay, chemiluminometric sandwich immunoassay, and chemiluminescent microparticle immunoassay. This discrepancy is even more evident when titers greater than 500 IU/L are measured. This variation in titers is probably caused by the limitation of equipment calibration. At our institution, some enzyme immunoassay methods gave anti-HBs titers that were 20% to 40% lower than the values obtained from the radioimmunoassay method; therefore, it is our practice to use the results from only one method of radioimmunoassay as reference values. It would be advisable to carefully compare anti-HBs titer values when a patient is being monitored in different institutions.
In cases in which HBV recurrence occurs despite high-dose HBIG therapy, it would be prudent to infer that HBIG is no longer effective. In such patients who experience HBV breakthrough while on high-dose HBIG therapy, it is our practice to taper off the HBIG therapy and replace it with antiviral agents. In patients who demonstrate seroconversion to HBsAg negativity after antiviral treatment, we have never attempted to reinstate HBIG monotherapy because the outcome in such cases cannot be guaranteed, although the possibility of inducing wild-type HBV after the prolonged use of antiviral agents cannot be excluded.
Seroconversion to HBsAg negativity after HBV recurrence has often been achieved by the use of antiviral agents. However, the clinical significance of this development should not be overestimated because it is known that patients can exhibit alternating positive and negative seroconversions during long-term follow-up. A paradoxical situation was observed in our study, in which resolution of HBsAg positivity occurred in a patient with an HBV DNA load of 4 log10 copies/mL and anti-HBs titer > 1000 IU/L. It would therefore be advisable to include HBV DNA PCR assay as a part of the routine evaluation of patients with HBV recurrence.
HBV recurrence has been found to be associated with chemotherapy administered to treat recurrent HCC.13 In our series, 24.2% of the patients with terminal HCC metastasis demonstrated HBsAg positivity. This might have been caused by poor host defense mechanisms in these patients and have been partially due to reluctance on the part of the patient to undergo frequent HBIG infusions in light of reduced life expectancy. Having considered those factors, we now regard HCC recurrence as an important indication for starting combination or antiviral therapy as HBV prophylaxis.
High-dose HBIG monotherapy, although highly effective, is not very cost-effective. The high cost of HBIG limits the routine use of this therapy in many countries. However, strictly speaking, cost effectiveness per se is a socioeconomic factor. In Korea, this problem of the high cost of HBIG therapy has been solved by the extraordinarily extended coverage of the HBIG cost by the social medical insurance program, and so the financial burden does not vary greatly among patients receiving different regimens of posttransplant HBV prophylaxis. This extensive financial support enables us to continue life-long, high-dose HBIG therapy.12, 23
In conclusion, primary high-dose HBIG monotherapy and rescue antiviral therapy resulted in a 5-year HBV recurrence rate of 7.3% and a 13.2% mortality rate after HBV recurrence. The indications for combination therapy were the need for frequent HBIG infusions, active pretransplant HBV replication, and HCC recurrence. Use of antiviral agents such as lamivudine and adefovir or entecavir appears to be the first-line rescue therapy for recurrent HBV infection.