Combined Lamivudine and Hepatitis B Immunoglobulin for the Prevention of Hepatitis B Recurrence after Liver Transplantation: Long-Term Results
* Corresponding author: Jérôme Dumortier, jerome. firstname.lastname@example.org
For the prevention of recurrent hepatitis B virus (HBV) infection after liver transplantation (LT), the efficacy of hepatitis B immunoglobulin (HBIg) has been largely demonstrated. The aim of this pilot study was to determine if the addition of lamivudine to HBIg in the prevention of HBV recurrence after LT could be more effective. Sixty HBsAg-positive/HBV DNA-negative patients underwent LT from October 1990 to December 2001. All 60 patients received intravenous HBIg to maintain serum anti-HB levels above 500 IU/L, indefinitely. Since 1997, 17 patients have received combined oral lamivudine (150 mg/day) and HBIg, and were compared with the historical cohort of 43 patients. In the historical control group, the recurrence rate was 10/43 (23%) after a 98-month median follow-up. Five patients died from HBV-related liver disease. After a 30-month median follow-up, none of the 17 patients in the combined prophylaxis group experienced HBV recurrence, and HBV DNA was undetectable by PCR in at least three serum samples per patient. HBV recurrence was significantly lower when compared with the historical control group (10/43 vs. 0/17, p < 0.01). Our results suggest that combined lamivudine and HBIg can avoid the recurrence of HBV infection in patients who are HBsAg-positive/HBV DNA negative before LT.
Liver disease due to hepatitis B virus (HBV) is a frequent indication for liver transplantation (LT). After LT for HBV-related liver disease, the recurrence of HBV infection is universal in the absence of prophylaxis, concerning more than 80% of patients and the risk of progression from chronic hepatitis to cirrhosis is well known (1,2). The use of hepatitis B immunoglobulin (HBIg), indefinitely after surgery, aiming to maintain serum anti-HBs levels above 100 IU/L, was established as a standard prophylaxis for all HBsAg-positive liver transplant recipients (3–5). However, prophylaxis with HBIg is not always effective and the 5-year actuarial rate of HBV recurrence, defined as re-appearance of HBsAg in the serum, is 20–35% for HBV DNA-negative patients (6). New antiviral agents, including lamivudine, and more recently adefovir dipivoxil and entecavir, are considered a major advance in anti-HBV therapy, and have made HBV liver disease an indication for LT irrespective of viral replication status, and therefore a complete turn around from 10 years ago (7,8). An alternative approach for prophylaxis of HBV recurrence after liver transplantation could be to use lamivudine, an oral nucleoside analogue, a potent inhibitor of HBV replication with an excellent safety profile, alone or in association with HBIg (9). The long-term results of such a new prophylaxis strategy are partial. Between 1997 and 2001, we conducted a prospective pilot study aiming to investigate the efficacy of a new prophylactic regimen involving a combination of HBIg and lamivudine for the prevention of HBV recurrence in HBsAg-positive/HBV DNA-negative liver transplanted patients. We present here our experience, from this cohort of patients regarding the recurrence of HBV infection after LT, in comparison with the patients previously treated with HBIg as a monotherapy, before 1997.
Patients and Methods
From October 1990 to December 2001, LT were performed in 563 patients in our institution, and 62 were HBsAg-positive/HBV DNA-negative at the time of LT. Two patients died during the early post-operative period and were excluded from the study. There were 50 men and 10 women; the mean age was 43 years (range 17–60 years). All the patients were HBeAg-negative and had antibodies to HBeAg (anti-HBe) and HBcAg (anti-HBc). The patients were treated after LT with a standard immunosuppressive regimen (cyclosporin or tacrolimus and azathioprine or mycophenolate mofetil with steroids, see Table 1). In patients from the historical cohort, the immunosuppressive regimen usually consisted of cyclosporin, steroids and azathioprine. Steroids were tapered from day 0 to day 5 at 25 mg/day and discontinued 3–12 months after LT. Acute rejection was treated with pulses of 1 g of methylprednisolone intravenously (i.v.) Azathioprine was stopped after 1 year and suspended if indicated (leukopenia). From 1997, the amount of steroids was the same but the immunosuppressive regimen usually consisted of tacrolimus (trough level of 8–12 ng/mL during the first 3 months, 3–6 ng/mL thereafter), steroids and mycophenolate mofetil.
Table 1. Clinical and virological characteristics of the 60 liver transplant patients included in the study
|Number of patients||43||17|
|Median age (years, range)||40.1||49.9|
| Steroids disclosure (months, range)||7.6 (3.8–12)||5.8 (3.5–12)|
|Acute rejection episodes||11||3|
|Median follow-up (months, range)||98 (57–137)||30 (12–48)|
Historical control group Forty-three patients transplanted between 1990 and 1996 were used as controls. The patients received i.v. HBIg to maintain serum anti-HB levels above 500 IU/L. HBIg administration consisted of 10 000 IU during the anhepatic phase, followed by 10 000 IU daily for the next week, and 10 000 IU at day 15 and day 30. Another 10 000 IU was then given routinely over the long term, every 4–8 weeks according to the anti-HBs level, indefinitely.
Combined prophylaxis group Since January 1997, we prospectively studied 17 HBsAg-positive/HBV DNA-negative liver transplanted patients, who received lamivudine (GlaxoWellcome, Greenford, UK), given orally at a daily dose of 150 mg, after LT, in addition to passive immunoprophylaxis with i.v. HBIg, according to the protocol given to the historical control group. The lamivudine dose was adjusted in the case of renal insufficiency. The protocol of the study was approved by the Ethics Committee of Hospices Civils de Lyon.
Follow-up. Patients were seen for their HBIg i.v. administration every 4, 5 or 6 weeks and blood samples were taken for full blood counts, liver function tests, serum HBsAg, anti-HBs, serum HBeAg and anti-HBe and HBV DNA. HBV recurrence was suspected in the case of reappearance of HbsAg in serum, which was monitored in all patients and then confirmed by detection of HBsAg and/or HBcAg on liver biopsy and serum HBV DNA detection. Liver biopsies were taken whenever graft dysfunction was suspected; they were obtained routinely 1, 5 and 10 years after LT in all patients.
HBsAg, anti-HBs, anti-HBc, HBeAg, anti-HBe, anti-HCV and anti-Delta antibodies were detected with enzyme immunoassays (Bio-Rad, Marnes la Coquette, France, Abbott, Rungis, France and DiaSorin, Saluggia, Italy). Between 1990 and December 1996, serum HBV DNA was quantitated with the solution hybridization assay Genostics (Abbott), which has a lower limit of detection of 2.5 pg/mL. The HBV DNA testing in the serum samples was performed after December 1996 using the signal amplification assay (Quantiplex bDNA assay, Chiron Corp. then Versant HBV Bayer, Puteaux, France), which has a lower limit of detection of 0.7 MEcopies/mL (2.5 pg/mL).
In the combined prophylaxis group, PCR testing of all serum was performed blindly in a central laboratory, without any knowledge of the patients' details or the results of the other HBV markers. Total DNA was extracted from 200 µL serum using the QIAmp whole blood kit (Qiagen Ltd, Crawley, UK). The sensitivity of the PCR reaction used in this study has been estimated at 400 copies/mL.
The liver graft biopsies were stained using a rabbit polyclonal antibody to HBsAg (Janssen Laboratories, Beerse, Belgium) and a rabbit polyclonal antibody to HBcAg (Dako Ltd, Glostrup, Denmark). The immunostaining of liver sections for all patients in this study was performed by the same method in a central laboratory.
Statistical analysis was performed with the StatView package using t-test for dependent or independent samples whenever appropriate, and for the Kaplan–Meier actuarial survival curves. Univariate analysis was performed to determine the significant prognostic factors for survival using the log-rank test: an associated hepatocellular carcinoma (HCC) at the time of LT, the absence of HDV co-infection and the recurrence of HBV infection. The chi-square test was used to assess the difference in the HBV recurrence rate between the combined prophylaxis and historical control groups.
Clinical and virological characteristics of the 60 LT patients included in the study are summarized in Table 1.
In the historical control group, the recurrence rate was 10/43 (23%) after a median follow-up of 98 months (range 57–137 months), in a median delay of 19 months (range 4–110 months). Two patients died from HBV-related fibrosing cholestatic hepatitis before 6 months post-LT. Seven patients experienced a progressive evolution to liver HBV-related cirrhosis, leading to death in three cases, and re-LT in one (after lamivudine rescue therapy and negativation of HBsAg).
All 17 patients who survived surgery in the combined prophylaxis group did not experience HBV recurrence, after a median of 30 months of follow-up (range 12–48 months). Two of them died after LT from sepsis. HBV recurrence was significantly lower (p < 0.01; chi-square test) when compared with the historical control group (10/43 or 23%). In all the 17 patients in the combined prophylaxis group, HBV DNA was undetectable by PCR in at least three serum samples per patient.
Sixty-four liver biopsies obtained from the 60 patients after LT were stained by immunohistochemistry for HBcAg and HBsAg, and both HBcAg and HBsAg were found in the liver of all the patients who experienced HBsAg re-appearance, but in none of the other patients.
Actuarial survival rate was 79.8% at 5 years and 76.9% at 10 years. Significant prognostic factors were: an associated HCC at the time of LT, the absence of HDV co-infection and the recurrence of HBV infection (p < 0.05 using log-rank test).
The major goal after LT in HbsAg-positive patients is to prevent the recurrence of HBV infection, and thereafter the graft loss due to fibrosing cholestatic hepatitis or cirrhosis. Our study strongly confirms the clinical usefulness of a combination prophylaxis, based on data obtained from a large number of patients and a prolonged follow-up of an average of 2.5 years. Because we compared the results of the combined prophylaxis regimen to a historical cohort, it must be stated that the effect of HBV prophylaxis on post-transplant outcomes may be confounded by changes in donor quality, immunosuppression regimens and other facets of perioperative and post-transplant care between the two eras.
Several studies have shown that, using prophylaxis with long-term HBIg administration, HBV recurrence can be reduced to 20–30% in HBV-infected patients, related to the initial liver disease and the level of HBV replication at the time of transplantation. The rate of HBV infection recurrence was found to be 0–10% in fulminant hepatitis B patients, 10–20% in HBV-HDV co-infected cirrhotic patients, 20–35% in HBV cirrhotic patients without HBV replication, and 30–80% in HBV cirrhotic patients with HBV replication (6). Our results from the historical cohort of HBsAg-positive/HBV DNA-negative patients are in accordance with these previous reports, with a 23% recurrence rate. Both delta antibody positivity and fulminant hepatic failure have been associated with an improved outcome after LT in HBsAg positive patients. Thus, one would have expected an even lower rate of recurrence in our HBIg-only group. In this cohort, recurrence was observed in patients transplanted for HBV-related cirrhosis in 8 out of 10 of the cases.
Despite a good long-term tolerance, long-term HBIg administration has several drawbacks: HBIg administration is expensive, constraining because of the need for frequent injections depending on serum anti-HBs antibody levels, and, last but not least, HBV infection recurrence remains possible, even in low-risk groups, as observed in our series. Moreover, several studies have shown that HBIg therapy imposes a selection pressure on the HBV S gene, and that the emergence of mutations in this region would make re-infection after LT possible (10).
The advent of lamivudine, a potent anti-HBV drug, has been a major breakthrough in controlling HBV replication prior to transplantation. Lamivudine is well tolerated even in severely ill cirrhotic patients, and is effective by achieving a negativation of HBV DNA in 90% of patients (11–13). These promising results led to propose LT in HBsAg-positive/HBV DNA-negative patients under lamivudine and to continue lamivudine after transplantation as prophylactic monotherapy without concomitant HBIg administration (14). Despite good initial results with a low rate of HBV recurrence at 1 year, recent data have shown that monotherapy prophylaxis with lamivudine failed in 30–50% of the patients, because of HBV recurrence with the YMDD escape HBV mutant, leading in some cases to a severe clinical outcome (15–17). A combination of lamivudine pre-transplantation and lamivudine and HBIg post-transplantation appeared then to be a better prophylaxis regimen, with a disappearance of HBV DNA prior to transplantation and a HBV recurrence rate at 1–2 years less than 20% (9,18–22). Using such an approach, it has been suggested that lamivudine co-administration could permit the reduction of the overall amount of HBIg given after LT in comparison with a prophylaxis regimen with HBIg alone (22). We report here the first clinical experience concerning the long-term combination prophylaxis of HBV recurrence after liver transplantation demonstrating that HBIg and lamivudine is a safe and very effective alternative. This approach gave excellent results with no recurrence of HBV infection after a follow-up of 2.5 years post-LT. Interestingly, based on PCR methods, we did not find HBV DNA in the serum of all these patients.
Recently, the possibility to stop HBIg and to replace it by lamivudine has been proposed (23). This strategy of HBIg substitution with lamivudine for the long-term prophylaxis has several advantages including convenience for the patients, drug availability and a considerably lower cost. Naoumov and colleagues compared a HBV infection recurrence rate in a group of transplanted patients, with a low risk of HBV re-infection, randomized to receive HBIg or lamivudine after a minimum of 6 months administration of HBIg (24). At 1 year, the HBV re-infection rate was 2 out of 12 (16.7%) and 1 out of 12 (8.3%) in the lamivudine and HBIg groups, respectively. These results confirmed, after only a 1-year follow-up, that this sequential approach is less effective than a combination therapy. An other approach is the conversion from intravenous to intramuscular HBIg in combination with lamivudine, which may be safe, effective, and cost-effective (25). A novel approach could be the association with anti-HBV vaccination. In a recent study, HBIg administration was discontinued, after at least 18 months, in 17 patients and replaced by anti-HBV vaccination, with a median follow-up of 14 months (26). The antibody level was low in most patients and declined with time in the majority of patients, but no HBV recurrence occurred. The role of anti-HBV vaccination in association with lamivudine could be investigated in the future.
In conclusion, our results confirm the major therapeutic improvement made in the prevention of HBV infection recurrence after liver transplantation, using lamivudine. Now, HBV re-infection could probably be avoided by a combination of lamivudine and HBIg prophylaxis.