New sensitive tools for hepatitis B virus (HBV) detection in liver transplantation: What will be their impact on the prophylaxis of HBV infection?


  • Didier Samuel,

    Corresponding author
    1. Institut National de la Santé et de la Recherche Médicale (INSERM), U785, Villejuif, France
    2. Université Paris-Sud, UMR-S785, Villejuif, France
    3. Assistance Publique–Hôpitaux de Paris (AP-HP) Hôpital Paul Brousse, Centre Hépato-Biliaire, Villejuif, France
    • Centre Hépatobiliaire, INSERM U 785, Hôopital Paul Brousse, 12 Avenue Paul Vaillant Couturier, 94800 Villejuif, France
    Search for more papers by this author
    • Telephone: 33 1 45 59 34 03; FAX: 33 1 45 59 38 57

  • Anne Marie Roque-Afonso

    1. Institut National de la Santé et de la Recherche Médicale (INSERM), U785, Villejuif, France
    2. Université Paris-Sud, UMR-S785, Villejuif, France
    3. Assistance Publique–Hôpitaux de Paris (AP-HP) Hôpital Paul Brousse, Service de Virologie, Villejuif, France
    Search for more papers by this author

  • See Article on Page 1137

The spontaneous risk for hepatitis B virus (HBV) reinfection after transplantation is around 80%.1 The first therapeutic breakthrough was the long-term administration of high doses of anti-HB surface antigen immunoglobulins (HBIG) that reduce drastically the rate of HBV recurrence.2, 3 In a recent reported series of 284 patients, the overall 10-yr HBV recurrence rate was 27%. Recurrence rate was higher in patients with cirrhosis (50%) than in those with fulminant hepatitis (0%) or with viral B-Delta cirrhosis (15%).4 Moreover, it was significantly higher in patients with viral B cirrhosis who were serum HBV–deoxyribonucleic acid (DNA)–positive before transplantation than in those who were HBV-DNA negative (80% vs. 27%). The advent of nucleos(t)ide analogs was the second breakthrough.5 Lamivudine monotherapy to prevent recurrence results in high rates of HBV reactivation reaching 40% at 3-4 yr posttransplantation. HBV reactivation is more frequent in HBV DNA-positive patients at transplantation,6, 7 due to expected emergence of resistance mutations in the YMDD locus of the viral polymerase. Results of adefovir or entecavir monotherapy posttransplantation are not yet available, but it may be expected that the development of adefovir mutations will similarly become a problem. The potential benefits of nucleoside analogs in combination therapy should also be investigated. The third breakthrough was the combination of HBIG with nucleos(t)ide analog, which is able to reduce the recurrence risk to less than 10%, even in patients with high risk of HBV recurrence.8–10 As a result of these therapeutical advances, HBV patients receiving adequate immunoprophylaxis have a similar or better 10-yr survival rate than patients receiving transplantation for other liver diseases.11 These excellent results reflect the additive effects of nucleoside analogs and HBIG. To reduce the high cost of combination therapy, it has then been suggested that the dose of HBIG could be reduced, and that intramuscular HBIG administration could safely replace intravenous HBIG. There are also ongoing studies on the possibility of HBIG discontinuation after several weeks or months, with introduction of HBV vaccine, or maintenance of nucleos(t)ide analogs. The duration of HBIG therapy remains a controversial issue that requires further investigation.


HB, hepatitis B; HBV, hepatitis B virus; HBIG, anti-HB surface antigen immunoglobulins; DNA, deoxyribonucleic acid; cccDNA, covalently closed circular DNA.

Most studies on HBV recurrence posttransplantation have been performed using non-polymerase chain reaction HBV DNA testing. Thus it is important to reassess the risk of HBV recurrence in the current therapeutic era with new tools for HBV detection. Current real-time polymerase chain reaction assays for HBV DNA achieve cutoff detection limits of 12 IU/mL. Until recently, non-polymerase chain reaction methods were unable to detect HBV DNA level below 105-106 copies/mL. Thus, the strongest predictive factor of HBV recurrence was the presence of a viremia above 105-106 copies/mL at transplantation. In the current period, most patients on the waiting list receive nucleos(t)ide analogs to decrease HBV DNA levels. It is not known if the best predictive factor is the level of serum HBV DNA before antiviral treatment or at time of liver transplantation. In other words, do patients with a spontaneously undetectable HBV DNA at time of transplantation have an identical risk of HBV recurrence as patients on antiviral treatment?

The outcome of HBV infection is the result of complicated viral-host interactions that involve different actors, including the host antiviral responses, viral mechanisms, and the liver microenvironment. HBV is the prototype of the hepadnavirus family, sharing a distinctive strategy for replication. The critical step for genome amplification that determines the rate of HBV replication is the transcription of the 1.1 genome-sized pregenome ribonucleic acid (RNA) from the covalently closed circular DNA (cccDNA) in the nucleus of the hepatocyte. In the cytoplasm of infected cells, within viral capsids, this replication intermediate is converted without amplification into an open circular duplex DNA by the virally-encoded polymerase that has several functions: reverse transcriptase, ribonuclease, and DNA-dependent DNA polymerase. A continued productive HBV infection requires a persistent population of transcriptionally active cccDNA molecules to ensure a stable source of pregenome RNA for replication and the templates for messenger RNA synthesis and production of viral proteins.12, 13 Nuclear cccDNA molecules are organized into a chromatin-like structure consisting of both histone and nonhistone proteins, including the virally encoded core protein, the cellular acetyltransferases PCAF and CBP and the histone deacetylase 1. cccDNA transcription is regulated by the acetylation status of the cccDNA-bound H3/H4 histones. HBV cccDNA has proven to be relatively insensitive to the therapeutic regimens used to suppress HBV replication in chronically infected patients. Recent studies have shown that cccDNA does decline after 48 weeks of treatment with adefovir dipivoxil or lamivudine plus pegylated interferon (from 1 to 2 log) but it can be inferred, based upon a mathematical modeling, that it would take more than 14 yr to completely clear a HBV-chronically infected human liver of intracellular cccDNA.14 The persistence of HBV cccDNA in permissive cells is the basis for HBV reactivation after discontinuation of nucleoside analogs, or for HBV recurrence after transplantation if prophylaxis is stopped. Indeed, cccDNA has been shown to be present in patients at long term after liver transplantation.4 During treatment, with ongoing residual replication, the error prone reverse transcription step results in the emergence of resistant strains that can infect other cells or that directly amplify the pool of cccDNA with templates harboring mutations in both HB surface antigen and polymerase open reading frames.

Current HBV DNA quantification assays are standardized and highly sensitive. Assays for cccDNA detection and/ or quantification in liver tissue are in-house assays and their respective sensitivity and specificity are difficult to compare. However, these assays are new tools to assess the predictive value of viral load and cccDNA on the risk of HBV reinfection after transplantation and to investigate the possibility to modify or decrease HBV prophylaxis regimens.

Recently, several authors have looked at the relationship between serum HBV DNA levels measured by polymerase chain reaction at transplantation and HBV reinfection. In the work by Marzano et al.15 there was a direct relationship between HBV DNA level at transplantation and HBV reinfection. This relationship was stronger in patients receiving lamivudine monotherapy as posttransplantation prophylaxis than in those receiving combined prophylaxis with lamivudine and HBIG. A viral load above 105 copies/mL was associated with a higher risk of HBV recurrence. In this issue of the Journal, Hussein et al.16 showed very elegantly that total HBV DNA was found by polymerase chain reaction in explant livers of 17 out of 18 patients including 7 patients treated with nucleoside analogs who had undetectable serum HBV DNA. In addition, cccDNA was found in the liver of 16 patients at transplantation. This result confirms that despite effective antiviral therapy, total HBV DNA and cccDNA remains positive in the liver of most patients. This is not surprising since as discussed above, the decrease of HBV DNA and of cccDNA liver content in patients treated with nucleos(t)ide analogs is slow and it would take at least 10 yr to clear the virus from the liver.14 The duration of antiviral therapy prior to transplantation was much shorter in this series, and it would have been surprising to not detect HBV DNA and cccDNA in the liver. In this series, there was no correlation between the level of hepatic HBV replication measured by HBV DNA and cccDNA in the liver and the rate of HBV reinfection posttransplantation. Indeed, as shown in previous series, the best predictor of HBV recurrence was the level of HBV DNA in serum at transplantation.2 The removal of the major viral reservoir and the neutralization of circulating virions by HBIG administration at the anhepatic phase modify the natural history of posttransplantation HBV infection. The efficacy of HBIG neutralization at the anhepatic phase certainly depends on viral burden, explaining why serum HBV DNA remained the best predictive factor of recurrence. Nonneutralized viral particles can then infect the graft. This is the second important point of this study: there is a high rate of HBV reinfection after transplantation. HBV DNA detection in the graft liver has been previously demonstrated in patients without apparent HBV recurrence.4, 9 However, this series showed for the first time the correlation between cccDNA and HBV DNA in the graft and the longitudinal evolution of these markers in posttransplantation liver graft biopsies at least during the first 3 posttransplantation years. At 18 months posttransplantation, 70% and 17% of the patients had HBV DNA and cccDNA detectable in their liver graft, respectively. The longitudinal analysis of the posttransplantation liver biopsies showed the persistence of HBV DNA in the majority of liver biopsies while cccDNA detection rate was declining at medium term. These results looked similar to those of Roche et al.,4 where at 10-yr posttransplantation with ongoing immunoprophylaxis, 44% of the patients had HBV DNA in their liver while cccDNA was detected in only 10%. The progressive decline of cccDNA detection in the graft of patients without recurrence is encouraging. If confirmed, it could resemble the situation of nontransplantation patients treated with nucleoside analogs where HBV clearance can be expected after several years of viral replication suppression. The presence of total HBV DNA detection in the absence of cccDNA raises the question of the sensitivity of the cccDNA assay and the question of HBV DNA integration. If the total HBV DNA detected does not correspond to integrated DNA, low amounts of cccDNA certainly persist to allow HBV replication. In this case HBV replication can be reactivated if HBV prophylaxis is stopped. This issue is fundamental in the decision to reduce or discontinue the prophylaxis against HBV. To answer this question an improvement of the current tests and/or the implementation of assays for pregenomic RNA and replication intermediates detection in the liver are required. In addition, it has been shown by others that HBV DNA can be detected at long term also in extrahepatic sites such as peripheral blood mononuclear cells; their implication in late HBV recurrence is still unclear.4, 17 It would be important in the future to know the longitudinal evolution of the amount of total HBV DNA in the liver graft and to determine if an increase of HBV DNA in the graft, or the reemergence of cccDNA will predict a late HBV recurrence.

HBV DNA and cccDNA detection in the liver are promising new tools in the management of HBV infection in this new era of the large use of potent antiviral drugs. Shortcomings of these new tools are the difficulty of performing frequent posttransplantation liver biopsies, the low number of patients with HBV recurrence, and assay standardization. In addition, the number of patients studied in different series is too small to draw any definite conclusion on their place in the management of HBV patients in and out the context of transplantation. In conclusion, these new tools still provide limited information due to the absence of technique standardization. However, at a time where physicians are potentially overconfident about the low risk of HBV recurrence after transplantation, these tools are extremely important and clearly show that most patients will require long-term prophylaxis. Indeed, the temptation to use only nucleoside analogs without combination with HBIG is high within many teams. These teams should keep in mind that on nucleoside analogs only, the risk of HBV reactivation will increase with the time posttransplantation (in contrast to what was observed in the combination prophylaxis HBIG + nucleoside analogs) due to escape mutations on one part and from noncompliance on the other part. In the future, HBV DNA testing in liver tissue will help to determine those patients in whom prophylaxis can be stopped or reduced. We must also keep in mind that HBV can persist indefinitely despite apparent full recovery and immunization. This has been shown in anti-HB surface antigen–positive and anti-HB core antigen–positive patients who reactivate on immunosuppression,18 and in liver grafts from anti-HB core antigen–positive, which can transmit HBV infection to the recipient.19, 20 These new tools will require standardization and better sensitivity but will probably never guarantee the absence of HBV reactivation.21 The previously published sentence is still valid: In liver transplantation for HBV infection “the situation seems under control but the virus is still there.”22