Current limitations to nucleoside analogue therapy for chronic hepatitis B virus infection in the liver transplant and non-transplant settings


Correspondence: ProfessorGMcCaughan AW Morrow Gastroenterology and Liver Centre, Royal Prince Alfred Hospital, Camperdown NSW 2050, Australia. Email: g.mccaughan@centenary.

See article on page 304

It is now recognized that nucleoside analogue therapy is proving extremely effective for the treatment of chronic hepatitis B virus (HBV) infection. 1–3 The agents currently under study include famciclovir, lamivudine and adefovir diproxil. In particular, lamivudine has been shown, in randomized, controlled trials to dramatically decrease serum HBV-DNA levels, normalize alanine aminotransferase (ALT), improve hepatic histology and induce hepatitis B early antigen (HBeAg) to antibody to HBeAg (anti-HBeAg) seroconversion. 4

Such effects have proven to be potentially lifesaving in the liver transplant setting. However, they still need to be balanced against current strategies. Early experience of liver transplantation in patients with HBV-associated cirrhosis was of frequent allograft reinfection leading to fibrosing cholestatic hepatitis (FCH), graft failure and early death. 5,6 Reinfection was observed more frequently in patients with high levels of circulating virus in the pretransplant state. 7,8 Also, the selection of precore mutant strains from an actively replicating and genetically diverse HBV pool has been associated with a poor outcome in one, but not all, studies. 9,10 Consequently, HBV infection became a relative contraindictation to transplantation in many units.

As a result of these problems, strategies have been developed to limit HBV reinfection. High-dose hepatitis B immunoglobulin infusion (HBIG) commencing in the anhepatic phase and continuing indefinitely was shown to reduce the incidence of reinfection. 7,8 However, HBIG is in scarce supply, extremely expensive, may be associated with mercury toxicity 11 and has limited efficacy in HBeAg-positive, HBV-DNA-positive patients. 7 New HBIG preparations with a prolonged half-life have since been formulated; this may reduce costs and the frequency of repeat injections. 12

In contrast to this approach, nucleoside analogues, particularly lamivudine, have been used to prevent HBV infection, as well as to treat active hepatitis B after liver transplantation. In order to prevent recurrent HBV, lamivudine was commenced at least 1 month before liver transplantation and was continued (without HBIG) afterwards. Such approaches have reduced HBV reinfection from historical levels of 80–90% to 20%. 13 The further addition of HBIG, administered either at high or very low doses, has completely prevented HBV reinfection of the allograft. 14–16 These are important new data as low-dose HBIG is less expensive and, given monthly as an intramuscular injection, is a more practical approach than current intravenous regimes. Titres of antibodies to hepatitis B surface antigen (anti-HBs) of 50–100 U/L are achieved by such an approach, and levels seem reasonably stable over time. Patients treated with combination lamivudine and HBIG remain negative for HBV-DNA in serum as determined by polymerase chain reaction (PCR). 15

Lamivudine therapy has also been used for post liver transplant patients already infected with HBV. In one multicentre study, a HBeAg seroconversion rate of 31% was reported, and 6% of patients also lost hepatitis B surface antigen (HBsAg). 17 However, YMDD variants of HBV associated with drug resistance emerged in 27% of patients. We have found that lamivudine can control severe recurrent hepatitis B infection, but the emergence of YMDD variants has occurred in 60% of patients treated for longer than 6 months. 15,16 Such breakthroughs were associated with death or liver failure in up to half of these patients. 15,16

Thus, the major initial limitation to the use of lamivudine therapy ( Table 1) is the emergence of drug-resistant variants. 18–20 These have been classified as type I when the variant has amino acid changes at position 526 (leucine to methionine) and position 552 (methionine to valine). The type II mutation involves a change of methionine at position 552 to isoleucine. This latter single change, in itself, is enough to result in drug resistance. The emergence of these variants is time dependent. 21

Table 1.  Current limitations to nucleoside analogue therapy for hepatitis B virus (HBV)
LimitationPossible solution
  1. ALT, alanine aminotransferase; HBeAg, hepatitis B early antigen; anti-HBeAg, antibody to hepatitis B early antigen; ccc DNA, covalently closed circular DNA.

Emergence of drug resistantLimit duration of therapy
variants (YMDD)New agents (e.g. adefovir)
Difficulty predicting HBeAg toHigh levels of ALT
anti-HBeAg seroconversionpretreatment
 Low levels of HBV-DNA
 Examine T-cell responses
 to HBV before therapy
Uncertain duration of therapyTreat until HBeAg to
Lack of effect on ccc DNA? Combination therapy
Cell-specific (bile duct cell)? Combination therapy
resistance? Other agents

Studies in the non-transplant setting have demonstrated HBeAg seroconversion, and loss of HBsAg in a small proportion of treated patients. 4 A HBeAg seroconversion or loss of HBsAg maybe the result of restoring an efficient antiviral T-cell response, and this has recently been demonstrated to accompany the reduction in viraemia induced by antiviral therapy. 22 Also, emerging data suggest that HBeAg seroconversion on lamivudine may be predicted by a high level of ALT and low level of HBV-DNA pretreatment. 4 An important feature of HBeAg seroconversion is that it increases with time. 21 Thus, there are potentially two conflicting time-related outcomes: seroconversion versus resistance. It is not yet clear what the duration of therapy should be, and whether or when antiviral therapy can or should be stopped.

A further limitation to nucleoside analogue therapy is the resistance of the HBV-covalently closed circular DNA (ccc DNA) to single agent therapy. However, even this problem is under study. Combination therapy with famciclovir and lamivudine in duck hepatitis B virus (DHBV) infection has been shown to significantly inhibit formation or replenishment of ccc DNA compared with single agents. 23 However, in humans, this benefit maybe lost as famciclovir, when used in combination with lamivudine, accelerates the emergence of YMDD-resistant variants. 24

In a manuscript published in this issue of the Journal, Nicoll et al. have now added another potential limitation to therapy. 25 These authors have provided evidence that resistance to antiviral therapy may be cell specific. In a series of experiments, they used the duck hepatitis model, an invaluable tool for studying the efficacy and mode of action of antiviral agents, 1 to address this question. Duck HBV infection within bile duct epithelial cells (BDEC) was shown to be unaffected by the nucleoside analogue famciclovir (penciclovir), a finding that contrasts strikingly with the clear cut inhibition of viral replication within hepatocytes. This lack of efficacy was even more obvious under conditions of biliary ductule proliferation induced by bile duct ligation. As cholangiolar proliferation is a prominent feature of FCH, it was postulated that these cells may be a source of persistent HBV in that condition.

Although these experiments are reasonably convincing, there are some potential criticisms. First, the authors have not formally shown DHBV replication in these cells, although the increase in viral load in the bile duct-ligated ducks that correlated with increased expression in BDEC is suggestive of it. Second, it is unclear whether the cells are resistant to famciclovir alone or to other nucleoside analogues, such as lamivudine. Third, the mechanism of resistance has not been demonstrated. As the authors point out, this resistance may be due to either failure of BDEC to phosphorylate the nucleoside analogue to its active form, or to lack of active drug in the bile. Finally, the extrapolation of findings in the avian model to the clinical arena should be made with great reservation because congenitally infected ducklings exhibit immune tolerance to DHBV and do not develop hepatitis as assessed histologically or by serum ALT. 26 It is also not clear whether, in humans, HBV persists within the biliary epithelium and proliferating bile ductules.

It can be concluded that nucleoside analogues are a major breakthrough in the therapy of hepatitis B. They have been shown to improve outcomes in chronic hepatitis B and they reduce or possibly abolish HBV reinfection after liver transplantation, particularly when used in combination with HBIG. Despite these advances, limitations still exist because of drug resistance, lack of proven combination therapies and the possible existence of specific sites of cellular resistance. These limitations challenge us to intensify the search for the ‘holy grail’ of viral eradication for patients with chronic HBV infection.