Resistance to lamivudine therapy: is there more than meets the eye?

The goals of treatment of hepatitis B are to prevent progression of the disease or to slow the disease process. Hepatitis B virus (HBV) is a DNA virus which integrates into the host genome. Thus it is difficult to eradicate viraemia. However, it is possible, albeit in a minority, to reduce levels of viraemia to relatively low threshold levels after finite courses of treatment with either interferon alpha or nucleoside analogues, and to lessen the induced necroinflammatory and immune response.¹ Two major forms of active chronic hepatitis B are recognised: wild-type (or hepatitis B e antigen (HBeAg) positive chronic HBV infection) and anti-HBe positive or precore mutant disease. The latter disease is caused by variants of HBV that contain nucleotide substitutions in the core promoter/precore regions of the viral genome.²

Wild-type (HBeAg positive) chronic hepatitis B can be treated with either interferon alpha or nucleoside analogues. Loss of HBeAg and associated suppression of viral replication with pegylated interferon alpha and new nucleoside analogues such as lamivudine, adefovir, tenofovir, telbuvidine, entecavir, and emtricitabine leads to biochemical remission, histological improvement, and in a small percentage, loss of HBsAg.^3–5 Durable responses can occur. Continuous therapy is frequently required for the majority of anti-HBe positive patients with chronic hepatitis B. Thus in most patients, longer term therapy is required to suppress viral replication and thereby slow the disease process.

Therapy with lamivudine alone to suppress viral replication, leading to normalisation of alanine aminotransferase (ALT) levels and improvements in liver histology, unfortunately leads to the emergence of lamivudine resistant mutants in over 50% of treated patients within three years of treatment.^6,7 The immediate clinical impact of resistance is usually relatively minor, although ALT flares and hepatic decompensation can occur.⁸ Earlier short term studies suggested that resistant HBV mutants were less fit (that is, were less replication competent) and consequently less pathogenic. Subsequent studies have indicated that incomplete suppression of HBV replication reduces the degree of benefit. Drug resistant virus favours the selection of increasingly fit and equally pathogenic virus by viral adaptation, and given the complex immune response to hepatitis B, such variants are indeed pathogenic over time.^9–12 Resistance to lamivudine results in lower seroconversion rates in HBeAg positive patients, lower rates of virological and biochemical remission, and less favourable histological change and thus an adverse effect on treatment outcome.¹³

Lamivudine resistance has been mapped to mutations in the tyrosine-methionine-aspartate-aspartate (YMDD) motif of the reverse transcriptase (rt) domain of HBV DNA polymerase.¹⁴ Methionine 204 is mutated to isoleucine or valine (rtM204I/V) in patients with increasing viraemia. Although predictors of resistance have been incompletely defined, immunosuppressant therapy, precore mutants or core promoter variants, duration of therapy, higher baseline ALT, higher baseline HBV DNA, body mass index, and HBV subtype adw have all been implicated.¹⁵ Mutational patterns may also differs between genotypes A and D.¹⁶

Lamivudine treatment have been shown to enhance CD4 and CD8-T cell response to HBV antigens.^17,18 In this issue of Gut, Lin and colleagues¹⁹ suggest that a CD8+ T cell response to lamivudine resistant polymerase epitope influences the response to lamivudine (see page 152). They examined the function and phenotype of specific T cells and demonstrated that functional anti-YMDD cytotoxic T lymphocytes (CTL) correlated with response and outcome. The authors capitalised on the fact that the YMDD motif of the rt domain of HBV DNA polymerase within the nonapeptide YMDDVVLGA (amino acid residues 203–211) in the catalytic site of the HBV DNA polymerase, is an HLA-A2-restricted CTL epitope.²⁰ Thus quantitative measurement of the numbers of peptide specific CTLs is feasible by MHC tetramer-peptide complex staining. The peptides used for analysis included the dominant HLA-A2 restricted peptide from the HBcAg18–27 (FLPSDFFPSV), the wild-type YMDD motif nonapeptide (YMDDVVLGA), the mutant YVDD peptides (YVDDVVLGA), and YIDD peptide (YIDDVVLGA), to construct HLA-A2-peptide tetrameric complexes.

The authors demonstrated that the frequency of YMDD, YIDD, and YVDD motif specific tetramer positive cells within the HLA-A2 CD8 T cell population was increased in “sustained responders” (that is, those with clearance of HBeAg and sustained normalisation of ALT after the end of lamivudine therapy) but not in “non-responders”. The interesting scenario proposed by the authors is that these anti-mutant CTLs may also contribute to clearance of emerging mutant viruses and a successful response to lamivudine treatment. The implication is that treatment responses are improved (and conversely rates of resistance reduced) if CD8+ T cells respond to treatment and proliferate, resulting in a higher level of functional anti-mutant CTL activity during and after therapy.

Some aspects however need to be considered. Firstly, the HBV specific CD8+ T cells were detected in this study only after repetitive in vitro expansion, a technique that detects a small number of precursors. Thus the frequency of polymerase specific CD8+ cells was lower than frequencies usually detected in patients controlling HBV infection.²¹ In addition, the in vivo antiviral efficacy of polymerase specific CD8+ T cells is under debate. The ability of polymerase specific CD8+ cells to control viral replication seems absent in HBV transgenic mice²² and in patients with chronic hepatitis B.²¹ Thus the increased presence of YMDD, YIDD, and YVDD specific CD8+ cells might be merely an association, and not the primary basis of the sustained response to lamivudine treatment. Other CD8+ T cells, not necessarily polymerase specific, could be present in sustained responders to control viral replication.

The ability of YMDD specific CTLs to cross react with YIDD and YVDD mutant epitopes is also not totally unexpected. As the authors point out in their discussion, the amino acid substitution in the YMDD epitope is located at position 204, which corresponds to the anchor position of the epitope and does not affect T cell recognition. Thus demonstration that YMDD specific CTLs can be activated by YIDD and YVDD mutant epitopes shows that the isoleucine and valine mutations at position 204 do not abrogate the binding of the epitope to HLA-A2, and that recognition of lamivudine resistant mutant epitopes does not require induction of a new cross reactive CTL response. However, it must be considered that these mutations might have an impact on the immunogenicity of the YMDD region. It will be important, for example, to directly test whether these YIDD or YVDD lamivudine resistant epitopes bind with higher affinity than YMDD to HLA-A2 molecules. Improved binding of the mutant peptide to HLA-class I molecules would increase the presentation of the lamivudine resistant epitope to specific CTL. This could potentially enhance the ability of polymerase specific CD8+ cells to recognise HBV infected cells in vivo and thus explain the association of polymerase specific CD8 and sustained viral control during lamivudine treatment in HLA-A2+ patients.

Demonstration that polyclonal activation of anti-wild type and CTL specific immune responses to YMDD, YIDD, and YVDD epitopes correlates with treatment outcome is an intriguing and new finding, and may point to an important aspect of the role of T cell responsiveness in lamivudine therapy. If confirmed, these new findings may assist in devising a strategy for treatment of patients and recognition of patterns of genotypic and phenotypic resistance. Lamivudine is a useful treatment for chronic hepatitis B, but its usefulness as a single therapy is limited by the frequency with which resistance occurs. It will be important to ascertain how treatment outcomes can be improved by further insights into immune responsiveness during antiviral therapy.