1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Numerous studies have shown that resistance to long-term lamivudine therapy occurs in as many as ⅔ of hepatitis B virus (HBV) chronic carriers. Additional studies have shown that reversion of HBV mutations in the precore/core promoter region conferring an HBeAg-negative phenotype/genotype can occur in up to 30% of lamivudine-treated patients. In this study, sequences of the HBV polymerase and precore/core coding regions in 26 HBV-infected patients (24 with HBeAg-negative virus infection, 25 genotype D, 1 genotype A) treated for 27 to 53 months with lamivudine were analyzed to determine the relationship between pretreatment HBV DNA sequence patterns and long-term treatment response, and the effect of therapy on the status of HBV precore mutations. Reversions of precore mutations A1762T/G1764A and G1896A were observed in 29% and 25% of patients, respectively, but none became HBeAg-positive. These data are consistent with previously published reversion frequencies for 2 other groups of lamivudine-treated patients. Two naturally-occurring DNA polymorphisms at aa91 and aa256 of the HBV polymerase were observed in the pretreatment serum samples, which correlated with extended treatment failure. In conclusion, reversion of mutations conferring an HBeAg-negative phenotype occur relatively frequently under lamivudine therapy. Furthermore, at least in HBeAg-negative patients infected predominately with HBV genotype D, specific viral DNA sequences which are present before therapy appear to be useful as predictors of long-term response to lamivudine treatment. (HEPATOLOGY 2004;39:64–73.)

Lamivudine monotherapy is currently considered a primary therapeutic option for patients affected with chronic hepatitis B irrespective of hepatitis B e antigen (HBeAg) status.1 Although the initial response to lamivudine therapy in HBeAg-negative patients is favorable with the majority of patients clearing HBV DNA from serum, long-term response rates are aggravated by the reappearance of HBV DNA in serum and increasing serum alanine aminotransfease (ALT) levels during treatment due to the selection of mutations in the HBV polymerase gene conferring drug-resistance.1–5

Patients with chronic hepatitis B are usually HBeAg and HBV DNA positive, although HBV is able to maintain a chronic infection in the absence of detectable HBeAg.6, 7 HBeAg-negative chronic hepatitis B predominates in the Mediterranean area and has an aggressive course with rapid progression to cirrhosis.7 The abolition of HBeAg production is associated with the appearance of mutations in the pre-core/core region. The predominant mutations involve a G-to-A change at nucleotide 1896 (G1896A), which creates a premature stop codon at codon 28, or a double promoter mutation, A1762T/G1764A.8–10

Recent studies have indicated that lamivudine therapy of chronically-infected HBeAg-negative patients is associated with a significant reversion frequency (up to 33%) of precore/core promoter mutations to wild-type and, in some cases, the re-acquisition of circulating HBeAg.11, 12 These reversions did not appear to be correlated with the development of the primary lamivudine-resistance HBV polymerase mutations.

Due to the high frequency of long-term failure rates for lamivudine therapy, there has been a continuing attempt to identify markers that are predictive of response to treatment. Clinical profiles of low initial HBV DNA levels and moderately elevated ALT in serum have been correlated with sustained response and enhanced rates of the loss of circulating HBeAg and anti-HBe sero-conversion.13, 14 Although some treatment studies have demonstrated the application of these profiles for therapy of HBe-negative HBV infection14–16, these parameters often do not apply well to therapy for these individuals.4, 5, 17

In this study, the sequences of the HBV polymerase and precore/core regions in serial serum samples of 26 patients treated with lamivudine were analyzed to determine if specific HBV DNA sequence patterns could be identified that would be linked to the appearance of the predominant HBV drug-resistant mutations in the B and C viral polymerase domains. The study also sought to determine if a significant frequency of reversion of HBV precore/core promoter mutations occurs under lamivudine therapy, if these reversions have a significant clinical impact, and whether such reversions were correlated with the appearance of established lamivudine-resistance HBV polymerase mutations.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References


Twenty-six patients with chronic hepatitis B, 24 men and 2 women [mean age of 47.3 years (range 21–57 years)], were consecutively enrolled at the Department of Gastroenterology, Molinette Hospital (Turin, Italy). All patients had detectable HBV DNA by PCR (HBV Monitor, Roche Diagnostics, Inc., Indianapolis, IN; detection limit of 1000 copies/ml), and elevated serum alanine transaminase (ALT) levels (concentrations > 1.5× upper limit of normal = 35 IU/L), at baseline and for at least 6 months before therapy. All patients had hepatic histological findings of chronic hepatitis or early stage of cirrhosis (class A Child-Pugh score). All patients were negative for hepatitis C, hepatitis D, and human immunodeficiency virus antibodies. None received interferon or other antiviral agents during the study, or within 6 months before the beginning of lamivudine treatment.

Study Design

Patients received lamivudine orally, at a daily dose of 100 mg (mean/median duration of treatment 38 months [range 27–53 months]). An aliquot of serum was collected monthly and stored at −70°C until used for analysis. DNA sequence analyses for the for HBV precore/core promoter region and the HBV polymerase gene were performed on all samples collected at baseline, and at 6 month intervals on all HBV DNA positive samples during treatment. All monthly serum samples were assessed for biochemical markers of liver disease (including ALT), and HBV serologic markers: HBeAg and anti-HBeAg (DiaSorin, Inc., Saluggia (VC), Italy), HBsAg (Abbott Laboratories, Inc.), serum HBV DNA (HBV Monitor).

HBV DNA Amplification for Sequence Analysis

HBV DNA was extracted from 50μl of serum as previously described.18 During the extraction procedure, standard precautions were taken to prevent cross-contamination. Amplification was performed by an in-house nested polymerase chain reaction assay. For the first round of PCR amplification, 5 μl of resuspended DNA was used in a final volume of 35 μl containing 20mM Tris-HCl (pH 8.3), 100 mM KCl, 15 mM MgCl2, 100 mM of deoxynucleotide mix, 0.5 U of Taq polymerase (Roche) and 20 pmol of each primer. The reaction was performed for 35 cycles of 94°C for 1 minute, 53°C for 1 minute for precore/core region or 51°C for the polymerase region, 72°C for 2 minutes, with an extension of 7 minutes at 72°C following the last cycle. For the second round PCR amplification, 1 μl of PCR product from the first reaction was added to 49 μl of the same PCR mixture and was amplified under the same conditions except the annealing temperature was 41°C for precore/core region or 48°C for the polymerase.

Primers used for the amplification of the HBV precore/core region were FHBVpc2 (5′-CTTCGCTTCACCTCTGCACG-3′, bp1584-1604) and BHBVpc8 (5′-ACCACCCACCCAGGTAGCTA-3′, 2123-2104) for the first PCR amplification, and FHBpc1 (5′-TTACATAAGAGGAC TCTTGG-3′, 1650-1669) and BHBpc1 (BHBpc1 5′-AGAAGATCTCGTACTGAAGG-3′, 1992-1973) for the nested PCR. For amplification of the HBV polymerase gene, primers were FHBVs1 (5′-AGAATCCTCACAATACCGCAG-3′, 224-244) and BHBVs10 (5′-AGCAAAACCCAAAAGACCCA-3′, 1020-1001) for the first PCR and FHBp3 (5′-TTTCTAGGGGGAACTACCGT-3′, 275-294) and BHBp3 (5′-GATGTGATCTTGTGGCAATG-3′, 924-905) for the nested PCR.

PCR products were analyzed by gel electrophoresis in 1.5% agarose stained with ethidium bromide. PCR products were then purified by the Quick-Step 2 Purification Kit (Edge BioSystem, Inc., Gaithersburg, MD), according to the manufacturer's instruction, and then directly sequenced (MWG Biotech, Inc., High Point, NC). For all samples, sequencing was performed using 2 to 4 different primers directed to both DNA stands, and sequencing chromatograms were examined for nucleotide heterogeneity at individual positions.

All pretreatment serum samples, and HBV DNA positive samples during treatment, were also analyzed for mutations in the HBV polymerase associated with lamivudine resistance by INNO-LiPA HBV DR analysis (Innogenetics, Inc., Ghent, Belgium) according to the manufacturer's instructions. HBV genotype was determined at baseline and during therapy by comparison of homology between patient HBV sequences and previously published sequences.19


In this study, “responder” patients were classified as attaining both a sustained biochemical and virological response: normalization of ALT levels and the absence of serum HBV DNA by Monitor PCR assay in 2 consecutive determinations during treatment with no further reappearance of HBV DNA. We considered “breakthrough” as those patients in whom HBV DNA reappeared by Monitor PCR assay after an initial response (serum HBV DNA negative by Monitor PCR and normalization of ALT), irrespective of ALT levels. “Non-responder” patients were those who failed to attain normal ALT levels and who failed to clear serum HBV-DNA by Monitor PCR in any determinations during therapy. In no case did we observe a biochemical response to therapy in the absence of a virologic response.

Statistical Analysis

Statistical analyses (Mann-Whitney Rank Sum, 2-tailed t-test, Fisher's exact test) were performed using Minitab (Minitab, Inc., College Park, PA).


The amino acid positions for the HBV polymerase gene (HBpol) are numbered to be consistent with the newly established scheme21 designed to standardize the nomenclature for lamivudine-resistance mutations, pL180M and pM204V/I (originally designated3 as pL528M or pL526M, and pM552V/I or pM550V/I).


  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Virological and Biochemical Response

Before treatment (baseline), all 26 patients were HBV-DNA positive (mean 2.7 × 106 copies/ml, median 1.0 × 106 copies/ml, range .001 – 9.6 × 106 copies/ml), and had elevated ALT levels (mean 219 IU/L, median 173 IU/L, range 62-953 IU/L; upper limit of normal = 35 IU/L) (Tables 1 and 2). Twenty-two of 26 patients (85%) were HBeAg negative and anti-HBeAg positive in sera collected at baseline and during treatment (Table 2). Two patients (#1, #4) were HBeAg/anti-HBeAg negative, and 1 (#23) was HBeAg positive/anti-HBe negative. One patient (#3) was HBeAg/anti-HBe positive at baseline and became anti-HBe negative after 12 months of therapy.

Table 1. Attributes of the 3 Categories of Treatment Responses
OutcomeNo.Initial Serum HBV DNA (log10)*Initial ALT (IU/L)*Duration of Treatment (mo.)*
  • *

    Mean and median values were not significantly different among the outcome groups (Mann-Whitney rank sum).

  • Median initial ALT for the breakthrough group was significantly different from the responder and non-responder groups by Mann-Whitney (P = .047), but was not significantly different by two-sample t-test (P = .29). Mean and median values were not significantly different among the remaining outcome groups (Mann-Whitney rank sum).

  • 9 out of 12 patients experienced breakthrough by 21 months ([mo. 12 {1 pt.}, mo. 18 {5 pts.}, mo 21 {3 pts.}]). The remaining 3 patients experienced breakthrough at months 24, 27, and 38.

All patients266.234.863.00–7.0221917362–953383827–53
Table 2. HBV Core/Precore and Drug-Resistant Polymerase Sequences Before and During Treatment
OutcomePatientSexPretreatmentTreatment (last follow-up)
ALT (IU/L)HBV-DNA (log10/ml)HBeAgAnti-HBePrecore/CorePolymeraseTreatment Duration (months)Precore/CorePolymerase
176217641896L180MM204V/IALT (IU/L)HBV DNA (log10/ml)HBeAgAnti-HBe176217641896L180MM204V/I
  • Although DNA sequence analysis was performed at multiple time points, unless otherwise noted, DNA sequences listed represent the latest bi-annual analysis during therapy (minimum of 24 months). No nucleotide heterogeneity was observed at any of the positions discussed in any of the patient samples. Bold print for patient sequences indicates divergence from the consensus (wild-type) sequence listed at the top of each column.

  • Abbreviations: R, responder; NR, non-responder; BK, breakthrough.

  • *

    Sequence not determined since samples were PCR negative (ud, undetectable, limit of detection 1000 copies/ml).

  • Transient detection of G1896A mutation.

  • Detection of wile-type at 18 months.

  • §

    Transient detection of wild-type.

  • Transient detection of M204I; ND, not detected.

Consensus         CTG         CTGATG
Sequence      AGGTTGATG     AGGTTG 

By week 24 of treatment, HBV DNA became undetectable in 19 patients (73%), and serum ALT normalized in 13 (50%). After a minimum of 27 months of therapy, 7 patients (27%) were responders, 12 (46%) were breakthroughs, and 7 (27%) were non-responders (Table 2). Virologic breakthrough was observed after 12 months of therapy in 6 patients, between 12 and 18 months in 5 patients, and after 38 months in 1 patient (#26). In the 7 non-responder patients, HBV-DNA concentrations and ALT levels were generally lower than baseline (but always detectable) for each patient during the first 6 to 12 months of therapy (data not shown), but then increased, in nearly all cases becoming equal to or greater than baseline levels (Table 2).

Following identification of long-term treatment outcome, an analysis of the pretreatment serum HBV DNA and ALT values revealed that there were no significant differences in pretreatment serum HBV DNA or ALT levels, or the duration of therapy among the three response categories (Table 1).

Analysis of Changes in HBV Polymerase Gene Sequences

At baseline, no patients carried detectable mutations in the HBV polymerase associated with lamivudine resistance (pL180M, pM204V/I) by DNA sequence analysis (Table 2, Fig. 1) and by INNO-LiPA analysis.20, 21 Alignment of HBsAg gene sequences20 in serum samples collected at baseline revealed that 25 of 26 patients were HBV genotype D, and 1 (#4) was genotype A (data not shown). No changes in genotype occurred in these patients during treatment.

thumbnail image

Figure 1. Locations of the major sequences associated with treatment outcome. Panel A depicts domains A to E of the HBV polymerase (amino acids 410 to 592 for genotype D21). Sites are designated with both the new and old nomenclatures for amino acid numbering in this region of the HBV polymerase. Panel B depicts the HBV precore/core region sequenced. The start/stop sites of the overlapping genes and the corresponding location of the HBV epsilon (ε) step-loop RNA structure are shown for reference.

Download figure to PowerPoint

During treatment, 6 patients (4 breakthroughs, 2 non-responders) acquired pL180M (Tables 2 and 3). Five of these 6 patients (4 breakthroughs, 1 non-responder) also developed pM204V, and 1 (#22) acquired pM204I. Five additional patients (3 breakthroughs, 2 non-responders) developed pM204I after 12 months of therapy (Table 2). One breakthrough patient (#24) acquired pM204I after 18 months of treatment and was wild-type at p180L, but after 24 months of therapy, a change to pM204V and pL180M was observed. In this study, pM204V/I and pL180M were not detected in 5 of the 12 patients (40%) experiencing virologic breakthrough during lamivudine therapy (Tables 2 and 3). No sequence heterogeneity was observed at these positions, or at any of the other positions discussed in any of the patient samples.

Table 3. Relationships of Precore/Core and Polymerase Region Mutations
 No. of Patients (%)Number of Patients With
ALT ElevationSustained ALT NormalizationPC/C MutationsYMDD Mutations
  • *

    Sequences from responder patients were not analyzed since samples were HBV DNA negative.

  • PC/C, precore/core promoter; this group includes includes A1762T, G1764A, and G1896A.

  • This group includes M204V and M204I.

Pretreatment (26 pts)     
 A1762T/G1764A only5 (19%)50 0
 A1762T/G1764A + G1896A11 (42%)110 0
 G1896A only5 (19%)50 0
 Wild-type precore/core5 (19%)50 0
 Wild-type p204M26 (100%)26021 
 M204I (YIDD)0000 
 M204V (YVDD)0000 
Treatment (18 pts)*     
 A1762T/G1764A only2 (11%)11 1
 A1762T/G1764A + G1896A6 (33%)41 3
 G1896A only5 (28%)40 4
 Wild-type precore/core5 (28%)30 2
 Wild-type p204M8 (44%)535 
 M204I (YIDD)5 (28%)325 
 M204V (YVDD)5 (28%)413 
Outcome summary (26 pts)     
 Responder7 (27%)70**
 Non-responder7 (27%)7064
 Breakthrough12 (46%)5767

Two naturally-occurring DNA polymorphisms19, 20 HBVpol in pretreatment serum samples were found to be significantly (P < .03 to < .0003) correlated with the failure (breakthrough or non-response) of long-term response to lamivudine therapy (Tables 4 and 5, Fig. 1). One polymorphism was the presence of either a leucine or isoleucine at amino acid 91 of HBpol (p91L or p91I, respectively). P91L was present in 16 of 26 patients (61%) (11/12 breakthroughs, 5/7 non-responders), and was maintained in 15 of these 16 patients during treatment. The remaining 10 patients (1/12 breakthrough, 2/7 non-responder, 7/7 responders) carried p91I. A change from p91I to p91L occurred in 1 breakthrough patient (#13) at month 11 of treatment (the first sample in which the nested PCR reaction could detect HBV DNA during therapy), 1 month before virological rebound (as defined by the Monitor PCR assay) and the emergence of pM204I.

Table 4. Newly Identified HBV Sequences in Pretreatment Serum Potentially Correlating With Treatment Response
OutcomePatientPretreatmentTreatment (last follow up)
1739175219091960-6291 I/L*Q 213S256 S/C1739175219091960-6291 I/L*Q 213S256 S/C
  • Although DNA sequence analysis was performed at multiple time points, unless otherwise noted, DNA sequences discussed represent the latest bi-annual analysis during therapy (minimum of 24 months). No nucleotide heterogeneity was observed at any of the positions discussed in any of the patient samples. Bold print for patient sequences indicates divergence from the consensus (wild-type) sequence listed at the top of each column.

  • Abbreviations: R, responder; NR, non-responder; BK, breakthrough.

  • *

    ATT 91I, CTT 91L.

  • AGT 256S, TGT 256C.

  • Sequence not determined since samples were PCR negative.

  • §

    A1750T instead of A1752C/T.

Consensus     (I) ATT(Q) CAA(S) AGT    (I) ATT(Q) CAA(S) AGT
Table 5. Summary of HBV Sequence Markers Absent in Pretreatment Serum Samples of Sustained Responders
Outcomep91I/Lp256S/CpQ213SPrecore/Core Promoter Region
  • The frequency of 4 sets of HBV sequence markers in pretreatment serum samples not observed in any of the patients exhibiting sustained response to lamivudine (minimum 27 months treatment), in the 3 response categories are listed. “All Tx failures” designates a combination of patients in the non-response and breakthrough groups.

  • *

    All significance listed is versus responder group (Fisher's 2-tailed exact t-test). n.s., not significant (P < .05).

  • Frequency of the following group of mutations: G1739T, A1752C/T, T1909C, T1960G, T1961A/G, C1962A. Patients listed under mutations (mut.) carried virus containing one or more of the 8 mutations listed in this group.

  • HBV in this patient converted from p91I (pretreatment) to p91L at breakthrough which occurred after 12 months of treatment.

Responder770 70 70 70 
All Tx failures19316(<.0003)613(<.006)127(ns.)910(<.03)

The other pretreatment DNA sequence polymorphism in HBpol correlated with long-term treatment response was a cysteine or serine at amino acid 256 of HBpol (p256C or p256S, respectively). p256C was present in 13 of 26 patients (50%) (4/7 non-responders, 9/12 breakthroughs), and was sustained in 11 of these patients during therapy (4 non-responders, 7 breakthroughs). HBV in the remaining 13 patients (3/12 breakthrough, 3/7 non-responder, 7/7 responders) was p256S. All of the 13 patients carrying p256C at baseline also carried p91L. The breakthrough patient (#13), who switched from p91I to p91L during therapy, maintained p256S before and during therapy (Tables 4 and 5) One breakthrough patient (#4), switched from p91L to p91I and from p256C to p256S during therapy, and carried wild-type sequences at p180L and p204M throughout treatment (Tables 4 and 5). During therapy, another breakthrough patient (#5), switched from p256C (observed at baseline) to p256S, maintained p91L (observed at baseline), and acquired pM204I (Tables 4 and 5).

A cluster of previously unreported mutations was detected in HBpol in the pretreatment sera of 7 patients (5 breakthrough, 2 non-responder), producing a glycine to serine substitution (pQ213S) (Tables 4 and 5, Fig. 1). These mutations were maintained during therapy in all 7 patients. In all 7 patients, all three nucleotides in this codon were changed (CAA[RIGHTWARDS ARROW]TCG). Six of these 7 patients also acquired mutations in the YMDD motif during therapy. None of the 7 responder patients carried this cluster of mutations at baseline or during therapy. However, mutations at this codon were not statistically correlated with treatment response (Table 5).

Published reports have described several changes in HBpol other than those discussed above in individual patients exhibiting HBV resistance to lamivudine.24–32 Only a few of these previously described mutations were observed in our patients. All were present at baseline and were maintained during therapy. Q130P (Q478R)25 was present in all 26 patients, S211N (S559N)27 and S219A (S567A)29 were observed in 4 patients, L82M (L430M)2 and L229M/V (L577M/V)24, 29 were present in a non-responder (#11), and Q215H26 was found in 1 breakthrough patient (#13).

Analysis of HBV Precore/Core Sequences

A total of 22 of 26 patients carried HBV with mutations conferring an HBe-negative genotype (Tables 2 and 3). DNA sequence analysis of the precore region in sera collected at baseline showed that the precore stop codon mutation (G1896A)5, 6 was present in 16 of 26 patients (62%) (8 breakthroughs, 4 non-responders, 4 responders) (Tables 2 and 3, Fig. 1). All of these 16 patients were HBeAg negative and anti-HBeAg positive. After a minimum of 27 months of treatment, HBV DNA was detectable in 14 of these 16 patients. Mutations in 4 of these 14 patients (29%) (3 breakthrough, 1 non-responder) reverted to wild-type at these positions (Table 2). Two (#6, #24) of the 10 patients (20%) carrying virus wild-type at this position at baseline, acquired G1896A after 12 months of therapy (Table 2).

DNA sequence analysis of the core promoter region in baseline sera showed that 16 of the 26 patients (61%) (8 breakthroughs, 4 non-responders, 4 responders) carried the double promoter mutation, A1762T/G1764A10, at baseline (Tables 2 and 3, Fig. 1). Eleven of these 16 patients (69%) also carried G1896A at baseline. After a minimum of 27 months of treatment, HBV DNA was detectable in 12 of these 16 patients. Three of these 12 patients (25%) (all breakthrough) reverted to wild-type (Table 2). After 12 months of therapy, 1 non-responder patient (#9) who harbored wild-type virus sequence at nucleotide 1764 at baseline acquired G1764A. In a breakthrough patient (#17), A1762T/G1764A present at baseline were replaced by wild-type sequences after 12 months of treatment, but by 18 months of therapy, these mutations re-emerged in the dominant circulating HBV. Reversion at either G1896A or A1762T/G1764A was not correlated with the acquisition of established mutations in the HBV polymerase B or C domains that confer resistance to lamivudine (Tables 2 and 3) in this study.

Several mutations were identified in the precore/core region that were never observed in any of the 7 responder patients (Tables 4 and 5, Fig. 1). One or more of these mutations were collectively observed in pretreatment sera from 10 patients (5/7 non-responder, 5/12 breakthrough), and in 2 additional breakthrough patients during therapy. When taken collectively, this group of 8 precore/core promoter region mutations was significantly (P < .03) correlated with long-term treatment failure (Table 5).

Two mutations in the HBV precore region recently associated with occult HBV infections (T1802C, T1803G),31 were present in all 26 patients in our cohort in the pretreatment serum samples. A polymorphism of cytosine or thymidine at nucleotide 1733 in the core promoter region of HBV previously described in HBe-negative chronic infections,32 was found in 22 of 26 patients (11 breakthrough, 7 non-responders, 4 responders) in the pretreatment samples. However, none of the patients in our cohort carried either the multi-base substitution or deletion at nucleotides 1753-1764 that were associated with this polymorphism.32


  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In this study, only 7 of 26 patients on lamivudine therapy experienced sustained suppression (>27 months) of HBV with no rebound in viral or disease markers. These data are consistent with previously reported studies and further illustrate the difficulty in treating a population of patients predominantly infected with HBV precore variants using current therapies. Elevated ALT levels and low virus titers have emerged as pretreatment markers of successful prolonged response to lamivudine therapy in chronic carriers infected with HBe-positive HBV and in some studies on HBeAg-negative HBV infection.13, 14–16 However, consistent with previous reports,4, 5, 17 in this study long-term response to lamivudine therapy for patients infected with HBeAg-negative virus was not correlated with pretreatment ALT levels or initial serum HBV titers.

This study has identified several viral sequence patterns that can potentially be used as pretreatment markers to predict the long-term response of a chronic HBV carrier to therapy with lamivudine. The specific DNA sequences that had the highest degree of correlation with the failure of long-term response to lamivudine, p91L/I and p256C/S in domains A and E of HBpol respectively, are especially notable as these were observed in the serum of patients before the initiation of therapy, and represent naturally-occurring polymorphisms.19, 20 These sequences apparently do not need to be maintained during therapy, although they were maintained in the overwhelming majority (17/19) of cases in this study. While in our patients the presence of these sequences in the pretreatment serum were 100% correlated with the failure of long-term drug response, the absence of these markers did not appear to guarantee successful therapy. One patient (#13) who suffered a relapse in HBV DNA level while under lamivudine therapy lacked both HBpol markers (p91L, p256C) in the pretreatment sera, although virus with p91L became the dominant circulating species prior to breakthrough. Additional investigations with other treatment options will be needed to determine if the DNA sequence patterns identified here are specific to lamivudine therapy.

The presence of a leucine at position 91 and a cysteine at position 256 of HBpol may represent compensatory changes that may help to restore replication competence in virus acquiring the drug-resistance mutations pM204V/I and/or pL180M. Some evidence for such compensatory mutations has been reported,25, 28 although neither of the 2 amino acid positions identified in this study have been previously implicated. Positions 91 and 256 are located in the palm region of the HBpol, flanking, in close proximity, the catalytic (containing p204M) and binding (containing p180L) domains.33 It is possible that the local configurations of the catalytic and binding domains that are altered by pL180M and/or pM204V/I and result in a reduction in HBV replication fitness,3, 33 are partially restored by the presence of a leucine at aa91 and/or a cysteine at aa256.

In our study, several mutations in the HBV precore/core region, when taken collectively, also appeared to be associated with long-term treatment failure. However, not all HBV sequence patterns that were absent from the sustained responder group (e.g., pQ213S) were linked to treatment outcome. The apparent requirement for a three-nucleotide change to induce pQ213S in 7 of 7 patients indicates a strong selective pressure at this site, the role of which is currently unknown.

Recent studies have indicated that lamivudine therapy of individuals chronically-infected with HBV precore/core promoter variants (A1762T, G1764A, G1896A) is associated with a significant reversion frequency (up to 33%) of these mutations to wild-type and, less frequently, the re-acquisition of circulating HBeAg.11, 12 Furthermore, these reversions were not correlated with the development of the primary lamivudine-resistance polymerase mutations (pL180M, pM204V/I). In this study, reversion to wild-type sequences at these positions was also observed in a significant fraction of patients (25–29%). Consistent with previous reports, reversions at these positions did not appear to be correlated with the acquisition of drug-resistance mutations in the HBV polymerase B and C domains. Since we were unable to sequence HBV in the serum of the sustained responders, no conclusions with regard to these reversions and long-term treatment outcome can be made in this study. There is little published information regarding the relative frequency of mutant and wild-type HBV at these positions in individual patients. In at least 1 study, approximately 15% of patients analyzed by INNO-LiPATM technology harbored mixed populations.34 Although there was no immediate change in clinical status at the time of the emergence of the precore mutation reversions in the current and previous studies,11, 12 the reappearance of HBeAg-positive virus as the dominant species has the potential to alter the future course of HBV infection in these patients.

In this study, mutations in the YMDD motif of HBpol could not be identified in 5 of 12 patients displaying virus breakthrough during treatment. Several previous studies have reported a similar lack of ability to detect mutations in the YMDD motif in an average of 25% (15–50%) of patients experiencing rebounds in HBV titers while under lamivudine therapy.14–16, 35 In this study, the presence or absence of mutations in the B and C domains of HBpol was confirmed using LiPA analysis, which has a substantially higher level of sensitivity in the detection of targeted mutations in mixed populations than DNA sequence analysis.21, 22 We could not identify any additional mutations in domains A to E (and the intervening sequences) of HBpol that could account for virus breakthrough in the patients without pM204V/I or pL180M.

In our cohort, we did not detect most of the purported drug resistance-associated HBV mutations listed in the published literature other than pL180M and pM204V/I. In our study, there is evidence that S211T (S559T)27 and Q130P (Q478R)25 are not directly associated with drug resistance since, in pretreatment sera, S211T was present in a responder patient and Q130P was present in all 26 individuals. The role of several other previously cited mutations in HBV drug resistance23, 24, 26, 28–31 is still inconclusive since, like pQ213S, these mutations were found only in limited numbers of breakthrough or non-responder patients in our study.

The association of T1802C and T1803G in the HBV precore region with stages of HBV infection other than occult infections31 may require further investigation as these 2 mutations were present in all 26 patients in our cohort, all of whom had active HBV replication and disease. In our cohort, the identification of 1733C in the HBV core promoter region in 22 of 26 patients in the absence of any base substitutions in the HBV core promoter indicates that the polymorphism previously reported32 at this position (1733C,T) and associated with multi-base substitutions or deletions at nucleotides 1753-1764 in HBe-negative chronic infections, may not be as well-defined as previously indicated.

In conclusion, this study has emphasized several important treatment-related virologic features of HBV chronic infections that should continue to be further investigated. Most importantly, it has identified pretreatment HBV sequence patterns that have the potential to serve as predictors of long-term drug response to lamivudine. The ability to target lamivudine therapy to those individuals who are most likely to have long-term benefit, and away from individuals who have little chance of long-term sustained response, could significantly improve the utility and cost-effectiveness of this important anti-HBV agent.


  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The authors would like to thank G. Verme and the members of the Rotary Club Torino Nord Ovest for their constructive support and P. Cote (Georgetown University) for assistance with the statistical analyses. The authors confer special gratitude to E. Hoffman, K. Farrar, F. Wells, K. Gaye and J. Nupp for their technical support.


  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    Lok AS, Heathcote EJ, Hoofnagle JH. Management of hepatitis B: 2000 summary of a workshop. Gastroenterology 2001; 120: 18281853.
  • 2
    Gauthier J, Bourne EJ, Lutz MW, Crowther LM, Dienstag JL, Brown NA, Condreay LD. Quantitation of hepatitis B viremia and emergence of YMDD variants in patients with chronic hepatitis B treated with lamivudine. J Infect Dis 1999; 180: 17571762.
  • 3
    Allen MI, Deslauriers M, Andrews CW, Tipples GA, Walters KA, Tyrrell DL, Brown N, et al. Identification and characterization of mutations in hepatitis B virus resistant to lamivudine. Lamivudine Clinical Investigation Group. HEPATOLOGY 1998; 27: 16701677.
  • 4
    Hadziyannis SJ, Papatheodoridis GV, Dimou E, Laras A, Papaioannou, C. Efficacy of long-term lamivudine monotherapy in patients with hepatitis B e antigen-negative chronic hepatitis B. HPATOLOGY 2000; 32: 847851.
  • 5
    Rizzetto M, Volpes R, Smedile A. Response of pre-core mutant chronic hepatitis B infection to lamivudine. J Med Virol 2000; 61: 398402.
  • 6
    Bonino F, Hoyer B, Nelson J, Engle R, Verme G, Gerin J. Hepatitis B virus DNA in the sera of HBsAg carriers: a marker of active hepatitis B virus replication in the liver. HEPATOLOGY 1981; 1: 386391.
  • 7
    Hadziyannis SJ. Hepatitis B e antigen negative chronic hepatitis B: from clinical recognition to pathogenesis and treatment. Viral Hepatitis Reviews 1995; 1: 736.
  • 8
    Carman WF, Jacyna MR, Hadziyannis S, Karayiannis P, McGarvey MJ, Makris A, Thomas HC. Mutation preventing formation of hepatitis B e antigen in patients with chronic hepatitis B infection. Lancet 1989; 2: 588591.
  • 9
    Lok AS, Akarca U, Greene S. Mutations in the pre-core region of hepatitis B virus serve to enhance the stability of the secondary structure of the pre-genome encapsidation signal. Proc Natl Acad Sci U S A 1994; 91: 40774081.
  • 10
    Fang ZL, Yang J, Ge X, Zhuang H, Gong J, Li R, Ling R, Harrison TJ. Core promoter mutations (A1762T and G1764A) and viral genotype in chronic hepatitis B and hepatocellular carcinoma in Guangxi, China. J Med Virol 2002; 68: 3340.
  • 11
    Lok AS, Hussain M, Cursano C, Margotti M, Gramenzi A, Grazi GL, Jovine E, et al. Evolution of hepatitis B virus polymerase gene mutations in hepatitis B e antigen-negative patients receiving lamivudine therapy. HEPATOLOGY 2000; 32: 11451153.
  • 12
    Cho SW, Hahm KB, Kim JH. Reversion from precore/core promoter mutants to wild-type hepatitis b virus during the course of lamivudine therapy. HEPATOLOGY 2000; 32: 11631169.
  • 13
    Perrillo RP, Lai CL, Liaw YF, Dienstag JL, Schiff ER, Schalm SW, Heathcote EJ, et al. Predictors of HBeAg loss after lamivudine treatment for chronic hepatitis B. HEPATOLOGY 2002; 36: 186194.
  • 14
    Yuen MF, Sablon E, Hui CK, Yuan HJ, Decraemer H, Lai CL. Factors associated with hepatitis B virus DNA breakthrough in patients receiving prolonged lamivudine therapy. HEPATOLOGY 2001; 34: 785791.
  • 15
    Nafa S, Ahmed S, Tavan D, Pichoud C, Berby F, Stuyver L, Johnson M, et al. Early detection of viral resistance by determination of hepatitis B virus polymerase mutations in patients treated by lamivudine for chronic hepatitis B. HEPATOLOGY 2000; 32: 10781088.
  • 16
    Papatheodoridis GV, Dimou E, Laras A, Papadimitropoulus V, Hadziyannis SJ. Course of virologic breakthrough under long-term lamivudine in HBeAg-negative pre-core mutant HBV liver disease. HEPATOLOGY 2002; 36: 219226.
  • 17
    Hadziyannis SJ, Papatheodoridis GV, Vassilopoulos D. Treatment of HBeAg-negative chronic hepatitis B. Semin Liver Dis 2003; 23: 8188.
  • 18
    Smedile A, Rizzetto M, Denniston K, Bonino F, Wells F, Verme G, Consolo F, et al. Type D hepatitis: the clinical significance of hepatitis D virus RNA in serum as detected by a hybridization-based assay. HEPATOLOGY 1986; 6: 12971302.
  • 19
    Norder H, Courouce AM, Magnius LO. Complete genomes, phylogenetic relatedness and structural proteins of six strains of hepatitis B virus, four of which represent two new genotypes. Virology 1994; 198: 489503.
  • 20
    Stuyver LJ, Locarnini SA, Lok A, Richman DD, Carman WF, Dienstag JL, Schinazi RF. Nomenclature for antiviral-resistant human hepatitis B virus mutations in the polymerase region. HEPATOLOGY 2001; 33: 751757.
  • 21
    Lok AS, Zoulim F, Locarnini S, Mangia A, Niro G, Decraemer H, Maertens G, et al. Monitoring drug resistance in chronic hepatitis B virus (HBV)-infected patients during lamivudine therapy: evaluation of performance of INNO-LiPA HBV DR assay. J Clin Microbiol 2002; 40: 37293734.
  • 22
    Stuyver L, Van Geyt C, De Gendt S, Van Reybroeck G, Zoulim F, Leroux-Roels G, Rossau R. Line probe assay for monitoring drug resistance in hepatitis B virus-infected patients during antiviral therapy. J Clin Microbiol 2000; 38: 702707.
  • 23
    Bartholomeusz A, Locarnini SA. Mutations in the hepatitis B virus polymerase gene that are associated with resistance to famciclovir and lamivudine. International Antiviral News 1997; 5: 123124.
  • 24
    Naoumov NV, Chokshi S, Smith H, Williams R. Emergence and characterization of lamivudine resistant hepatitis B variant. HEPATOLOGY 1996; 24: 282A.
  • 25
    Torresi J, Earnest-Silveira L, Civitico G, Walters TE, Lewin SR, Fyfe J et al. Restoration of replication phenotype of lamivudine-resistant hepatitis B virus mutants by compensatory changes in fingers subdomain of the viral polymerase selected as a consequence of mutations in the overlapping S gene. Virology 2002; 299: 8899.
  • 26
    Chayama K, Suzuki Y, Kobayashi M, Kobayashi M, Tsubota A, Hashimoto M, Miyano Y, et al. Emergence and take over of YMDD motif mutant hepatitis B virus during long-term lamivudine therapy and re-takeover by wild-type after cessation of therapy. HEPATOLOGY 1998; 27: 17111716.
  • 27
    Bartholomew MM, Jansen RW, Jeffers LJ, Reddy KR, Johnson LC, Bunzendahl H, Condreay LD, et al. Hepatitis B virus resistance to lamivudine given for recurrent infection after orthotopic liver transplantation. Lancet 1997; 349: 2022.
  • 28
    Fu L, Cheng YC. Role of additional mutations outside the YMDD motif of hepatitis B virus polymerase in L(-)ddC (3TC) resistance. Biochem Pharmacol 1998; 55: 15671572.
  • 29
    Pillay D, Bartholomeusz A, Cane P, Mutimer D, Schinazi RF, Locarnini S. Mutations in the hepatitis B virus DNA polymerase associated with antiviral resistance. International Antiviral News 1998; 6: 167169.
  • 30
    Yeh CT, Chien RN, Chu CM, Liaw YF. Clearance of the original hepatitis B virus YMDD-motif mutants with emergence of distinct lamivudine-resistant mutants during prolonged lamivudine therapy. HEPATOLOGY 2000; 31: 13181326.
  • 31
    Kao JH, Chen PJ, Lai MY, Chen DS. Sequence analysis of pre-S/surface and pre-core/core promoter genes of hepatitis B virus in chronic hepatitis C patients with occult HBV infection. J Med Virol 2002; 68: 216220.
  • 32
    Sallam TA, Tong CY. Two distinct types of hepatitis B virus core promoter variants in Yemeni blood donors. J Med Virol 2002; 68: 328334.
  • 33
    Das K, Xiong X, Yang H, Westland CE, Gibbs CS, Sarafianos SG, Arnold E. Molecular modeling and biochemical characterization reveal the mechanism of hepatitis B virus polymerase resistance to lamivudine (3TC) and emtricitabine (FTC). J Virol 2001; 75: 47714779.
  • 34
    Grandjacques C, Pradat P, Stuyver L, Chevallier M, Chevallier P, Pichoud C, Maisonnas M, et al. Rapid detection of genotypes and mutations in the pre-core promoter and the pre-core region of hepatitis B virus genome: correlation with viral persistence and disease severity. J Hepatol 2000; 33: 430439.
  • 35
    Fontaine H, Thiers V, Chretien Y, Zylberberg H, Poupon RE, Brechot C, Legendre C, et al. HBV genotypic resistance to lamivudine in kidney recipients and hemodialyzed patients. Transplantation 2000; 69: 20902094.