SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

Background : Long-term effect of YMDD mutations on liver histology in Chinese hepatitis B patients is unknown.

Aim : To examine the effect of prolonged lamivudine treatment on liver histology in Chinese patients with and without YMDD mutations.

Methods : Liver histology was assessed in 85 patients on long-term lamivudine at baseline and year 1, and at year 3 for 25 patients.

Results : Comparing patients with and without YMDD mutations at year 1, the former had higher baseline median necroinflammatory (11 vs. six respectively, P = 0.014) and fibrosis scores (three vs. one respectively, P = 0.001). The proportion of patients with improvement in necroinflammation and worsening of fibrosis was comparable for patients with and without YMDD mutations at year 1 (57.1%, 14.3% vs. 55%, 15% respectively) and year 3 (57.9%, 26.3% vs. 50%, 16.7% respectively). Comparing the histology at year 1 and 3, more patients with YMDD mutations developing after year 1 had worsening of necroinflammation than patients with persistent YMDD wild type (53.8% vs. 25% respectively).

Conclusions : Patients who developed YMDD mutations had higher baseline histological scores. With YMDD mutations, the liver histology became less favourable after 3 years than at the first year, although there was still improvement when compared with that at baseline.


Background

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

Serological markers including reduction of hepatitis B virus (HBV) DNA to an undetectable level, loss of hepatitis B e antigen (HBeAg) with or without development of antibody to HBeAg (anti-HBe, HBeAg seroconversion), and normalization of serum alanine aminotransferase (ALT) level are often taken as short-term end-points of treatment of chronic hepatitis B (CHB).1, 2 Improvement in liver histology is also an important parameter for assessment of different therapies for CHB and is still a required end-point for the purpose of licensing nucleoside analogues.3

Two studies of 1-year treatment of lamivudine in Asian and Caucasian populations have shown that lamivudine treatment is associated with a significant improvement of necroinflammation.4, 5 Suzuki et al. have also shown that improvement of hepatic necroinflammation and fibrosis are observed in 95% and 35% of patients respectively after 1 year of lamivudine treatment.6 Established fibrosis can even be reversed by extending the treatment to 3 years.7 Lamivudine has been shown to reduce levels of circulating immunohistological markers of fibrosis, including α-smooth muscle actin and procollagen III.8

As the prevalence of lamivudine-resistant virus, YMDD (tyrosine-methionine-aspartate-aspartate) mutants increases with the duration of lamivudine treatment,9 it is important to determine whether emergence of YMDD mutations has any adverse effect on liver histology. A study of 13 patients with serial liver biopsies at baseline, year 1 and 3 shows that liver histology worsens from year 1 to 3 in patients with YMDD mutations.10 A more recent study however, shows that liver histology continues to improve for Caucasian patients with or without YMDD mutations, although the former blunts the benefits.7

In the present study, we aimed to examine the effect of prolonged lamivudine treatment on the liver histology in Chinese patients with and without YMDD mutations after 2–3 years of treatment.

Patients

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

Eighty-eight patients were originally recruited from Department of Medicine, The University of Hong Kong, Queen Mary Hospital, Hong Kong of the total of 358 patients in the multicentre trial NUCB30094 and the follow-on trial NUCB3018.10 Eighty-five patients had a second liver biopsy at year 1 and were eligible for recruitment into the present study. The trials were approved by the Ethics Committee of the University of Hong Kong, Hong Kong. In the NUCB3009 trial,4 patients were randomized to receive lamivudine 25 mg daily or lamivudine 100 mg daily or placebo for 1 year (Figure 1). After completion of the first year treatment, patients were enrolled into the NUCB3018 trial in which patients previously receiving lamivudine 100 mg daily were given either lamivudine 100 mg or placebo for 1–2 years followed by open-label lamivudine 100 mg daily for a further 2 years. Patients previously receiving lamivudine 25 mg daily were re-randomized to receive lamivudine 25 mg daily or placebo for 1–2 years followed by open-label lamivudine 100 mg daily. All the patients re-randomized to receive placebo during the second and third year had rebound of HBV-DNA 4 weeks after receiving placebo. These patients were given open-label lamivudine 100 mg daily. All the patients receiving placebo in the NUCB3009 trial, were given lamivudine 100 mg daily in NUCB3018. Patients were followed up every 2 weeks till week 4, every 4 weeks till week 52, every 8 weeks till week 104 and every 16 weeks till week 156.

image

Figure 1. Details of the treatment regime over 3 years of the 85 patients. LAM, lamivudine; OL, open-label.

Download figure to PowerPoint

Liver histology assessment

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

Paired liver biopsies (at baseline and year 1) were performed in 85 patients; biopsies at year 3 were performed in 31 patients (the remaining 54 patients did not agree to a third liver biopsy; a total of 82 patients of the 358 patients from the multicentre trial NUCB3009 and the follow-on trial NUCB3018 had year 3 biopsies). All the liver biopsies were scored according to the criteria of Knodell scoring system11 by a single pathologist who was blinded as to the clinical and laboratory data of patients as well as the timing of the biopsy samples. Briefly, the liver biopsy was scored by the histological activity index (HAI) divided into two major components namely necroinflammation and fibrosis. For necroinflammation, the scores were as follows: piecemeal necrosis, score 0–10; lobular necrosis and inflammation, score 0–4 and portal inflammation, score 0–4. For fibrosis, the scores were as follows: absence of fibrosis, score 0; fibrous portal expansion, score 1; bridging fibrosis, score 3 and cirrhosis, score 4. Improvement or worsening of necroinflammation is defined by changes of at least two points in the scores. Improvement of fibrosis is defined as the changes from score 1 to 0, 3 to 0 or 1, and 4 to 3, 1 or 0. Worsening of fibrosis is defined as changes from 0 to 1, 3 or 4, 1 to 3 or 4 and 3 to 4. Unlike the present study, ranked response, direct comparison between the two biopsies, for necroinflammatory activity and fibrosis, was used for the histological assessment in the original publication of the entire series of 358 patients participating in NUCB3009.4

Determination of HBV-DNA level and YMDD mutation

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

The HBV-DNA levels on presentation were measured by Cobas Amplicor HBV Monitor Test (Roche Diagnostics, Branchburg, NJ, USA) (lower limit of detection 300 copies/mL) from the stored serum kept at −20 °C. The analysis of the YMDD motif was performed on the serial serum at baseline, year 1, year 2 and year 3. HBV mutant and/or wild type at YMDD motif (rt204) were determined by polymerase chain reaction-based methods as described previously by Virology Department, GlaxoSmithKline, Research Triangle Park, NC, USA.12, 13 The lower limit of detection of the assay was approximately 500 copies/mL. The assay is able to discriminate the mutant or wild type if either one is present for >5% of the total viral population.

Statistical analysis

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

Continuous variables with skewed distribution were expressed in median (range). These variables were compared by Mann–Whitney U-test. The median scores of necroinflammation and fibrosis of the paired liver biopsies were compared by Wilcoxon signed ranks test. Categorical data were tested by chi-square test or Fisher exact test. A two-tailed P-value of <0.05 was considered to be statistically significant.

Patient characteristics

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

The details of the treatment regimes over 3 years of the 85 patients are depicted in Figure 1. Thirty-one patients had the third liver biopsies done at year 3. The demographic data of all these patients are listed in Table 1. There were no differences in the demographics between patients with baseline and year 1 biopsies and patients with baseline, year 1 and year 3 biopsies.

Table 1.  Demographic data of the study population on presentation
 Patients with liver biopsies performed at baseline and year 1Patients with liver biopsies performed at baseline, year 1 and year 3
  1. Continuous variables are expressed in median (range).

  2. HBeAg, hepatitis B e antigen; HBV-DNA, hepatitis B virus deoxyribonucleic acid.

Number of patients8531
Male:female60:2522:9
Age (years)31.1 (16.3–47.3)33.0
Albumin (U/L)47 (39–55)49 (42–51)
Bilirubin (μm)10 (4–94)11 (5–27)
Alanine aminotransferase (U/L)59 (14–385)48 (14–298)
HBeAgPositivePositive
HBV-DNA (copies/mL)7.2 × 108 (17 600–5.3 × 1011)1.4 × 109 (9.6 × 105–1.3 × 1011)

Histology assessment at year 1

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

The details of the changes of the necroinflammatory and fibrosis scores over the first year are listed in Table 2. Of the 18 patients receiving placebo, there were no significant changes in the median necroinflammatory and fibrosis scores between baseline and year 1. Among the 67 patients receiving 1 year of lamivudine, there was a significant improvement in the median necroinflammatory score [7 (range: 2–14) to 5 (range: 0–11), P < 0.001] at baseline and year 1 respectively. Patients receiving lamivudine had a significantly higher chance of improvement of necroinflammation compared with those receiving placebo [37 of 67 (55.2%) vs. three of 18 (16.7%) respectively, P = 0.004, odds ratio (OR) 6.2, 95% confidence interval (CI): 1.6–23.3]. The median fibrosis score however, remained the same (1 for both baseline and year 1). There was no significant difference in the frequency of worsening of fibrosis between patients receiving lamivudine and patients receiving placebo [nine of 67 (13.4%) vs. three of 18 (16.7%) respectively, P = 0.71].

Table 2.  Comparison of the histological findings at baseline and year 1 of patients receiving lamivudine or placebo
 Patients receiving lamivudinePatients receiving placebo (n = 18)
Total (n = 67)No YMDD mutations (n = 60)YMDD mutations (n = 7)
  1. *P < 0.001, **P < 0.001, ***P = 0.043, ****P = 0.004.

Change in necroinflammation
 Median score (range)7* (2–14) to 5* (0–11)6** (2–14) to 5** (0–11)11*** (6–14) to 9*** (5–11)7.5 (2–14) to 5.5 (2–14)
 Improvement37 (55.2%)****33 (55%)4 (57.1%)3 (16.7%)****
 No change27 (40.3%)24 (40%)3 (42.9%)9 (50%)
 Worsening3 (4.5%)3 (5%)0 (0%)6 (33.3%)
Change in fibrosis
 Median score (range)1 (0–4) to 1 (0–4)1 (0–3) to 1 (0–3)3 (1–4) to 3 (1–4)1 (0–4) to 1 (1–4)
 Improvement1 (1.5%)1 (1.7%)0 (0%)1 (5.6%)
 No change57 (85.1%)50 (83.3%)6 (85.7%)14 (77.8%)
 Worsening9 (13.4%)9 (15%)1 (14.3%)3 (16.7%)

Among the 67 patients receiving 1-year treatment of lamivudine, seven patients (10.4%) had YMDD mutations at year 1. The details of the changes of the necroinflammatory and fibrosis scores of patients with or without YMDD mutations over the first year are also listed in Table 2. There was significant improvement in the median necroinflammatory score from baseline to year 1 in both patients with and without YMDD mutations [11 (range: 6–14) to 9 (range: 5–11), P = 0.043 for patients with YMDD mutations; 6 (range: 2–14) to 5 (range: 0–11), P < 0.001 for patients without YMDD mutations]. There were no changes in the median fibrosis score for both patients with and without YMDD mutations (median fibrosis score: 3 for patients with YMDD mutations; 1 for patients without YMDD mutation). There was no difference in the frequency of improvement of necroinflammation and worsening of fibrosis between patients with and without YMDD mutations [necroinflammation: four of seven (57.1%) vs. 33 of 60 (55%) respectively, P = 1.0; fibrosis: one of seven (14.3%) vs. nine of 60 (15%) respectively, P = 1.0]. Progression to cirrhosis was not observed in patients with or without YMDD mutation.

Patients who subsequently developed YMDD mutations by the end of 1 year had significantly higher baseline median necroinflammatory and fibrosis scores compared with patients without YMDD mutations [necroinflammation score 11 (range: 6–14) vs. 6 (range: 2–14), P = 0.014; fibrosis score 3 (range: 1–4) vs. 1 (range: 0–3), P = 0.001 respectively]. Of the seven patients who had YMDD mutations, six (85.7%) had a baseline necroinflammatory score >7 and 5 (71.4%) had a baseline fibrosis score >3. Patients with baseline necroinflammatory score >7 had a significantly higher chance of development of YMDD mutations during the first year compared to those with score <7 [six of 30 (20%) vs. one of 37 (2.7%) respectively, P = 0.039, OR 9.0, 95% CI: 1.0–79.5]. Patients with baseline fibrosis score >3 also had a significantly higher chance of development of YMDD mutations compared to those with score <3 [five of 15 (33.3%) vs. two of 52 (3.8%) respectively, P = 0.005, OR 12.5, 95% CI: 2.2–73.7].

Although patients who developed YMDD mutations by the end of 1 year had a higher median ALT level [98 U/L (range: 28–180)] compared to patients without YMDD mutations by the end of 1 year [56 U/L (range: 14–385)], the difference was not statistically significant (P = 0.1). The baseline median HBV DNA level for patients with and without YMDD mutations by the end of 1 year were comparable [3.5 × 108 (range: 3.8 × 107–3.6 × 109) vs. 8.4 × 108 (range: 17 600–1.3 × 1011) copies/mL respectively, P = 0.52].

Histology assessment after 2–3 years of lamivudine treatment

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

Thirty-one patients had liver biopsies at year 3. The serial liver biopsies were assessed and compared in 25 patients who were treated continuously with lamivudine (with either the 25 mg or 100 mg dose) for 2–3 years. These 25 patients included 20 patients on lamivudine for 3 years and five patients on placebo for the first year and lamivudine for the remaining 2 years (details mentioned below). The liver biopsies in the remaining six patients were not eligible to be assessed for the prolonged effect of lamivudine therapy because four patients received lamivudine 100 mg, placebo and lamivudine 100 mg daily for the first, second and third year respectively and two patients received lamivudine 100 mg daily for the first 2 years only, the lamivudine being stopped after HBeAg seroconversion in the third year. Of the 25 patients receiving continuous lamivudine treatment, seven were treated with lamivudine 100 mg for 3 years; 12 with lamivudine 25 mg for the first 2 years and followed by 100 mg for the third year; one with lamivudine 25 mg for 3 years; five with placebo for the first year and followed by lamivudine 100 mg for the remaining 2 years. Only one of these 25 patients had HBeAg seroconversion before the year 3 liver biopsy. The details of the histological findings at baseline and year 3 of the 25 patients are listed in Table 3. The ALT and HBV-DNA levels at year 3 are also listed in Table 3. There were no differences in the median necroinflammatory and fibrosis scores at baseline and year 3. Improvements in necroinflammation and fibrosis were observed in 14 (56%) and six (24%) patients respectively (one had reversion of cirrhosis from fibrosis score 4 to 3). Six patients (24%) had progression of fibrosis (one progressed to cirrhosis from fibrosis score 3 to 4). YMDD mutations were detected in 19 patients (76%) whereas YMDD wild type were detected throughout in six patients (24%). The details of the histological findings in patients with and without YMDD mutations are also listed in Table 3. Although the median necroinflammation score improved from 5 to 3.5 for patients with YMDD wild type and worsened from 6 to 9 for patients with YMDD mutations, the differences were not statistically significant. Patients with and without YMDD mutations had comparable frequency of improvement of necroinflammation [11 of 19 (57.9%) vs. three of six (50%) respectively, P = 1.0] and worsening of fibrosis [five of 19 (26.3%) vs. one of six (16.7%) respectively, P = 0.71]. There were also no differences in the changes of necroinflammation and fibrosis between patients with YMDD mutations of <1 year and of more than 1 year. Among the 19 patients with YMDD mutations, one had regression of cirrhosis with fibrosis score changing from 4 to 3 and one had progression to cirrhosis with fibrosis score changing from 3 to 4.

Table 3.  Comparison of the histological findings at baseline and year 3 and the alanine aminotransferase and the HBV-DNA levels at year 3 of 25 patients (20 on lamivudine for 3 years and five on lamivudine for 2 years) with and without YMDD mutations on prolonged lamivudine treatment
 Total patients (n = 25)Patients without YMDD mutation (n = 6)Patients with YMDD mutations
Total (n = 19)YMDD mutation <1 year (n = 6)YMDD mutation ≥1  year (n = 13)
  1. Values of ALT and HBV-DNA are expressed in median (range).

  2. ALT, alanine aminotransferase; HBV-DNA, hepatitis B virus deoxyribonucleic acid.

Change in necroinflammation
 Median score (range)5 (1–13) to 7 (1–13)5 (4–6) to 3.5 (1–7)6 (3–14) to 9 (2–13)5 (4–10) to 3 (2–13)8 (3–14) to 9 (2–13)
 Improvement14 (56%)3 (50%)11 (57.9%)4 (66.7%)7 (53.8%)
 No change4 (16%)1 (16.7%)3 (15.8%)1 (16.7%)2 (15.4%)
 Worsening7 (28%)2 (33.3)5 (26.3%)1 (16.7%)4 (30.8%)
Change in fibrosis
 Median score (range)1 (0–4) to 1 (0–4)1 (0–1) to 1 (0–1)1 (0–4) to 1 (0–4)1 (0–1) to 1 (0–3)1 (1–4) to 1 (0–4)
 Improvement6 (24%)1 (16.7%)5 (26.3%)1 (16.7%)4 (30.8%)
 No change13 (52%)4 (66.7%)9 (47.4%)3 (50%)6 (46.2%)
 Worsening6 (24%)1 (16.7%)5 (26.3%)2 (33.3%)3 (23.1%)
ALT level (U/L)41 (11–418)31 (17–51)47 (11–481)42.5 (11–112)55 (25–481)
HBV-DNA level (copies/mL)1.1 × 107 (<300–3.3 × 1010)6.2 × 105 (<300–3.4 × 106)7 × 107 (2.5 × 105–3.3 × 1010)3 × 107 (2.5 × 107–2 × 108)9.2 × 107 (5.8 × 105–3.3 × 1010)

Discounting the five patients who received placebo in the first year, 20 patients had continuous lamivudine treatment for 3 years. Of these, four patients had persistent YMDD wild type by the end of year 3. The remaining 16 developed YMDD mutations at various time-points. Of these 16 patients, three patients developed YMDD mutations by the end of year 1 and 13 patients developed YMDD mutations after year 1. The details of the necroinflammatory score and the total HAI score at baseline, year 1 and 3 of each patient are depicted in Figure 2. For the four patients with persistent YMDD wild type, one (25%) had worsening of necroinflammatory and total HAI scores by comparing the baseline and year 3 liver histology. Similarly, for the 16 patients with YMDD mutations, four (25%) had worsening of necroinflammatory and total HAI scores.

image

Figure 2. The details of the necroinflammatory and total histological activity index (HAI) scores for patients with persistent YMDD wild type (diagram on the left) and patients with YMDD mutations (diagram on the right). Thin lines represent YMDD wild type and thick lines represent YMDD mutants developed during the particular years of follow-up.

Download figure to PowerPoint

To examine whether there were differences in the liver histology at year 1 and 3 in patients who developed YMDD mutations after 1 year of lamivudine treatment, the paired liver biopsy at year 1 and 3 of 17 patients of the 20 on continuous lamivudine treatment for 3 years (13 developed YMDD mutations after year 1, four with persistent YMDD wild type) were compared in Table 4 (the histology of the remaining three patients who developed YMDD mutants by the end of year 1 was excluded for this comparison). There was improvement of the median necroinflammation score (from 4.5 to 1.5) and fibrosis score (from 1 to 0.5) in the four patients with persistent YMDD wild type, although the differences were not statistically significant. The median necroinflammatory and fibrosis scores for patients who developed YMDD mutations after 1 year of lamivudine treatment were comparable at year 1 and 3 (median necroinflammatory score 6 and 7, respectively; median fibrosis score 1 and 1, respectively). Comparing patients with and without YMDD mutations, there were no statistical differences in the frequency of improvement of necroinflammation [four of 13 (30.8%) vs. three of four (75%) respectively, P = 0.25]. Conversely, more patients with YMDD mutations had worsening of the necroinflammation compared to patients without YMDD mutation [seven of 13 (53.8%) vs. one of four (25%) respectively, P = 0.58]. The lack of statistical significance was likely to be due to the relatively small number of patients in each category. However, the frequency of worsening of fibrosis between patients with and without YMDD mutations was similar [three of 13 (23.1%) vs. one of four (25%) respectively, P = 1.0].

Table 4.  Comparison of histological findings at year 1 and 3 in patients with persistent YMDD wild type and patients developing YMDD mutations after 1 year of lamivudine treatment
 Patients with persistent YMDD wild type (n = 4)Patients with YMDD mutations after 1  year (n = 13)
Change in necroinflammation
 Median score (range)4.5 (4–5) to 1.5 (1–7)6 (0–9) to 7 (2–13)
 Improvement3 (75%)4 (30.8%)
 No change0 (0%)2 (15.4%)
 Worsening1 (25%)7 (53.8%)
Change in fibrosis
 Median score (range)1 (0–1) to 0.5 (0–1)1 (0–3) to 1 (0–3)
 Improvement2 (50%)2 (15.4%)
 No change1 (25%)8 (61.5%)
 Worsening1 (25%)3 (23.1%)

Discussion

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

In the present study, lamivudine was given in dosages of either 25 mg or 100 mg daily. According to the multicentre 1-year lamivudine trial in Asia, there was no significant difference in the effects in the liver histology between these two doses.4 In addition, among the 13 patients (with three liver biopsies) receiving lamivudine 25 mg initially, all except one were given lamivudine 100 mg after the second year. We have therefore analysed the patients on either doses of lamivudine together. The major limitation of the present study is that, although a total of 85 patients were available for analysis of the first year biopsies, the number became much smaller for the third year biopsies. Statistical analyses of the third year biopsies, and of the biopsies of the small number of patients with YMDD mutations at year 1, are limited by the small sample size. However, it is unlikely that there will be any larger series of patients on lamivudine for 3 consecutive years in the future. Another limitation of the study is that the biopsies were only reviewed by one pathologist. However, the use of a two-point change in the scoring, and the blinding of the pathologist as to the treatment assignments and the sequence of the specimens (as described in the original trial report14), would tend to reduce observer variability.

One novel finding of the present study is the association between more severe baseline necroinflammation and fibrosis, and the chance of YMDD mutations. Further studies are required to confirm this finding, particularly taking into consideration of the effects of other covariates, e.g. ALT and HBV DNA levels. These two factors are known to be associated with YMDD mutations.15 But because of the relatively small number of patients (n = 7) with YMDD mutations by the end of year 1, we were unable to show significant differences in the baseline ALT and HBV DNA levels between patients with and without YMDD mutations. Should it be known that more severe liver histology is an independent factor for YMDD mutations, these patients may have greater clinical benefit if they are treated with other nucleoside analogues with higher anti-viral potency and lower risk of development of viral resistance.

The present study showed that 1-year treatment of lamivudine improved the liver histology with the median necroinflammatory score changing from 7 to 5 (Table 2). Improvement of more than two points of necroinflammatory score was observed in 55.2% lamivudine-treated patients compared with only 16.7% placebo patients (P = 0.004). This improvement of the necroinflammation was observed regardless of the presence or absence of YMDD mutations.

Upon 2–3 years of lamivudine treatment, 19 of 25 (76%) had the YMDD mutations. There was a comparable proportion of patients with improvement of necroinflammation when comparing the baseline and year 3 liver histology between patients with and without YMDD mutations (57.9% and 50% respectively; Table 3). This improvement was consistent with the finding in a previous study by Dienstag et al.7 More importantly, improvement of fibrosis was observed in 24% of patients (26.3 and 16.7% for patients with and without YMDD mutations). Although we did not have year 3 liver biopsies of patients on placebo, this considerable percentage of patients with improvement of fibrosis was likely to be attributable to lamivudine treatment because improvement of fibrosis was observed in only 5.6% of the placebo patients at 1 year (Table 2). Worsening of fibrosis occurred in 24% of patients (26.3 and 16.7% for patients with and without YMDD mutations) by the end of year 3. As only 16.7% patients without YMDD mutations had worsening of fibrosis at year 3 [16.7% of placebo patients already had worsening of fibrosis at year 1 (Table 2)], prolonged lamivudine treatment in patients who did not develop YMDD mutation was associated with a significant retardation of the fibrosis process. This finding was in agreement with a previous study that was conducted in Caucasian population.7 Therefore, the effect of lamivudine treatment on liver histology was similar in Asian and Caucasian patients. There were no differences in the frequency of improvement and worsening of both the necroinflammatory and fibrosis scores between patients harbouring the YMDD mutants for <1 year and for more than 1 year (Table 3). This finding supports that of Suzuki et al. that improvement of liver histology after 3-year treatment of lamivudine is independent of the appearance of YMDD mutants.16

However, when the liver histology was compared between year 1 and 3 in patients with and without YMDD mutations, patients with YMDD mutations developing after year 1 had an increase in median necroinflammatory score and higher chance of worsening of the necroinflammation when compared to patients without YMDD mutation (Table 4). The lack of statistical significance was likely to be related to the relatively small number of patients. Nevertheless, there was no increase in the frequency of worsening of fibrosis in patients with YMDD mutations. As demonstrated in Figure 2, for the majority of patients (10 of 13, 76.9%) with YMDD mutations, the liver histology at year 3 was still better than that at baseline.

The present study suggests that patients who developed YMDD mutations had higher baseline necroinflammatory and fibrosis scores. Patients on prolonged lamivudine treatment had continuous improvement of liver histology for those who did not develop the YMDD mutations. Liver necroinflammation improved after 1-year of lamivudine treatment regardless of the presence or absence of YMDD mutation. With longer duration of infection with YMDD mutations, the liver histology became less favourable when compared with the histology after 1 year of treatment, but still showed improvement when compared with the pre-treatment histology.

Acknowledgements

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References

The authors are indebted to Dr Pui-Chee Wu for scoring the liver histology of all the patients in this study and Dr J. S. Dixon for providing data on the liver histology and the YMDD mutations.

References

  1. Top of page
  2. Summary
  3. Background
  4. Patients and methods
  5. Patients
  6. Liver histology assessment
  7. Determination of HBV-DNA level and YMDD mutation
  8. Statistical analysis
  9. Results
  10. Patient characteristics
  11. Histology assessment at year 1
  12. Histology assessment after 2–3 years of lamivudine treatment
  13. Discussion
  14. Acknowledgements
  15. References
  • 1
    Yuen MF, Lai CL. Debates in hepatitis: how to assess HBV DNA reductions in association with therapy. Viral Hepat Rev 1999; 5: 15975.
  • 2
    Yuen MF, Lai CL. Treatment for chronic hepatitis B. Lancet Infect Dis 2001; 1: 23241.
  • 3
    Yuen MF, Lai CL. Design for clinical trials in Far East. In: Lau, JYN, ed. HBV Protocol. Humana Press, NJ, 2003: 4756.
  • 4
    Lai CL, Chien RN, Leung NWY, et al. A one-year trial of lamivudine for chronic hepatitis B. N Engl J Med 1998; 339: 618.
  • 5
    Dienstag JL, Schiff ER, Wright TL, et al. Lamivudine as initial treatment for chronic hepatitis B in the United States. N Engl J Med 1999; 341: 125663.
  • 6
    Suzuki Y, Kumada H, Ikeda K, et al. Histological changes in liver biopsies after one year of lamivudine treatment in patients with chronic hepatitis B infection. J Hepatol 1999; 30: 7438.
  • 7
    Dienstag JL, Goldin RD, Heathcote J, et al. Histological outcome during long-term lamivudine therapy. Gastroenterology 2003; 124: 10517.
  • 8
    Kweon YO, Goodman ZD, Dienstag JL, et al. Decreasing fibrogenesis: an immunohistochemical study of paired liver biopsies following lamivudine therapy for chronic hepatitis B. J Hepatol 2001; 35: 74955.
  • 9
    Lai CL, Dienstag J, Schiff E, et al. Prevalence and clinical correlates of YMDD variants during lamivudine therapy for patients with chronic hepatitis B. Clin Infect Dis 2003; 36: 68796.
  • 10
    Leung NWY, Lai CL, Chang TT, et al. Extended lamivudine treatment in patients with chronic hepatitis B enhances hepatitis B e antigen seroconversion rates: results after three years of therapy. Hepatology 2001; 33: 152732.
  • 11
    Knodell RG, Ishak KG, Black WC, et al. Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology 1981; 1: 4315.
  • 12
    Allen MI, DesLauriers M, Andrews CW, et al. Identification and biologic characterization of mutations in HBV resistant to lamivudine. Hepatology 1998; 27: 16707.
  • 13
    Allen M, Gauthier J, DesLauriers M, et al. Two sensitive PCR-based methods for the detection of hepatitis B virus variants associated with reduced susceptibility to lamivudine. J Clin Microbiol 1999; 37: 333847.
  • 14
    Lai CL, Chien RN, Leung NWY, et al. A one-year trial of lamivudine for chronic hepatitis B. N Engl J Med 1998; 339: 618.
  • 15
    Yuen MF, Sablon E, Hui CK, Yuan HJ, Decraemer H, Lai CL. Factors predicting hepatitis B virus DNA breakthrough in patients receiving prolonged lamivudine therapy. Hepatology 2001; 34: 78591.
  • 16
    Suzuki Y, Arase Y, Ikeda K, et al. Histological improvements after a three-year lamivudine therapy in patients with chronic hepatitis B in whom YMDD mutants did not or did develop. Intervirology 2003; 46: 16470.