Clinical trial: effects of adefovir dipivoxil therapy in Asian and Caucasian patients with chronic hepatitis B


Dr S. G. Lim, Department of Medicine, Division of Gastroenterology, National University Hospital, 5 Lower Kent Ridge Road, Singapore 119074, Singapore.


Background  Two-thirds of the 350 million people infected with chronic hepatitis B virus live in the Asia-Pacific region.

Aim  To compare the effects of adefovir dipivoxil therapy between Asian and Caucasian patients with chronic hepatitis B.

Methods  The safety and efficacy of 10 mg of adefovir dipivoxil was compared to placebo in 501 Asian (n = 259) or Caucasian (n = 242) HBeAg+ and HBeAg− chronic hepatitis B virus patients treated for 48 weeks in two randomized, double-blind, placebo-controlled studies.

Results  At week 48, histological improvement was observed in 60% and 56% of Caucasian and Asian patients, respectively. Change in serum hepatitis B virus DNA from baseline to week 48 for the adefovir dipivoxil-treated patients was −3.89 and −3.70 log10 copies/mL in Caucasian and Asian patients, respectively, while 34 per cent of Caucasian patients and 39 per cent of Asian patients had undetectable serum hepatitis B virus DNA (<400 copies/mL) at week 48. The percentage of patients achieving alanine aminotransferase (ALT) normalization at week 48 was similar in both groups (Caucasian 64 per cent, Asian 63 per cent). No patients developed resistance through week 48. No differences in adverse events or grade 3 or 4 laboratory abnormalities were observed between groups.

Conclusions  There were no significant differences in treatment response between Asians and Caucasians. Adefovir dipivoxil was well tolerated and no resistance developed up to week 48 in both racial groups.


More than 350 million people worldwide are chronically infected with the hepatitis B virus (HBV).1 Most patients with chronic hepatitis B in the world are Chinese, and China remains a country in which HBV endemicity is high (prevalence of 7–20%).2 The prevalence of chronic hepatitis B in South-East Asia varies from 1% to >10%, with most infections occurring in the neonatal period or during early childhood. By contrast, prevalence is low (0.2–0.5%) in North America; northern, western and central Europe; and Australia, where most infections are transmitted during adolescence or adulthood through sexual contact or intravenous drug use.2 The natural history of HBV infection differs between Asian and Caucasian patients,3, 4 with Asians more likely to acquire infection perinatally, while Caucasians may acquire it in adulthood.

Although HBV vaccination has been shown to be effective in decreasing the rate of chronic hepatitis B infection, the large number of HBsAg carriers remain at increased risk of dying from hepatocellular carcinoma and liver disease, in particular those with high levels of HBV DNA.5 Therefore, effective treatment of chronic HBV infection to reduce morbidity and mortality from liver cirrhosis and hepatocellular carcinoma is an urgent medical need worldwide, especially in Asian countries.

Until the late 1990s, the only generally approved treatment for chronic hepatitis B was alpha interferon (IFN-α). However, treatment with IFN-α frequently causes undesirable side effects, results in long-term response in only a small proportion of patients, and must be given by injection. Early trials showed that IFN-α therapy is least effective in Asian patients,6 except for those with elevated liver enzyme levels. The poor response is thought to be due to immune tolerance to HBV after infection at birth or in early childhood.7

Lamivudine (Epivir-HBV), a nucleoside analogue, has been shown to suppress HBV replication, improve liver necrosis and inflammation, reduce hepatic fibrosis, and enhance rates of HBeAg loss and seroconversion compared to placebo after 1 year of treatment irrespective of ethnic origin or HBeAg+ or HBeAg− disease status at baseline.8–11 With prolonged dosing, lamivudine monotherapy is associated with the emergence of resistant HBV with mutations in the tyrosine-methionine-aspartate-aspartate (YMDD) motif at catalytic domain (C domain) of the viral DNA polymerase. The incidence of YMDD mutations rises from 15-32 per cent in the first year to 67–69 per cent by the fourth year of treatment10, 11 and is associated with viral and biochemical breakthrough. The benefit of long-term therapy must therefore be weighed carefully against the concern about YMDD mutations and the durability of therapeutic response.12

Adefovir dipivoxil (ADV) (Hepsera, Gilead Sciences, Foster City, CA, USA) is a pro-drug of adefovir, a nucleotide analogue of adenosine monophosphate that has demonstrated in vivo and in vitro activity against both lamivudine-resistant HBV and wild-type HBV.13, 14 Large placebo-controlled clinical trials have demonstrated potent activity against HBV in both HBeAg+ and HBeAg− nucleoside-naïve patients, including significant biochemical, virological and histological improvement as well as 0% resistance after 48 weeks of therapy.15, 16 In an ongoing randomized, double-blind, placebo-controlled, five-year study involving 480 HBeAg+ Chinese subjects (15–22% of whom were shown to be infected with YMDD-mutant HBV at baseline), ADV was shown to be safe and effective after 1 year of therapy without any evidence of genotypic resistance to adefovir.17 ADV has also been shown to be particularly effective against lamivudine-resistant HBV when added to lamivudine as soon as genotypic resistance is detected instead of delaying therapy until marked increases in serum HBV DNA and alanine aminotransferase (ALT) are apparent.18 Moreover, in an open-label, multi-center, international study in pre- and post-liver transplantation patients with recurrent chronic hepatitis B and evidence of lamivudine-resistant HBV, ADV administration resulted in a significant decrease in viral load, improvement in Child-Pugh-Turcotte score, and high rates of survival in both cohorts, with no detectable adefovir resistance after 48 weeks of therapy.19, 20

In this analysis, we compared the treatment response between Asians and Caucasians with respect to histological, virological, and biochemical parameters after 48 weeks of ADV 10 mg therapy; and within each ethnic group, response rates between ADV 10 mg and placebo were compared.



Six hundred ninety-five chronic hepatitis B virus patients were treated for 48 weeks in two randomized, double-blind, placebo-controlled studies.

Study GS-98-437 was a phase 3, randomized, double-blind, multicenter, placebo-controlled study to evaluate the safety and efficacy of ADV 10 mg once daily in patients with HBeAg positive chronic HBV infection.15 A secondary efficacy objective was to also evaluate ADV 30 mg once daily. Five hundred fifteen patients (HBV DNA ≥ 106 copies/mL, Roche Amplicor polymerase-chain-reaction (PCR (Roche Diagnostics, Branchburg, NJ, USA)), LLQ 400 copies/mL, ALT 1.2–10 x upper limit of normal (ULN)) were randomly assigned in a 1:1:1 ratio to ADV 30 mg daily, ADV 10 mg daily, or placebo for the first 48 weeks of study; 511 patients received at least one dose of study drug (167, 171, and 173 patients in the placebo, ADV 10 mg, and ADV 30 mg groups, respectively). During the second 48 weeks of the study, patients originally randomized to ADV 10 mg were re-randomized to receive either ADV 10 mg or placebo once daily; patients who had received ADV 30 mg received placebo; and patients who had received placebo received ADV 10 mg once daily. This study was initially planned to be a 2-year study of double-blinded treatment. However, during year two it was discovered that drug had been misallocated to a majority of patients due to a contractor error, but there were no such errors in year 1 of the study. Because of this misallocation, the objectives for the second 48 weeks could not be achieved so the blinded phase of the study was closed. The study was unblinded and extended, initially with an open-label period that allowed subjects to remain on ADV 10 mg treatment, and ultimately with an additional, 3-year, open-label, long-term safety and efficacy study period. The 48 week safety and efficacy data have been previously reported.15

Study GS-98-438 was a phase 3, double-blind, randomized, parallel-group, placebo controlled study.16 Study-Weeks 1 to 96 consisted of two successive 48-week double-blind treatment periods. One treatment group received ADV 10 mg once daily in both 48-week treatment periods (the ADV-ADV group); another treatment group received placebo in the first 48 weeks and ADV 10 mg once daily in the second 48 weeks (the Placebo-ADV group); a third treatment group received ADV in the first 48 weeks and placebo in the second 48 weeks (the ADV-Placebo group). This was followed by a long-term safety and efficacy study (study-weeks 97 to 240): This was an open-label uncontrolled study period in which all patients received ADV 10 mg once daily. Patients who received ADV during Study-Weeks 49 to 96 and completed Study-Week 96 could enter the long-term study (i.e., patients in the ADV-ADV and Placebo-ADV treatment groups). Patients who received ADV in the first 48 weeks and placebo in the second 48 weeks were not eligible to enter the long-term study, but they were offered the opportunity to enroll in an open-label continued access study. Five year efficacy and safety data have been previously reported.16

For both trials, male and female patients 16 to 65 years of age who had compensated liver disease were enrolled. Chronic hepatitis B was defined by the presence of detectable HBsAg for at least 6 months. Patients were to have a total bilirubin level of no more than 2.5 mg/dL (42.7 μmol/L), a prothrombin time that was no more than 1 second above the normal range, a serum albumin level that was at least 3 g/dL, a serum creatinine level of no more than 1.5 mg/dL (133 μmol/L), and an adequate blood count [absolute neutrophil count ≥ 1 × 103/mm3 (≥ 1 × 109/L); Platelets ≥ 100 × 103/mm3 (≥ 100 × 109/L); haemoglobin ≥ 10 g/dL (≥ 100 g/L) (males) or ≥ 9 g/dL (≥ 90 g/L)].

Criteria for exclusion included a coexisting serious medical or psychiatric illness; immuno-globulin, interferon, or other immune- or cytokine-based therapies with possible activity against HBV disease within 6 months before screening; organ or bone marrow transplantation; recent treatment with systemic corticosteroids, immunosuppressants, or chemotherapeutic agents; serum alpha-fetoprotein level of at least 50 ng/mL; evidence of a hepatic mass; liver disease that was not due to hepatitis B; prior therapy for more than 12 weeks with a nucleoside or nucleotide analogue with activity against HBV; and seropositivity for human immunodeficiency virus or hepatitis C or D virus.


The studies were conducted in compliance with the 1975 Declaration of Helsinki and approved by appropriate local regulatory bodies and Institutional Review Board/Ethics Committees. All patients gave written informed consent to participate.

Liver biopsies were required within 6 months prior to screening or before receiving study treatment and at week 48. Virological and biochemical assessments (serum HBV DNA and ALT levels, prothrombin time, and blood chemistry tests) and adverse-event monitoring were conducted every 4 weeks. Serum creatinine and phosphorus levels were assayed at baseline and every 4 weeks thereafter to monitor renal function. Genotypic analyses of HBV mutations were performed at baseline and week 48 for all patients with serum HBV DNA levels greater than the lower limit of quantification (see below).

The primary endpoint was histological improvement, defined as a reduction of at least 2 points in the Knodell necroinflammatory score with no concurrent worsening in the Knodell fibrosis score,21 missing biopsies were considered treatment failures. Paired baseline and week-48 liver biopsy specimens were assessed by a single histopathologist who was blinded to treatment assignment and sequence. Secondary endpoints included the effect of treatment on serum HBV DNA levels and ALT levels. Serum HBV DNA was measured by the Roche Amplicor Monitor PCR assay (lower level of quantification, 400 copies/mL).

The safety analysis included all patients who received at least one dose of study medication and all events that occurred during treatment or within 30 days of premature discontinuation of study drug. The severity of adverse events and laboratory abnormalities were graded according to the Common Toxicity Criteria of the National Institute of Allergy and Infectious Diseases.22

Clinical data were collected, monitored and entered into a database by Quintiles, a contract research organization. A central reference laboratory (Covance Laboratories, Indianapolis, IN, USA and Geneva, Switzerland) assessed all laboratory data. The sponsor held the data and conducted the statistical analyses.

Statistical analysis

Statistical analysis for the HBeAg+ (Study 437) and HBeAg− (Study 438) patients included Asian and Caucasian patients who received at least one dose of ADV 10 mg or placebo (n = 501). The analysis of histological end points included patients who had an assessable baseline biopsy specimen (n = 452), whereas HBV DNA and ALT endpoints included patients with assessable week 48 values (HBV DNA ≥ 400 c/mL and ALT > ULN at baseline additionally required for HBV DNA < 400 c/mL and ALT normalization) for the endpoint. The analyses compared the ADV 10 mg treatment group with the placebo group by race (Asians vs. Caucasians). All confidence intervals, significance tests, and resulting P-values were two sided, with an α level of 0.05. The Breslow-Day test allowed evaluation of homogeneity of the effect of race while adjusting for treatment, and was used to test for differences in response among the two races for the primary endpoint, histologic improvement. Change from baseline in HBV DNA was compared for Asian vs. Caucasian patients treated with ADV 10 mg using a Wilcoxon rank sum test. Comparison of adefovir 10 mg vs. placebo response for ALT normalization, HBV DNA undetectable, and histologic improvement was performed separately for Caucasians and for Asians using a Fisher’s Exact Test. Fisher’s exact test was also used to test for differences in adefovir 10 mg response rates between Asian and Caucasian patients for ALT normalization and HBV DNA undetectable. All serum HBV DNA values that were less than 400 copies/mL were considered to be at the lower limit of quantification.

A logistic regression analysis with the dependent variable being undetectable HBV DNA at week 48 and race and other potential confounders as independent variables was conducted. For Asian and Caucasian patients in studies 437 and 438 who received ADV 10 mg in year 1 of the study and had a week 48 HBV DNA measurement (n = 256), the following baseline characteristics were fit into logistic regression models to determine their importance with respect to predicting HBV DNA  < 400 copies/mL at week 48: race (Asian or Caucasian), gender (male or female), Knodell Fibrosis Score (Scores), Knodell Necroinflammatory Score (0–1, 2–4, 5–9, 10–14, 15–18) and HBeAg at Baseline (negative, positive/borderline). The following baseline characteristics were split into quartiles: age, ALT (xULN), HBV DNA and weight (kg).

All subjects who received study drug were included in the safety analyses.



Overall, fifty percent of patients enrolled in the 10 mg and placebo arms of the treatment studies were Asian and 46 per cent Caucasian [In Study 437 all were HBe Ag-positive and 203/338 (60%) were Asian. In Study 438 all subjects were HBe Ag-negative and 56/184 (30%) were Asian]. Seventy-five per cent of patients who received adefovir 10 mg or placebo were men. The mean age of these patients was 35 to 44 years across treatment and racial groups; ages ranged from 16 years to 66 years. Asian patients were slightly younger, had lower baseline ALT levels, and had a lower percentage of patients with prior interferon treatment. All other characteristics were similar among treatment/racial groups (Table 1).

Table 1.   Patient demographic and baseline HBV disease characteristics (ITT population)
 Caucasians (n = 242)Asians (n = 259)
Placebo (n = 100)Adefovir dipivoxil 10 mg (n = 142)Placebo (n = 121)Adefovir dipivoxil 10 mg (n = 138)
  1. HBV, hepatitis B virus; IFN, interferon; ITT, intent to treat; LAM, lamivudine.

  2. * Roche Amplicor Monitor polymerase-chain-reaction (LLQ = 400 copies/mL); † ULN 43 IU/L for males. 34 IU/L for females; ‡ IFN and LAM = discontinued > 6 months prior to baseline biopsy; § Defined as Knodell fibrosis score of 4.

Age (years)
 Mean ± s.d.43.9 ± 11.8942.0 ± 12.4534.6 ± 10.2235.9 ± 10.95
 Median (range) 43 (17–66)43 (16–65)34 (16–61)35 (16–63)
 Q1, Q334.5, 5332, 5027, 4027, 44
 Male (%)85 (85)111 (78)77 (64)110 (80)
HBV DNA (log10 copies/mL)*
 Mean ± s.d.8.00 ± 1.087.76 ± 1.067.62 ± 0.997.62 ± 1.12
 Median (range)8.18 (5.17–10.00)7.62 (5.51–10.16)7.72 (4.42–9.40)7.80 (3.67–9.61)
 Q1, Q37.31, 8.827.10, 8.596.91, 8.406.87, 8.50
 Mean ± s.d.168 ± 199.4143 ± 107.1122 ± 94.8142 ± 174.5
 Median (range)112 (44–1459)110 (20–742)90 (6–665)84 (31–1286)
 Q1, Q376, 17572, 17662, 14562, 149
Prior IFN (%)‡47 (47)70 (49)13 (11)14 (10)
Median total Knodell score10101010
Cirrhosis (%)§13 (13)15 (11)6 (5)9 (7)


After 48 weeks of treatment in the HBeAg+ and HBeAg− patients, ADV 10 mg resulted in histological improvement in 60 per cent compared to 26 per cent of placebo treated (P < 0.001, Fisher’s exact test) Caucasian patients and in 56 per cent of ADV 10 mg treated compared to 29 per cent of placebo (P < 0.001) treated Asian patients (Figure 1). Histological improvement rates were similar between Asian and Caucasian patients (P = 0.41, Breslow-Day Test). The reduction in HBV DNA from baseline at 48 weeks (3.89 vs. 3.70 log10 copies/mL for Caucasians and Asians, respectively) did not differ significantly between racial groups for patients treated with ADV 10 mg (P = 0.44, Wilcoxon rank sum test) (Figure 2). In addition, 34 per cent of Caucasian patients and 39 per cent of ADV treated Asian patients had undetectable serum HBV DNA (<400 copies/mL) at week 48 (Figure 3). The percentage of ADV treated patients achieving ALT normalization at week 48 was similar in both racial groups (Caucasian 64 per cent, Asian 63 per cent) (Figure 4). Among Asian patients, and among Caucasian patients, ADV 10 mg was significantly (P < 0.001, Fisher’s exact test) better than placebo for the proportion of patients with HBV DNA undetectable, ALT normalization, and histologic improvement.

Figure 1.

 Percent of patients with histological improvement at week 48. Improvement defined as ≥2 point reduction in the Knodell necroinflammatory score from baseline with no worsening in the Knodell fibrosis score; = 0.41 by Breslow-Day Test, missing or unassessable post-baseline biopsies treated as no improvement. Asian: adefovir dipivoxil (ADV) 10 mg vs. placebo (< 0.001); Caucasian: ADV 10 mg vs. placebo (< 0.001) Fisher’s Exact Test.

Figure 2.

 Median change in serum hepatitis B virus DNA at Week 48. Roche Amplicor Montitor PCR (LLQ = 400 copies/mL); = 0.44 by Wilcoxon rank sum test for comparison of Caucasians vs. Asians on adefovir dipivoxil (ADV) 10 mg.

Figure 3.

 Per cent of patients with undetectable serum hepatitis B virus DNA through week 48. Roche Amplicor Monitor PCR (LLQ = 400 copies/mL). Asian: adefovir dipivoxil (ADV) 10 mg vs. placebo at week 48 (< 0.001); Caucasian: ADV 10 mg vs. placebo at week 48 (< 0.001) Fisher’s Exact Test. Adefovir 10 mg group: Caucasian vs. Asian (= 0.44, Fisher’s exact test).

Figure 4.

 Per cent of patients with ALT normalization through week 48. ULN 43 IU/L for males; 34 IU/L for females. Asian: adefovir dipivoxil (ADV) 10 mg vs. placebo at week 48 (< 0.001); Caucasian: ADV 10 mg vs. placebo at week 48 (< 0.001) Fisher’s Exact Test. Adefovir 10 mg group: Caucasian vs. Asian (= 0.90 Fisher’s exact test).

Analysis of HBeAg or HBsAg seroconversion by race was not performed due to the overall low rate of occurrence at Week 48 in the pivotal studies. Among HBeAg− patients in study 438, only six patients had confirmed HBsAg loss or HBsAg seroconversion (four Caucasian – three genotype D and one genotype A and two Asian – both genotype C). [In study 437, 9/161 HBeAg+ placebo patients (six Asian, three Caucasian) and 20/171 (10 Caucasian – six genotype A, three genotype D and one genotype G, nine Asian – one genotype A, three genotype B, four genotype C, one genotype D and one Black – genotype E) patients treated with ADV 10 mg had HBeAg seroconversion]; no patient had HBsAg seroconversion at Week 48. Among patients treated with ADV 10 mg, there was no statistically significant differences in the proportion of Asian vs. Caucasian patients with HBeAg seroconversion (P = 0.20 from 2-sided Fisher’s Exact Test).

Logistic regression analysis

When each baseline variable was tested one at a time, higher baseline age, lower baseline HBV DNA and negative baseline HBeAg status at baseline were important (P < 0.05) univariate predictors of HBV DNA < 400 c/mL at week 48. When a stepwise logistic regression model was used with entry and exit criteria of P = 0.15 [same results observed with entry and exit criteria P = 0.10], variables entered into the model included baseline HBV DNA, baseline weight (lower weight associated with better response), and baseline HBeAg status. When entry and exit criteria of P = 0.05 were used, HBeAg status was removed from the model leaving only lower baseline HBV DNA and lower baseline weight as important predictors of HBV DNA < 400 c/mL at week 48.

Race was not an important predictor of HBV DNA < 400 c/mL at week 48 as a univariate predictor or when adjusting for important baseline predictors. The table below summarizes P-values for individual predictors and shows that race was not an important predictor of HBV DNA < 400 c/mL at week 48 as a univariate predictor or when adjusting for important baseline predictors.

Model #RaceBaseline weightBaseline HBV DNABaseline HBeAg Status


The adverse events occurring in ≥15 per cent of patients and all grade 3 and 4 laboratory abnormalities are shown in Table 2 and Table 3, respectively. ADV was well tolerated by both racial groups, with similar frequencies of adverse events and laboratory abnormalities in the ADV - and placebo-treated groups. However, pharyngitis and cough appeared to be more common in Asians as compared to Caucasians. Because the median change from baseline in serum creatinine was zero following 48 weeks of treatment with ADV 10 mg in both study 437 and 438, there was no comparison by race.

Table 2.   Adverse events reported by at Least 15 per cent of patients in a treatment group
Placebo (n = 100)Adefovir dipivoxil 10 mg (n = 142)Placebo (n = 121)Adefovir dipivoxil 10 mg (n = 138)
Flu syndrome19%13%20%16%
Abdominal pain12%16%17%15%
Cough increased4%3%17%12%
Table 3.   Grade 3/4 laboratory abnormalities
Placebo (n = 100)Adefovir dipivoxil 10 mg (n = 142)Placebo (n = 121)Adefovir dipivoxil 10 mg (n = 138)
  1. AST, aspartate aminotransferase; GGT, gamma-glutamyltranspeptidase; PT, prothrombin time.

Creatine kinase1%7%11%7%
Serum amylase3%<1%4%5%
Total bilirubin3%<1%2%0%
PT prolonged0%0%0%1%
Creatinine increase0%0%0%0%


No HBV with an adefovir-resistance mutation was detected in any HBeAg− or HBeAg+ Asian or Caucasian patient in the first 48 weeks of these studies.


There are differences in chronic hepatitis B in different ethnic populations that may result in differential response to treatment. For instance, chronic hepatitis B in Asians is perinatally acquired while chronic hepatitis B in Caucasians is more likely to be acquired in adulthood.4 Epidemiological data also demonstrate a higher prevalence of chronic hepatitis B in Asians.23 The reason(s) for this is unclear, however, there is a possibility that ethnic differences contribute to susceptibility towards persistence of chronic hepatitis B infection.24 A possible explanation is the association of genes with increased susceptibility to chronic hepatitis B, including histological activity index (HLA) class I and class II alleles, TNF polymorphisms and chemokine polymorphisms.24

While susceptibility to development of chronic hepatitis B and acquisition of HBV are important, how does this ethnic difference affect clinical outcome? There is circumstantial evidence from cohort studies in Caucasians25, 26 that the complications of liver disease are lower than in a similar Asian population. Another aspect of the difference between Asians and Caucasians relate to the genotype differences. Asians have predominantly genotypes B and C while Caucasians have predominantly genotypes A and D. Asian patients with genotype C appear to have a poorer response to interferon alpha,27 while in Caucasians genotype D is associated with a poorer response to interferon alpha.28 In addition, genotype A in particular, is associated with a high rate of HBsAg clearance particularly in patients treated with pegylated interferon.29

The results from our study show that four patients who cleared HBsAg after week 48 were Caucasians, while two were Asians. The numbers of these patients are too small to make definitive conclusions. Nonetheless, the body of evidence suggests that there are differences in Asians and Caucasians with regards to many aspects of chronic hepatitis B infection. Consequently, it is of interest to determine whether there are ethnic differences in the response to therapy. A previous analysis of the same adefovir dataset from studies 437 and 438 show that there were no genotypic differences in response to therapy.30 When we now examine the combined data from two randomized, double-blind, placebo-controlled studies in HBeAg+ and HBeAg− patients, treatment with ADV results in significant histological, virological, and biochemical improvement after 48 weeks in both Asian and Caucasian patients, with the ADV 10 mg response rates being similar among patients of either ethnic background.

Efficacy results of this study are similar to previously published results of lamivudine treatment in Asian and Caucasian patients.8, 9 In addition, 48 weeks of ADV treatment resulted in no adefovir-resistance mutations,15, 31 compared with resistance rates of approximately 20% per year reported with lamivudine. Although drug resistance is not a concern with IFN-α, it is associated with poor treatment response in Asians6 and has significant side effects. The results of this study contrast with interferon studies where ethnic background or genotype may affect the treatment response. To date, there have been no published studies suggesting that ethnic differences affect treatment response with nucleoside analogues. However, this is the first study that has examined this question specifically in nucleotides.

Adefovir dipivoxil was also well tolerated in Asians and Caucasians for up to 48 weeks of treatment. ADV is renally cleared, but no significant changes in the median serum creatinine levels up to week 48 were observed. In patients with creatinine clearance <50 mL/min, systemic clearance of adefovir is dramatically reduced. Dose interval modification is recommended for these patients as noted in the prescribing information for ADV.32

In summary, pooled data from the above two studies demonstrate that after 48 weeks of ADV therapy, significant histological, virological, and biochemical improvements were observed in both Asian and Caucasian HBeAg+ and HBeAg− patients; and response rates observed among patients treated with ADV 10 mg were similar across both ethnic groups. No resistant mutations were observed after 48 weeks of therapy in either group of patients. ADV is safe and effective for the treatment of HBV in both Asians and Caucasians. In conclusion, adefovir dipovoxil is suitable for therapy in Asians and Caucasians with HBeAg positive and negative chronic hepatitis B infection.


The authors thank the following investigators for their contributions: Dr. Paul Desmond; Dr. Ian Kronborg; Dr. Peter Angus; Dr. G.W. McCaughan; Dr. Graham Cooksley; Dr. David Shaw; Dr. Darrell Crawford; Dr. Joe Sasadeusz; Dr. Meng Ngu; Dr. Stephen Sacks (deceased); Dr. Morris Sherman; Dr. Elizabeth Heathcote; Dr. Eric Yoshida; Dr. Bincent Bain; Dr. Brian Conway; Dr. Kelly Kaita; Dr. Jean-Pierre Villeneuve; Dr. Christian Trepo; Dr. Dennis Vetter; Dr. JP Pascal; Dr. Samuel Didier; Dr. Jean Pierre Zarski; Dr. Marc Bourliere; Dr. Patrice Couzigou; Dr. Stanislas Pol; Dr. Andre Gerolami-Santandreas; Dr. JC Trinchet; Dr. Michael P. Manns; Dr. Wolfgang E. Fleig; Dr. Michael Gregor; Dr. Peter Buggisch; Dr. Georgios Kitis; Dr. Yoav Lurie; Dr. Elchanan Nussenson; Dr. Pietro Andreone; Dr. Mario Rizzetto; Dr. Giuseppe Pastore; Dr. Antonio Francavilla; Dr. Ferruccio Bonino; Dr. Antonio Craxi; Dr. Remo Naccarato; Dr. Giovanni B. Gaeta; Dr. Alfredo Alberti; Dr. Rosmawati Mohamed; Dr. Ismail Merican; Dr. Ji Jiang; Dr. Yun-Hong Wang; Dr. Judy Lao; Dr. Jose Sollano; Dr. Han Seong Ng; Dr. Rafael Esteban; Dr. Vicente Olaso; Dr. Jaime Enriquez; Dr. Manuel Rodriguez; Dr. Francisco Perez Roldan; Dr. Yun-Fan Liaw; Dr. Pinit Kullavanijaya; Dr. Pairoj Luengrojanakul; Dr. Richard John Gilson; Dr. Mark Nelson; Dr. Ellie Barnes; Dr. Thomas W. Warnes; Dr. Lennox Jeffers; Dr. Vinod Rustgi; Dr. Paul Martin; Dr. Adrian DiBisceglie; Dr. Vijayan Balan; Dr. Karen Lindsay; Dr. Mitchell Shiffman; Dr. Nezam Afdhal; Dr. Luis Balart; Dr. Michael T. Bennett; Dr. Robert Brown; Dr. Hari Conjeevaram; Dr. Tse-Ling Fong; Dr. Michael Fried; Dr. Hans Fromm; Dr. Robert Gish; Dr. Stuart Gordon; Dr. Tarek Hassanein; Dr. John Hoefs; Dr. Ira Jacobson; Dr. George Koval; Dr. Kris Kowdley; Dr. Anna Lok; Dr. John McHutchison; Dr. Ed Wakil; Dr. Caroline Riely; Dr. Robin Rutherford; Dr. Raj Satyanarayana; Dr. Paul Thuluvath; Dr. Stephen P. Esposito; Dr. David Bernstein; Dr. Marie Borum; Dr. Thomas Faust; Dr. Douglas Dieterich; Dr. Teresa Wright; Dr. Robert Perillo; Dr. Jules Dienstag; Francoise Schuller; Narinder Virk; Catherine Scherrer; Leslie Richardson; Dr. Oren Cohen.

Declaration of personal interests: S. G. Lim, P. Marcellin, N. Tassopoulos, S. Hadziyannis, T. T. Chang, M. Tong, W. Sievert, and P. Hu have served as clinical investigators and have received funding for their participation from Gilead Sciences. S. Arterburn and C. L. Brosgart are employees of Gilead Sciences and own stocks and shares in Gilead Sciences. Declaration of funding interests: The studies described in this manuscript were funded in full by Gilead Sciences. The writing/preparation of this paper was completed by the primary author with input from all other authors but without additional funding from any source. No additional personnel performed any data analyses.