Potential conflict of interest: Dr. Harb owns stock in Novartis.
The trial registration number is NCT00131742.
Chronic hepatitis B and its life-threatening sequelae are highly prevalent in China. There is a need for effective new therapies to suppress hepatitis B virus (HBV) replication and ameliorate liver disease. In this study, we compared the efficacy of telbivudine, a nucleoside analogue, with lamivudine in Chinese patients. In this phase III, double-blind, multicenter trial conducted in China, 332 patients with compensated hepatitis B e antigen (HBeAg)–positive or HBeAg-negative chronic hepatitis B were randomly assigned to treatment with 600 mg of telbivudine or 100 mg of lamivudine daily for 104 weeks. The primary efficacy endpoint was reduction in serum HBV DNA levels at week 52 of treatment. Secondary endpoints included clearance of HBV DNA to undetectable levels, HBeAg loss and seroconversion, therapeutic response, and alanine aminotransferase (ALT) normalization. Viral resistance and safety were assessed. At week 52, among 290 HBeAg-positive patients, mean reductions of serum HBV DNA were significantly greater in telbivudine recipients than lamivudine recipients (6.3 log10 versus 5.5 log10, P < 0.001), and HBV DNA was polymerase chain reaction–negative in significantly more telbivudine recipients than lamivudine recipients (67% versus 38%, P < 0.001). ALT normalization (87% versus 75%, P = 0.007), therapeutic response (85% versus 62%, P = 0.001), and HBeAg loss (31% versus 20%, P = 0.047) were also significantly more common in the telbivudine group. Treatment effects showed similar patterns in the smaller HBeAg-negative group (n = 42). Viral resistance in telbivudine recipients was approximately half that observed with lamivudine; however, this difference was not statistically significant. Clinical adverse events were similar in the two treatment groups. Conclusion: In Chinese patients with chronic hepatitis B, telbivudine treatment for 52 weeks provided greater antiviral and clinical efficacy than lamivudine, with less resistance. (HEPATOLOGY 2007.)
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Eradication of hepatitis B virus (HBV) infection remains an unfulfilled global health goal. The end-stage sequelae of chronic HBV infection, cirrhosis-related liver failure, and hepatocellular carcinoma are responsible for over 1 million deaths each year worldwide.1 More than one-third of the world's 350 to 400 million chronic HBV carriers live in China, where the average prevalence of chronic HBV infection is over 9%.2–4 Among patients of Chinese ancestry, chronic HBV infection leads to a striking increase in the life-time risk of cirrhosis, hepatic decompensation, hepatocellular carcinoma, and liver-related premature mortality.5, 6 It is estimated that over 500,000 Chinese die annually from end-stage hepatitis B complications.7 In China, as in other countries, progression to end-stage disease is associated with increased healthcare costs and other socioeconomic burdens.
The prevalence of specific HBV genotypes in China differs from that observed in Western countries and may contribute to differences in disease characteristics.8–10 Chinese HBV carriers are infected almost exclusively with genotype C or genotype B, whereas Western patients of non-Asian ethnicity display an array of HBV genotypes, primarily genotypes A, D, E, and F.
In chronic HBV carriers, the risks of progression of chronic hepatitis B to cirrhosis and to hepatocellular carcinoma have been linked to high levels of circulating HBV DNA,11, 12 and recent trials with antiviral agents have indicated that suppression of HBV replication can improve outcomes and reduce the risks of disease progression.13, 14 Unfortunately, many patients fail to respond optimally to current therapies because of efficacy limitations, tolerability issues, or the emergence of resistance. Thus, new antiviral agents continue to be needed to maximize longer term suppression of HBV replication and improve therapeutic outcomes.
Telbivudine, an orally bioavailable l-nucleoside analogue, is a specific and potent inhibitor of HBV replication in vitro.15 Toxicology studies revealed a favorable preclinical safety profile, with no evidence of significant organ toxicity, genotoxicity, carcinogenicity, mitochondrial toxicity in vitro, teratogenicity, or embryofetal toxicity.16 Telbivudine treatment produced rapid and profound reductions in serum HBV DNA levels in a 4-week phase I/II dose-escalation trial and a 1-year phase IIb trial conducted with Asian and Western hepatitis B patients.17, 18 These results were confirmed and expanded in a large international phase III trial (the GLOBE study), which compared telbivudine and lamivudine in a population of 1367 patients with HBeAg-positive or HBeAg-negative chronic hepatitis B; 373 of these resided in mainland China.19 In the GLOBE study, telbivudine treatment was associated with significantly greater responses, compared with lamivudine, for all direct measures of antiviral efficacy. To gain additional clinical data in Chinese patients, we have conducted a similarly designed phase III trial in China.
The goals of the present study were (1) to gain additional phase III data on telbivudine in Chinese patients with chronic hepatitis B, as required by Chinese regulatory guidelines; (2) to compare telbivudine and lamivudine in the relatively homogeneous Chinese patient population, in which the potentially confounding effects of ethnic variation could be minimized; and (3) to assess direct antiviral effect, reflected by serum HBV DNA reduction, as the primary efficacy measure, in contrast to the composite serologic efficacy measure used in the international GLOBE trial. The primary results at 1 year from this study are described in this report.
ALT, alanine aminotransferase; AST, aspartate aminotransferase; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; PCR, polymerase chain reaction; ULN, upper limit of normal.
Patients and Methods
This multicenter, double-blind, randomized phase III trial was designed to assess 2 years (104 weeks) of treatment with telbivudine versus lamivudine in Chinese adults with compensated chronic hepatitis B. The primary efficacy endpoint was serum HBV DNA reduction at 1 year (52 weeks), with treatment continuing for a second year to assess longer term efficacy and safety. Eligible patients were randomly assigned in a 1:1 ratio to 600 mg of telbivudine plus lamivudine placebo or 100 mg of lamivudine plus telbivudine placebo, once daily. The dose of telbivudine selected for study was based on earlier dose-response studies showing that this dose produced a nearly maximal antiviral effect.20 After patient eligibility was determined through clinical and laboratory screening assessments, a centralized telecommunication-based interactive voice response system was used for patient randomization. At randomization, patients were stratified by HBeAg status (positive or negative) and serum alanine aminotransferase (ALT) level [<2.5 or ≥2.5 times the upper limit of normal (ULN)]. The first dose of the study drug was taken at the baseline visit, and patients were followed up in the first year at weeks 2, 4, 8, 12, 16, 24, 32, 40, 48, and 52, with second-year visits at weeks 60, 68, 76, 84, 92, 100, and 104. At each visit, serum HBV DNA levels, routine laboratory tests, adverse events, and concurrent medications were assessed. Serum for HBeAg, hepatitis B surface antigen (HBsAg), and the corresponding antibodies (antibodies to HBeAg and antibodies to HBsAg) was obtained at weeks 12 and 24 and subsequently according to the aforementioned schedule. For patients completing 104 weeks of treatment, participation in a follow-up protocol was offered. Posttreatment safety assessments at 4-week intervals for 4 months were undertaken for patients electing not to participate in the follow-up study and for those who discontinued study participation prematurely. Hematologic parameters and serum chemistries were measured with standard assays at a central reference laboratory (Quintiles Laboratories, Beijing, China). Serum HBV DNA levels were quantified at the central laboratory with the Roche COBAS Amplicor polymerase chain reaction (PCR) assay (Roche Molecular Systems, Branchburg, NJ), with an estimated lower limit of quantitation of 300 HBV genome copies/mL. This trial was conducted under an Investigational New Drug application by the US Food and Drug Administration and with approval received from the Chinese State Food and Drug Administration. The trial protocol was approved by ethics committees at each participating center, and all patients gave written informed consent. Clinical data were collected, monitored, and entered into a database by Quintiles Transnational, Inc. (Research Triangle Park, NC), in accordance with standardized data management and quality assurance procedures.
Eligible patients were Chinese males or females, 16 to 70 years of age, with a clinical history compatible with chronic hepatitis B and active viral replication, documented by positive serum HBsAg, HBeAg-positive or HBeAg-negative, serum HBV DNA ≥6 log10copies/mL, serum ALT levels ≥1.3 times ULN but <10 times ULN at the screening visit, and a liver biopsy compatible with chronic hepatitis B obtained within 12 months prior to randomization. Exclusion criteria included history or evidence of decompensated liver disease; pregnancy or breastfeeding, unwillingness to use a double-barrier method of contraception; coinfection with hepatitis C virus, hepatitis D virus, or human immunodeficiency virus; previous treatment for HBV with nucleoside or nucleotide analogues; treatment with interferon or other immunomodulators in the 12 months prior to screening; abuse of alcohol or illicit drugs within the past 2 years; frequent or prolonged use of systemic corticosteroids, acyclovir or famciclovir; hepatocellular carcinoma or other malignancy requiring treatment; other serious medical conditions that might confound efficacy or safety assessments; use of anticoagulants; and a history of clinical pancreatitis. Laboratory exclusion criteria included hemoglobin <11 g/dL for men and <10 g/dL for women; neutrophil count <1500/mm3; platelet count <75,000/mm3; serum creatinine ≥1.5 mg/dL; serum amylase and lipase levels ≥1.5 times ULN; serum albumin <3.3 g/dL; prothrombin time prolonged by >3 seconds over the upper limit of the reference value; and total bilirubin ≥2.0 mg/dL. Patients with serum alpha-fetoprotein >50 ng/mL required exclusion of underlying hepatocellular carcinoma prior to randomization.
Efficacy and Safety Endpoints.
The primary efficacy endpoint was reduction in serum HBV DNA (from baseline level) at week 52 of treatment. Secondary efficacy measures included proportions of patients with serum HBV DNA reduction to <5 log10 copies/mL on two successive visits, HBV DNA reductions to PCR-undetectable levels, ALT normalization, HBeAg loss and seroconversion (for HBeAg-positive patients), and therapeutic response. Among HBeAg-positive patients, therapeutic response was defined as a composite serologic endpoint with HBV DNA reduced to <5 log10 copies/mL coupled with HBeAg loss or ALT normalization. In HBeAg-negative patients, therapeutic response was defined as HBV DNA reduced to <5 log10 coupled with serum ALT normalization. Primary treatment failure was defined as failing to achieve serum HBV DNA levels below 5 log10 copies/mL after at least 24 weeks of treatment. Safety endpoints, assessed in the combined intent-to-treat population, included serious and nonserious adverse events and graded laboratory abnormalities.
Viral Breakthrough and Resistance.
Viral breakthrough was defined as a persistent (2 consecutive determinations) on-treatment increase in HBV DNA of greater than 1 log10 from nadir, consistent with recent recommendations.21, 22 Resistance was defined as viral breakthrough with identified treatment-emergent resistance mutations.22 HBV DNA was amplified by PCR from screening/baseline and week 48 sera for all patients with viral breakthrough. The entire 344-codon reverse-transcriptase domain of the HBV polymerase gene was determined by automated sequencing at an independent reference laboratory (Delft Diagnostic Laboratory, Voorburg, The Netherlands). This method detects reverse-transcriptase mutations potentially associated with resistance and comprising at least 25% of the amplified HBV DNA copies. HBV genotype was determined by a comparison of sequences against reference standards.
All randomized patients who received at least one dose of study medication and had at least one observation after baseline were included in the intent-to-treat population. Descriptive statistics of demographic and baseline features were summarized by treatment arms. Categorical variables were compared by chi-square testing, and continuous variables were compared by the two-sided Student t test. The Cochran-Mantel-Haenszel method was used to combine stratified subgroups for testing response rates between treatment groups. For safety analyses, incidence of adverse events and frequency of grade 3 or 4 laboratory abnormalities were summarized by treatment groups and compared with Fisher's exact test. The study was powered for treatment differences on the primary efficacy endpoint at 1 year (HBV DNA reduction) in the overall study population. This design was based on the expectation that the primary endpoint would not show a statistical interaction with HBeAg status. However, current treatment guidelines recommend different management strategies for HBeAg-positive and HBeAg-negative patients; therefore, the data for these groups are reported separately.23 Although the HBeAg-positive population (87% of the total) was sufficient for preplanned statistical analyses, these analyses had low statistical power for the much smaller HBeAg-negative population.
A total of 332 Chinese patients comprised the intent-to-treat population, including 290 patients (87%) with HBeAg-positive chronic hepatitis B and 42 patients (13%) with HBeAg-negative chronic hepatitis B at screening. Of the 167 patients randomized to receive telbivudine, 147 were HBeAg-positive, and 20 were HBeAg-negative (Table 1). The 165 patients randomized to lamivudine comprised 143 HBeAg-positive and 22 HBeAg-negative patients. Nearly two-thirds of patients were infected by HBV genotype C, and one-third were infected by HBV genotype B (Table 1). No other HBV genotypes or mixed genotypes were found. Baseline demographic and disease characteristics were similar among patients in the telbivudine and lamivudine groups. HBV DNA levels were lower in HBeAg-negative patients, as seen previously, but more HBeAg-positive patients had ALT levels >2.5 times ULN.24 Four patients in the telbivudine group and 5 in the lamivudine group withdrew from the study before week 52, of which one in each treatment group (0.6%) was discontinued for adverse events, clinical disease progression, or lack of efficacy.
Table 1. Baseline Demographic and Disease Features
All patients were HBsAg-positive and Chinese.
Age: mean years (range)
Gender: male %
Weight: mean kg (range)
HBV DNA: mean log10 copies/mL (standard error)
HBV DNA: median log10 copies/mL (25%, 75%)
9.7 (9.0, 10.1)
9.7 (9.0, 10.1)
7.3 (6.5, 9.3)
7.2 (6.4, 9.3)
ALT: mean IU/L (standard error)
Efficacy in HBeAg-Positive Patients.
Telbivudine treatment resulted in greater suppression of HBV replication, reflected by significantly greater reduction in serum HBV DNA levels, compared with lamivudine (Fig. 1). This difference in HBV DNA suppression was significant by week 8 and continued to widen through week 52. At week 52, serum HBV DNA reduction from baseline was significantly greater for telbivudine (6.3 log10) than lamivudine (5.5 log10; P < 0.001; Table 2). Serum HBV DNA became PCR-negative (<300 copies/mL) more rapidly in telbivudine-treated patients (Fig. 2), and PCR negativity at week 52 was significantly more frequent with telbivudine treatment compared to lamivudine (67% versus 38%, P < 0.001). Correspondingly, the proportion of patients with primary treatment failure (serum HBV DNA remained above 5 log10 copies/mL throughout the 52 weeks of treatment) was significantly lower with telbivudine compared to lamivudine (4% versus 18%, P < 0.001). Therapeutic response was significantly more common in the telbivudine group (85%) compared with lamivudine (62%; P < 0.001), and serum ALT levels were normalized in 87% of telbivudine recipients versus 75% of lamivudine recipients (P = 0.007). HBeAg loss was significantly more frequent in the telbivudine group compared with lamivudine (31% versus 20%, P = 0.047). HBeAg seroconversion was more frequent with telbivudine (25%) compared with lamivudine (18%), but this difference was not statistically significant (P = 0.14). No patient experienced HBsAg loss or seroconversion.
Calculated for patients with serum ALT >1 times the ULN at baseline (n = 142 and 135 for telbivudine and lamivudine, respectively).
HBV DNA reduced to <5 log10 copies/mL coupled with HBeAg loss or ALT normalization.
Calculated for patients HBeAg-positive at baseline (n = 138 for each treatment group).
Serum HBV DNA remained above 5 log10 copies/mL through the 52 weeks of treatment.
Viral resistance is defined as breakthrough with treatment-emergent resistance mutations confirmed by genetic sequencing at week 48. In the study protocol, viral breakthrough was defined primarily as an increase of serum HBV DNA to at least 5 log10 copies/mL, following reduction to below that level. During the course of the study, this definition was superseded by the simpler and more widely accepted definition that pertains to the data reported above. Resistance rates per protocol were 5.1% and 10.7% for telbivudine and lamivudine, respectively, in HBeAg-positive patients.
Difference = telbivudine minus lamivudine, with 95% confidence interval.
Serum HBV DNA (mean log10 reduction from baseline)
In the substantially smaller HBeAg-negative patient group (n = 42 patients), efficacy responses were similar to those in HBeAg-positive patients. Telbivudine produced a greater mean reduction of serum HBV DNA (5.5 log10for telbivudine versus 4.8 log10 for lamivudine), higher rates of therapeutic response (100% versus 82%), ALT normalization (100% versus 78%), and PCR-negative HBV DNA (85% versus 77%), and less primary treatment failure (0% for telbivudine versus 5% for lamivudine). However, these efficacy differences were not analyzed statistically because of the limited power for statistical comparisons within the small HBeAg-negative patient population. No patient experienced HBsAg loss or seroconversion.
Viral Breakthrough and Resistance.
Among HBeAg-positive patients, viral breakthrough was significantly more common in the lamivudine arm at week 48 compared with telbivudine (17.5% versus 7.5%, P = 0.009; Table 2). Resistance (breakthrough with detection of treatment-emergent resistance mutations) was also more common with HBeAg-positive lamivudine recipients (21/143, 14.7%) compared with telbivudine (11/147, 7.5%; P = 0.06); however, this difference did not reach statistical significance. One HBeAg-negative lamivudine recipient experienced viral breakthrough in this study; however, no resistance was detected in HBeAg-negative patients in either treatment group.
The signature M204I telbivudine resistance mutation was found in all 11 HBeAg-positive telbivudine recipients with viral breakthrough at 1 year.22, 25 The M204I mutation was accompanied by L80I or L80V mutations in 8/11 telbivudine-resistant patients. L80 mutations were found only in association with M204I, suggesting a compensatory role. No other form of M204-associated resistance mutation was observed in telbivudine recipients, including the M204V-L180M double mutant that is commonly associated with lamivudine resistance. Of the 26 HBeAg-positive lamivudine recipients with viral breakthrough, 4 retained essentially wild-type sequences at week 48, possibly reflecting loss of response for noncompliance or other reasons unrelated to resistance. Signature (YMDD-related) resistance mutations were detected at week 48 in the remaining 22 lamivudine recipients with viral breakthrough. Mutation of the M204 locus was also detected before treatment in one of these patients, who was therefore excluded from subsequent resistance analyses. Among the remaining 21 lamivudine recipients with resistance, 3 different patterns of treatment-emergent genotypic mutations were observed at codon M204. The M204V mutation was found in 12/21 (57%) patients with lamivudine resistance, all of which were accompanied by an L180M compensatory change. The M204I mutation was found in 8/21 (38%) lamivudine-resistant patients and was accompanied by L80 mutations in 5/8 cases. A mixed M204M/I pattern was found in a single lamivudine patient (5%). There was no apparent association of HBV genotype with resistance mutation pattern. Five of the 11 patients with telbivudine resistance and 10 of the 21 patients with lamivudine resistance were HBV genotype B, with the remainder genotype C.
Safety and Tolerability.
Both study drugs were generally well tolerated. Adverse events were reported in about half of the patients in both treatment arms (Table 3); most adverse events were not attributed to the study drug by the clinical investigators. Nasopharyngitis was reported in 24% to 29% of lamivudine-treated and telbivudine-treated patients and was the most frequently reported event. Other clinical events occurred in <6% of patients. Complaints of myalgia and other muscle-related adverse events were uncommon and occurred with similar frequency in both treatment groups. One patient in the telbivudine group developed a polymyositis that was not attributed to the study drug; the event abated subsequently, with continued treatment. Grade 3 or 4 serum ALT and aspartate aminotransferase (AST) elevations were more common in the lamivudine group than the telbivudine group (9.1% versus 5.4% and 6.7% versus 5.4%, respectively). All grade 3 or 4 aminotransferase elevations seen in the lamivudine arm in the last 6 months of study treatment and the single aminotransferase elevation in the telbivudine arm were associated with viral breakthrough. Grade 3 or 4 elevations of creatine kinase were more common in the telbivudine group than the lamivudine group, but this difference did not reach statistical significance (8.4% versus 3.0%, P = 0.06).
Table 3. Most Frequent Clinical Adverse Events and Grade 3/4 Laboratory Abnormalities
Telbivudine (n = 167)
Lamivudine (n = 165)
Most frequent adverse events (patients with event, regardless of attributability to study drug)
Total patients with event
Upper respiratory tract infection
Abdominal pain upper
Grade 3/4 laboratory abnormalities (patients with new onset abnormalities)
Absolute neutrophil count
Maximizing and maintaining HBV suppression is emerging as a key goal for achieving long-term improvement in clinical outcomes in patients with chronic hepatitis B.13, 19 In this study of 332 patients from China with HBeAg-positive or HBeAg-negative chronic hepatitis B, telbivudine treatment produced rapid and profound HBV suppression and was superior to standard lamivudine therapy with respect to the primary efficacy endpoint (mean HBV DNA reduction from baseline to week 52) in HBeAg-positive patients. In this major subpopulation, telbivudine treatment was also associated with a significantly greater proportion of patients achieving PCR-negative HBV DNA, ALT normalization, therapeutic response, and HBeAg loss, with significantly less viral breakthrough and primary treatment failure (HBV DNA remaining above 5 log10 copies/mL).
Efficacy differences also generally favored telbivudine in the much smaller HBeAg-negative subpopulation. The study was designed without stratification of enrollment based on HBeAg status. Exploratory analysis of data from the telbivudine GLOBE study suggested that the primary endpoint in the present study (HBV DNA reduction) would not be susceptible to a statistical interaction with HBeAg status, in contrast to therapeutic response, which was the primary endpoint in the GLOBE study. Moreover, with a study conducted in mainland China, the proportion of eligible HBeAg-negative patients was expected to be relatively low. For these reasons, the present study was statistically powered for treatment differences in HBV DNA reduction in the overall study population comprising both HBeAg-positive and HBeAg-negative patients. Observations in the minor HBeAg-negative patient group therefore afford limited statistical inferences.
The results of this study in Chinese patients are consistent with those of other hepatitis B studies with telbivudine and other agents, which have shown an association between greater early viral suppression and subsequent higher rates of ALT normalization and histologic improvement, suggesting a link between the magnitude of viral suppression and improvement in HBV-related hepatic necroinflammation.26–29
Resistance to telbivudine after 1 year was less than half that observed with lamivudine, consistent with the previous phase III GLOBE trial with telbivudine, although the difference did not reach statistical significance in the present study, presumably because of the more limited sample size in comparison with the GLOBE study. The difference presumably reflects the greater antiviral efficacy of telbivudine as well as the fewer pathways to telbivudine resistance. Only the M204I primary resistance mutation emerged during telbivudine treatment in this study, consistent with results of the GLOBE trial and the earlier phase II trial, in contrast to the multiple mutations observed with lamivudine.18, 25
With the present results, telbivudine has now demonstrated greater antiviral efficacy than lamivudine in three randomized trials.18, 19 In addition, a 52-week randomized open-label comparison of telbivudine and adefovir indicated substantially greater and more consistent antiviral efficacy for telbivudine compared to adefovir, with similarly low 1-year viral breakthrough rates.30 There are no head-to-head comparative data for telbivudine versus entecavir. In a study of 519 Chinese patients similar to those enrolled in the present trial, viral load declined by 5.9 log10 copies/mL after 1 year with entecavir, 76% of entecavir-treated patients were PCR-negative, and ALT was normalized in 90%; all of these results were statistically greater than those obtained in the lamivudine comparator group.31, 32 However, HBeAg seroconversion was reported in 15% of entecavir-treated patients versus 18% for lamivudine. Two patients experienced viral breakthrough with entecavir but without detection of resistance mutations. No randomized comparisons of telbivudine or entecavir with interferon or pegylated interferon have been conducted in China or elsewhere. Interferons have been associated with higher rates of HBsAg responses, compared with nucleos(t)ide analogs.23 However, the frequency of these responses appears lower in Chinese interferon recipients than in Caucasian interferon recipients, and long-term follow-up suggests limitations of HBeAg or HBsAg seroconversion as treatment endpoints in this population.33, 34
Telbivudine and lamivudine demonstrated similar clinical adverse event profiles in this study. The clinical safety observations for telbivudine in this and other trials suggest that telbivudine has a promising safety profile for long-term treatment, and there was no evidence of carcinogenicity or embryofetal toxicities in preclinical studies.16 The incidence of myositis in one telbivudine recipient and occasional cases of myopathy in other studies suggest that persistent, unexplained muscle-related symptoms with telbivudine should be evaluated promptly.35
Second-year data from this study and from the GLOBE study will be helpful in determining the longer term efficacy and safety of telbivudine. Newer and more potent anti-HBV agents such as telbivudine offer the prospect of more effective control of HBV replication, with the potential to improve long-term outcomes for more patients.