Efficacy of telbivudine in Taiwanese chronic hepatitis B patients compared with GLOBE extension study and predicting treatment outcome by HBV DNA kinetics at Week 24

A total of 59 Taiwanese patients in the telbivudine arm and 56 Taiwanese patients in the lamivudine arm from the intent to treat (ITT) population of the GLOBE study were included in this study. Signed informed consent was obtained from the patients and the treatment was approved by the joint institutional review board (NV-02B-022) and the department of health (0940305529) in Taiwan. Four patients from the telbivudine arm discontinued treatment before the end of the 2-year study due to efficacy (achieved HBeAg seroconversion), and 10 patients from the lamivudine arm withdrew before the end of the 2-year study due to efficacy (n = 1) and protocol deviations or adverse events (n = 9). In total, 55 patients in the telbivudine arm and 46 patients in the lamivudine arm completed treatment in the GLOBE study. Of these, 48 patients in the telbivudine arm and 36 patients in the lamivudine arm who had polymerase chain reaction (PCR)-undetectable or PCR-detectable serum HBV DNA levels and did not show genotypic resistance to the two drugs entered the Phase IIIB extension 2303 study. Considering the significant superiority of telbivudine over lamivudine demonstrated by the international GLOBE trial, all patients who entered the 2303 study were treated with telbivudine for an additional two years due to ethical concerns. A total of nine patients withdrew from the study (7 in the telbivudine arm and 2 in the lamivudine-switch arm) due to protocol deviations and/or adverse events. In total, 34 (18 HBeAg-positive and 16 HBeAg-negative) of the 36 patients from the lamivudine-switch arm and 41 (16 HBeAg-positive and 25 HBeAg-negative) of the 48 patients from the telbivudine arm completed the 2303 study. In the present study, the 4-year efficacy and on-treatment durability of the 41 Taiwanese patients from the telbivudine arm of the GLOBE trial who completed the 2303 study were analyzed.

To study the HBV DNA kinetics, the data obtained from all of the 34 HBeAg-positive telbivudine-treated patients completing the 2303 study were analyzed. The baseline characteristics of the Taiwanese patients in the GLOBE trial, 2303 study, and HBV kinetics study are shown in Table 1.

Efficacy measures for the analyses were as described in a previous publication [6]. Standardized tests were performed centrally by Quintiles Transnational (Research Triangle Park, NC). Serum HBV DNA levels were quantified using a COBAS® Amplicor HBV Monitor PCR assay (COBAS-AM assay; Roche Molecular Systems, Pleasanton, CA, USA; detection limit, 300 copies/mL). Informed consent was obtained from each patient enrolled in the study. The study was conducted in compliance with the Declaration of Helsinki and in accordance with Good Clinical Practice guidelines and local regulations.

Because only patients treated with telbivudine were included in this analysis, no statistical comparisons between treatment groups were performed. Analyses of the 4-year efficacy results (HBV DNA levels of <300 log₁₀ copies/mL, alanine aminotransferase (ALT) normalization (the normal range of ALT was below 48 IU/mL in male patients and 37 IU/mL in female patients), HBeAg loss and seroconversion, and resistance) were based on observed data, and missing data were not imputed. The cumulative HBeAg seroconversion rate was defined as the percentage of HBeAg-positive patients (n = 16) with documented seroconversion at any point during the 4-year treatment period, including patients who subsequently seroconverted. Genotypic resistance was defined as the emergence of treatment-associated resistant mutations identified by direct sequencing of the amplified HBV DNA at baseline and from sera of all patients with serum HBV DNA levels of >3 log₁₀ copies/mL at Year 4 (Week 208) [6]. Viral breakthrough was defined as a persistent (two consecutive determinations) on-treatment increase in serum HBV DNA level of >1 log₁₀ copies/mL above the nadir level [12].

The correlations of viral load kinetics (exact reduction in HBV DNA) at Week 24 with PCR negativity, ALT normalization, HBeAg seroconversion, and treatment-emergent resistance at Years 1, 2, 3, and 4 of telbivudine treatment were also evaluated. Patients were categorised into three groups according to serum HBV DNA reduction at treatment Week 24: Group 1, ≤5 log₁₀ vs. >5 log₁₀ copies/mL; Group 2, ≤6 log₁₀ vs. >6 log₁₀ copies/mL; and Group 3, ≤7 log₁₀ vs. >7 log₁₀ copies/mL. Stepwise logistic regression analyses were performed to identify variables associated with treatment outcomes. An odds ratio of >1 indicated a direct relationship, whereas an odds ratio of <1 indicated an inverse relationship [8]. A P value of less than 0.05 was considered statistically significant.

The efficacy results of continuous telbivudine treatment in the 41 compensated CHB Taiwanese patients, the ITT population of the telbivudine arm of the GLOBE trial analysed in this study, were consistent with those of the worldwide GLOBE trial. In the 4-year worldwide GLOBE study (GLOBE study and extension 2303 study) of telbivudine [13], 79% of the HBeAg-positive patients achieved undetectable HBV DNA levels (<300 log₁₀ copies/mL) with 51% HBeAg seroconversion, and 84% of the HBeAg-negative patients achieved undetectable serum HBV DNA levels and 91% achieved ALT normalization at Year 4. In the present subanalysis of the Taiwanese patients, 50% of the HBeAg-positive patients had undetectable serum HBV DNA levels, 56.3% had normalized ALT, and 37.5% and 25% achieved HBeAg loss and seroconversion, respectively, at Year 4. The HBeAg-negative patients achieved better efficacy endpoints: 76% had undetectable serum HBV DNA levels and 83% had normalized ALT at Year 4. In the Taiwanese patients who had undetectable or detectable HBV DNA levels at study entry, genotypic resistance to telbivudine at Year 4 was observed in 43.8% of the HBeAg-positive patients, and 12.5% of the HBeAg-negative patients (Table 2). In Taiwanese treated patients had lower HBeAg seroconversion and higher genotypic resistance rate compare to GLOBE study might be prolonged HBV infection and genotype B and C.

Seven out of the 16 HBeAg-positive patients who achieved undetectable levels of serum HBV DNA at Week 24 sustained this response (100% response rate) at the end of Year 4. Among the HBeAg-negative patients, 16 patients achieved PCR negativity at Week 24, of whom 11 (78.3%) sustained the response at the end of Year 4. Overall, the proportion of patients achieving ALT normalization and HBeAg loss and seroconversion was higher in the subgroup of patients with undetectable serum HBV DNA levels at Week 24 (Table 2). Undetectable HBV DNA levels at treatment Week 24 were also associated with lower rates of genotypic resistance to telbivudine at Year 4 in the HBeAg-positive (0%) and HBeAg-negative patients (8.7%) (Table 2).

The rate of cumulative HBeAg seroconversion increased with continued telbivudine treatment from 12.5% at the end of Year 1 to 31.3% at the end of Year 4. In line with other efficacy results, the rate of seroconversion was higher (28.5% at Year 1 and 57.1% at Year 4) in the subgroup of patients with undetectable serum HBV DNA levels at Week 24 (Figure 1A).

In patients with HBeAg-positive CHB, telbivudine treatment significantly and progressively reduced serum HBsAg levels over a 3-year period in the subanalysis of the worldwide GLOBE study [14]. In the present study, six HBeAg-positive CHB Taiwanese patients were enrolled in an HBsAg sub-study of the 4-year GLOBE study. Of these, five patients showed a rapid decline (>0.5 log₁₀ IU/mL) and one patient showed a slow decline (<0.5 log₁₀ IU/mL) in HBsAg levels at Week 24. Patients who had a rapid decline in HBsAg levels had higher HBsAg loss and seroconversion rates (80% and 60%, respectively) compared to patients with a slow decline (0%), at that time point, their HBV DNA less than 300 copies/mL (Figure 1B). Interestingly, the proportion of patients with a rapid decline in HBsAg levels from baseline to Week 24 was higher in the Taiwanese subanalysis (83.3%) than in the worldwide GLOBE study (33%) [14]. However, HBsAg loss or seroclearance with S-gene mutations has recently been reported [15, 16], and an immune escape-like mutant has been proposed. Further studies are necessary for the simultaneous evaluation of HBsAg-negative and HBV DNA levels.

We investigated the viral load kinetics (exact decrement in HBV DNA levels) at Week 24 and the correlations with long-term efficacy endpoints, namely maintained PCR negativity, ALT normalization, HBeAg seroconversion, and treatment-emergent resistance at Years 1, 2, 3 and 4 of continuous telbivudine treatment. The average baseline HBV DNA level of the 34 HBeAg-positive patients in the correlation studies was ~9.6 log₁₀ copies/mL. Reductions of 5, 6, and 7 log₁₀ copies/mL in HBV DNA levels at Week 24 were compared against post-treatment outcomes at Years 1, 2, 3 and 4. Statistically significant correlations were observed only between HBV DNA reductions of 5 log₁₀ copies/mL and maintained PCR negativity (with the exception of treatment Year 1), and between HBV DNA reductions of 5 log₁₀ copies/mL and lack of resistance. At two years post-treatment, the odds ratios of an HBV DNA reduction of <5 log₁₀ copies/mL at Week 24 for maintained PCR negativity and treatment-emergent resistance were 0.075 and 9.953, with P values of 0.0288 and 0.0219, respectively (P < 0.05) (Table 3). At Years 3 and 4, the odds ratios for maintained PCR negativity were 0.056 and 0.053, with P values of 0.019 and 0.0164, respectively. The odds ratios for resistance at Years 3 and 4 could not be calculated given that the patients who had an HBV DNA reduction of <5 log₁₀ copies/mL at Week 24 developed resistance at Years 3 and 4 post-treatment. The higher cumulative telbivudine resistance rates at Years 1, 2, 3, and 4 in patients with an HBV DNA reduction <5 log₁₀ copies/mL compared to patients with an HBV DNA reduction >5 log₁₀ copies/mL (Table 4) were significantly correlated with HBV DNA reduction levels at Week 24, and further identified an HBV DNA reduction of >5 log₁₀ copies/mL at Week 24 as a good predictor of treatment outcomes at four years of continuous telbivudine treatment.