OBJECTIVETo investigate whether quantifiable changes in serum levels of hepatitis B e antigen (HBeAg) in response to 24 weeks of pegylated-interferon alfa-2a (Peg-IFN-a 2a) treatment are predictive of therapeutic efficacy at 48 weeks of treatment in HBeAg-positive chronic hepatitis B (CHB) patients and to investigate the efficacy of using an individualized antiviral treatment strategy.

METHODSNinety-six HBeAg-positive CHB patients with detectable HBeAg at week 24 of Peg-IFN-a 2a treatment were categorized according to the quantitative change in HBeAg (vs. pre-treatment baseline): group A, HBeAg decline more than 2 log; group B, HBeAg decline between 1 - 2 log; group C, HBeAg decline less than 1 log, which was then randomly divided into two sub-groups: C1 and C2. Group A, B, and C1 patients continued the original therapy for an additional 24 weeks, while group C2 patients were supplemented with lamivudine (3TC + Peg-IFN-a 2a) for the additional 24 weeks of treatment. All patients underwent liver biopsy at the end of treatment (week 48), and HBV covalently-closed circular (ccc)DNA was quantified as a measure of therapeutic efficacy. A, B, and C1 between-group multiple comparisons were made by the Nemenyi test; C1 and C2 between-group comparison was made by the Mann-Whitney U test. The significance of between-group differences in decreased HBV cccDNA vs. HBeAg/anti-HBe seroconversion was made by the Chi-squared test.

RESULTSAt week 48, the mean decrease of serum HBV cccDNA in each group was: A, 5.8 log10 copy/ml; B, 3.8 log10 copy/ml; C1, 2.8 log10 copy/ml; C2, 5.7 log10 copy/ml. Statistically significant differences were observed for group A vs. B and C1 (P less than 0.01) and C1 vs. C2 (P less than 0.01); however, the difference between group B and C1 did not reach statistical significance (P = 0.19). The mean decrease of HBeAg in each group was: A, 2.7 log10 S/CO; B, 1.9 log10 S/CO; C1, 0.9 log10 S/CO; C2, 1.6 log10 S/CO. Statistically significant differences were observed for group A vs. B and C1 (P less than 0.01) and C1 vs. C2 (P less than 0.01). The rate of patients who achieved undetectable HBV DNA in each group was: A, 87.5%; B, 34.5%; C1, 17.4%; C2, 85.0%. Statistically significant differences were observed for group A vs. B and C1 (P less than 0.01) and C1 vs. C2 (P less than 0.01). The HBeAg seroconversion rates were: A, 75.0%; B, 24.1%; C1, 13.0%; C2, 25.0%. Statistically significant differences were observed only for group A vs. B and C1 (P less than 0.01). Finally, group A achieved greater reduction in levels of cccDNA in liver tissues than B or C1 (P less than 0.01); however, the differences between B and C1 and between C1 and C2 did not reach statistical significance.

CONCLUSIONCHB patients who showed an HBeAg decline of more than 2 log at week 24 of Peg-IFN-a 2a treatment had better treatment outcome at week 48 than those who showed HBeAg decline less than 2 log at week 24. Augmenting the Peg-IFN-a 2a treatment with 3TC can improve the clinical response. A change of quantifiable HBeAg at week 24 of Peg-IFN-a 2a treatment may be a useful predictor of therapeutic efficacy of a 48-week antiviral regimen.