Background: Early identification of non-responders to interferon-alpha and development of stopping rules are needed in patients with chronic hepatitis B to reduce treatment-related costs and morbidity.
Methods: In total, 47 patients with hepatitis B e antigen (HBeAg)-positive chronic hepatitis B received pegylated interferon-alpha2b for 8 weeks, lamivudine plus pegylated interferon-alpha2b combination therapy for 24 weeks and lamivudine monotherapy for 28 weeks. Sustained virological response was defined as HBeAg seroconversion and hepatitis B virus (HBV) DNA < 10(5) copies/ml at the end of treatment and after 52 weeks of follow-up. The early HBV DNA data from the first 12 weeks of therapy were fitted by a viral kinetic model.
Results: Cutoff values for prediction of sustained virological response were defined as a rate of infected cell loss delta > or = 0.005 per day (negative predictive value [NPV] 100% and positive predictive value [PPV] 33.3%) and log values of the area under the mathematically predicted HBV DNA curve between baseline and week 12 of therapy < or = 8.9 log10 copies/ml x days (NPV 100% and PPV 50%). By the latter cutoff, 25/36 (69.4%) patients without sustained virological response could be identified after 12 weeks of therapy.
Conclusions: In the present study, mathematical modelling of viral dynamics allowed prediction of sustained virological response after 12 weeks of therapy. Virodynamic predictors for sustained virological response should be further validated. The area under the mathematically predicted HBV DNA curve seems a promising candidate for potential cutoff values as it summarizes the influence of baseline HBV DNA and treatment effects.