Shuangmin Ji (1), Chenhui Deng (2), Kehua Wu (1), Liang Li (1), Tianyan Zhou (1), Mats Karlsson (2), Wei Lu (1)
(1) School of Pharmaceutical Sciences, Peking University, Beijing, China; (2) Pharmacometrics Research Group, Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Objectives: The objectives of this study were to (1) develop a semi-mechanistic model to characterize the dynamics of viral DNA load in hepatitis B virus (HBV) infected patients who had received treatment of entrcavir (ETV) or adefovir (ADV), (2) to describe the life cycle of HBV for efficacy comparison of two drugs, and (3) to identify potential covariates of virus kinetics.
Methods: Publications were retrieved from PubMed database according to the pre-defined inclusion and exclusion criteria. Mean and individual data of HBV DNA after start of treatment were analyzed using a nonlinear mixed-effects modeling (NONMEM) approach and first-order conditional estimation with interaction (FOCE-I) laplacian method. The significance of potential covariates was evaluated. Variance of estimates and model robustness were assessed using sampling importance resampling (SIR) and individual visual predictive check (VPC) approaches.
Results: A three-compartment model, including target hepatocytes compartment, infected hepatocytes compartment and virus load compartment, was built to describe HBV DNA kinetics after administration of ETV or ADV, which is first and second line treatment, respectively. A density-dependent proliferation of hepatocytes is assumed in this model, and part of the healthy hepatocytes is assumed not to be infected . The death rate of infected hepatocytes for ETV and ADV is 0.0428 and 0.0311 per day respectively, almost 10 times of the healthy hepatocytes (0.004 per day). HBV DNA was significantly decreased by both treatments. ED50 of ETV in patients with or without prior treatments was 0.0055 mg and 0.0015 mg, respectively.
Conclusions: Model-based meta-analysis is useful to compare different drugs in the same class or identify significant covariates by pooling literature data and building pharmacological models. Further development of this viral kinetic model may provide information about drug resistance and therefore improve understanding of this disease.
Acknowledgement:
We thank Gailing Li, Jia Ji and ShanShan Bi for their support and contributions to this project.
Reference: PAGE 24 () Abstr 3437 [www.page-meeting.org/?abstract=3437]
Poster: Drug/Disease modeling - Infection