Modeling covalently closed circular DNA dynamics in assessing chronic hepatitis B prognosis.

The persistence of covalently closed circular DNA (cccDNA) in the nuclei of HBV-infected hepatocytes plays a critical role in the pathogenesis of chronic hepatitis B (CHB), and hepatitis B surface antigen (HBsAg) levels can be considered a surrogate marker for cccDNA quantification. In this paper, a mathematical model is proposed to mimic cccDNA kinetics in infected hepatocytes, and the basic reproduction rate of cccDNA is obtained. It is found that the backward bifurcation occurs. Importantly, the model predictions match well the clinical data from 96 newly treated CHB patients, collected at the authors' hospitals. Specifically, 18 patients (23.96 %) achieved HBsAg serologic negative conversion (SNC), excluding 5 patients with serologic relapse. Ten patients were predicted to be uncertain, including two clinically confirmed patients with SNC. Excluding the uncertain patients, our model gives a concordance rate of 93.75 % (15/16). Our results suggest that the baseline HBsAg, HBV DNA and hepatitis e surface antigen statuses are the major factors related to the accuracy of the model for the prediction of negative conversion. Therefore, our model is helpful to design effective clinical withdrawal indicators.