Modeling hepatitis D virus kinetics during bulevirtide monotherapy: challenges and solutions

The entry inhibitor Bulevirtide (BLV) was recently approved in Europe and the United States for treatment of chronic hepatitis D virus (HDV) infection, which is considered the most severe form of viral hepatitis infection. It is well established that a model that incorporates free virus and infected cells, but assumes fixed target cell number , is limited to predicting a monophasic viral decline for antiviral agents that act solely to block viral entry or infection. We investigated a recently published fixed target cell model against clinical data from HDV-infected individuals treated with BLV monotherapy for up to 96 weeks using non-linear mixed effects modelling (NLME). We found that although estimated parameters in the fixed target cell model had relative standard errors (RSE) below 50%, suggesting acceptable precision, the model failed to reproduce the non-monophasic HDV kinetic patterns observed in most patients. Consequently, the fixed target cell model led to inaccurate predictions of the treatment duration required to reach a theoretical cure boundary, defined as less than 1 virion in the patient's total extracellular body fluid. Furthermore, the model was unable to account for viral breakthrough, characterized by an initial decline followed by an increase in the virus during therapy, and incorrectly predicted that viral load will remain unchanged after treatment cessation. Lastly, we showed that a model that includes target cell dynamics can explain non-monophasic HDV decline patterns such as biphasic, flat-partial response and viral breakthrough. Including target cell dynamics also predicted a viral rebound once BLV is stopped as observed in clinical studies.