Predicting Systemic and Liver Bosentan Exposure Using Physiologically-Based Pharmacokinetic Modeling

Bosentan is the first approved oral medication for pulmonary arterial hypertension, yet the black-box warning on its labeling implies a substantial risk of liver injury associated with bosentan exposure. The risk assessment of bosentan-induced liver injury requires a thorough understanding of the underlying mechanisms, for which there is accumulating evidence. Integrating these mechanisms with clinical liver bosentan concentration would enable a more dynamic and relevant risk assessment. This study designed a workflow of physiologically-based pharmacokinetic (PBPK) model development to capture bosentan's hepatic disposition and predict the (intra)hepatic bosentan exposure. Specifically, clinical plasma and excretion data of bosentan were used to minimize the uncertainty in estimating the hepatic clearance. The model predictions were well overlapped with observations in the systemic circulation and excretion. Furthermore, the model-derived intrinsic hepatic clearance was comparable with the one derived from a clinical study. These results reflected confidence in the model's capability to predict hepatic bosentan exposure. The model-simulated steady-state unbound exposure to bosentan in hepatocytes and liver tissue ranged from 1.65 to 34.1 ng/mL following twice-daily 125-mg oral doses. The ratio of the simulated unbound concentration between the liver matrices and systemic plasma was between 0.80 and 2.93 across the therapeutic dosing regimens. In summary, a bosentan PBPK model was successfully developed with the designed workflow and was able to predict the hepatic disposition of bosentan. The developed model can be applied to generate hepatic bosentan exposure that bridges the toxicological mechanistic findings from in vitro to in vivo, assisting in risk assessment.