Evaluation of Adefovir PBPK Model to Assess Biomarker-Informed OAT1 Drug–Drug Interaction and Effect of Chronic Kidney Disease

Abstract

Evaluation of transporter-mediated drug–drug interactions (DDI) with endogenous biomarkers, coupled with physiologically-based pharmacokinetic modeling (PBPK) is envisioned to replace or reduce dedicated DDI clinical trials with clinical probe substrates. The current study developed a PBPK model of adefovir (OAT1 clinical probe) by incorporating experimental measurements of adefovir passive diffusion and OAT1-mediated transport in a mechanistic kidney model for in vitro–in vivo extrapolation of its secretion and renal clearance. The adefovir model was verified with clinical data from nine studies (188 subjects) investigating a range of adefovir intravenous and adefovir-dipivoxil oral doses in White, Japanese, and Chinese populations with healthy renal function. A previously verified probenecid model, incorporating in vivo OAT1/3 inhibitory constant estimated using OAT1/3 endogenous biomarker (4-pyridoxic acid) clinical data, was utilized. The ratio of adefovir maximal plasma concentrations (CmaxR) and area under the curve (AUCR) after mid-to-high single oral doses of probenecid was predicted within the stringent Guest criterion. With the lowest probenecid dose (0.5 g), adefovir AUCR and CmaxR were slightly overpredicted but still within a 1.5-fold error. Application of the adefovir model to patients with severe chronic kidney disease (CKD) accounted for our previously recommended additional 50% decrease in OAT1 activity beyond the decline of glomerular filtration rate. The model successfully predicted 3-fold higher adefovir Cmax and 6-fold higher AUC in patients with severe CKD relative to healthy individuals. This study reinforces the role of PBPK modeling to predict transporter-mediated DDI (coupled with biomarker-informed approach to optimize in vivo inhibitory constant) and the effect of renal impairment on OAT1/3 drugs in lieu of clinical studies.