Assessing the Effect of Organic Anion Transporter Inhibition on Circulating Pyridoxic Acid, an Endogenous Transporter Biomarker, using Physiologically‐based Pharmacokinetic Modeling

Abstract

Transporter‐mediated renal secretion plays a critical role in the systemic clearance of ~30% of the Food and Drug Administration (FDA)‐approved drugs and is attributed to clinically significant drug‐drug interactions (DDIs). For example, organic anion transporters (OATs) are inhibited by perpetrator drugs (e.g., probenecid), leading to higher circulating levels of OAT substrates (e.g., furosemide).1 Recently, OAT inhibition has been shown to be associated with increased plasma levels of endogenous substrates, e.g., pyridoxic acid (PDA) and homovanillic acid, which can serve as potential OAT biomarkers.2 The objective of our study was to develop and validate a physiologically‐based pharmacokinetic (PBPK) model of PDA to allow prospective evaluation of the magnitude of potential DDI of an OAT inhibitor‐substrate pair. Probenecid and furosemide were used as selective inhibitor and substrate of OATs, respectively. Physicochemical properties and in vitro data pertaining to the pharmacokinetics of probenecid and PDA were collated from literature. A PBPK model was first constructed to predict the plasma concentration‐time profiles of probenecid after oral administration using Simcyp (v20). Then, utilizing the fraction secreted versus filtered as well as the unbound fraction (fu = 0.092) data of PDA,2 the total elimination and synthesis rates of PDA at steady‐state were used to develop a baseline PBPK model. The effect of OAT inhibition by probenecid on the PDA levels was estimated and the predicted area under the plasma concentration‐time profile (AUC), change in clearance, and the highest plasma concentration (Cmax) values were compared with the observed data.2 The model was able to capture the changes within 2‐fold of the clinical data. Using the predicted AUC profile of PDA and assuming 100% inhibition at Cmax, the magnitude of inhibition was estimated across different time points. The biomarker‐informed magnitude of inhibition showed significant correlation (r2 = 0.9336) with the observed change in furosemide concentrations in the presence of probenecid (Fig. 1). PDA‐informed PBPK model developed here can be used to prospectively predict the magnitude of OAT inhibition, which is particularly applicable to assessing DDI potential of investigational drugs or prescribed drugs in special populations with sparse samples, without the requirement of an exogenous probe substrate.