A Pragmatic PBPK-Driven Strategy to Guide ADME Characterization and Early Human PK Prediction for ECCS Class 1B/3B Compounds

Predicting human pharmacokinetics (PK) for ECCS class 1B/3B compounds often demands resource-intensive experiments and empirical scaling, especially when transporter-mediated processes dominate. We propose a tiered physiologically based pharmacokinetic (PBPK) framework for early discovery that (i) aims to identify which ADME processes warrant characterization by chemical class and (ii) provides guidance for human PK prediction with either minimal inputs (zwitterions) or targeted measurements (acids). Benchmarking against well-described molecules (Fexofenadine, Valsartan, Rosuvastatin, and the active metabolite of Valategrast) supports each tier. For the zwitterions we investigated allometrically scaled rat biliary/renal or nonhuman primate plasma clearance alone predicted human PK within 2-fold. For the acids, adjusting the tissue partition coefficient (Kp) was required to recapitulate plasma profiles. This structured PBPK strategy helps to identify essential experiments for acidic compounds and enables reliable early human PK predictions for zwitterions, contributing to a more informed screening cascade design and ultimately accelerating discovery cycle times.