Prediction of human oral pharmacokinetics using nonclinical data: examples involving four proprietary compounds

Abstract

The oral pharmacokinetics (concentration-time profile) of four proprietary compounds in humans were predicted using the C(vss)-MRT method. The first step was to demonstrate superposition of intravenous (i.v.) pharmacokinetic profiles of preclinical species following mathematical transformation of their respective concentration-time curves using the corresponding C(vss) (where C(vss)=dose/Vss; Vss is the volume of distribution at steady state) and mean residence time (MRT) values. The resultant profiles were then back-transformed to estimate human i.v. plasma concentration-time profiles using human C(vss) and MRT values. Human C(vss) and MRT values were estimated from projected human Vss and CL values. Projection of CL was based on scaled (in vitro) metabolic clearance, simple allometry with and without various correction factors and the unbound fraction corrected intercept method. Vss values were estimated by allometric scaling with and without correction for interspecies differences in plasma protein binding. The predicted human i.v. profiles, in combination with the estimated mean absorption rate constants and bioavailability, were then used to simulate the oral pharmacokinetics in human using one- or multi-compartment kinetic models. Overall, with this approach, key oral pharmacokinetic parameters such as AUC, C(max), C(min) and oral plasma T((1/2)) were projected to be within two-fold of the actual values in humans.