Influence of the compound selection process on the performance of human clearance prediction methods

This is a commentary on the series of five manuscripts written as part of the Pharmaceutical Research and Manufacturers of America Clinical and Preclinical Development Committee initiative on predictive models of human pharmacokinetics (PK). In particular, we wish to comment on the third paper in the series, which describes the performance of prediction methods of human clearance (CL). Human CL prediction methods described in the third manuscript are fundamental to the work presented in manuscripts four and five on the prediction of human PK profiles. In this commentary, we examine the influence of the compound selection process by performing a probability analysis and examining the CL properties of compounds that are selected using an idealized drug discovery screening process focused on PK optimization. The results of the analysis suggest that the selection of screening species can influence the performance of various predictive models of human CL.