Predicting in vivo half-life is possible using a function taking microsomal clearance, affinity for human serum albumin, and artificial cell membranes into account. It allows a quick prediction of compound properties during the early phase of medicinal chemistry programs.

A reliable prediction of the in vivo plasma half-life of drug candidates, from easily accessible in vitro assays, is not yet possible. It is surmised that the existing models, which consider protein binding and metabolic clearance, will be improved if they also include the propensity of molecules to bind to cell membranes. To test this hypothesis, a proportionality equation taking these three parameters into account is developed. This article discusses that when this equation can be applied and how reliable its predictions are. It also shows that under controlled conditions, this equation can be used in early medicinal chemistry programs to predict the half-life of novel compounds in rats from in vitro assays.