Use of in vitro and in vivo data to estimate the likelihood of metabolic pharmacokinetic interactions

This article reviews the information available to assist pharmacokineticists in the prediction of metabolic drug interactions. Significant advances in this area have been made in the last decade, permitting the identification in early drug development of dominant cytochrome P450 (CYP) isoform(s) metabolising a particular drug as well as the ability of a drug to inhibit a specific CYP isoform. The major isoforms involved in human drug metabolism are CYP3A, CYP2D6, CYP2C, CYPIA2 and CYP2E1. Often patients are taking multiple concurrent medications, and thus an assessment of potential drug-drug interactions is imperative. A database containing information about the clearance routes for over 300 drugs from multiple therapeutic classes, including analgesics, anti-infectives, psychotropics, anticonvulsants, cancer chemotherapeutics, gastrointestinal agents, cardiovascular agents and others, was constructed to assist in the semiquantitative prediction of the magnitude of potential interactions with drugs under development. With knowledge of the in vitro inhibition constant of a drug (K-i) for a particular CYP isoform, it is theoretically possible to assess the likelihood of interactions for a drug cleared through CYP-mediated metabolism For many agents, the CYP isoform involved in metabolism has not been identified and there is substantial uncertainty given the current knowledge base. The mathematical concepts for prediction based on competitive enzyme inhibition are reviewed in this article. These relationships become more complex if the inhibition is of a mixed competitive/noncompetitive nature. Sources of uncertainty and inaccuracy in predicting the magnitude of in vivo inhibition includes the nature and design of in vitro experiments to determine K-i, inhibitor concentration in the hepatic cytosol compared with that in plasma, prehepatic metabolism, presence of active metabolites and enzyme induction. The accurate prospective prediction of drug interactions requires rigorous attention to the details of the in vitro results, and detailed information about the pharmacokinetics and metabolism of the inhibitor and inhibited drug. With the discussion of principles and accompanying tabulation of literature data concerning the clearance of various drugs, a framework for reasonable semiquantitative predictions is offered in this article.