Einolf, Heidi, Chen, Liangfu, Fahmi, Odette, Gibson, Christopher, Obach, Scott, Shebley, Mohamad, Silva, Jose, Sinz, Michael, Unadkat, Jashvant, Zhang, Lei and Zhao, Ping (2013) Assessing modeling methods for predicting clinical CYP3A induction: A collaborative effort among academic, government regulatory, and pharmaceutical industry scientists. Clinical Pharmacology and Therapeutics, 95 (2). pp. 179-188. ISSN 1532-6535

Abstract

Several drug-drug interaction (DDI) prediction models were evaluated for their capability of identifying drugs with cytochrome P450 (CYP) 3A induction liability based on in vitro mRNA data. These models included basic methods (e.g. Cmax/EC50, Relative Induction Score, and R3 approach) and mechanistic models (e.g. Net Effect and physiologically-based pharmacokinetic models). All methods performed with high fidelity with few, if any, false negatives or positives predicted for induction. The basic methods resulted in no false negatives when total Cmax was incorporated. Mechanistic models that include CYP inactivation had a slightly higher false negative rate, likely due to an over-prediction of the inactivation effect. Based upon this evaluation, a tiered approach using the basic R3 approach for initial DDI risk assessment followed by more mechanistic modeling is recommended. For compounds that are both CYP3A inducers and inactivators, DDI predictions of each mechanism should be considered separately for decisions of clinical trial initiation.

Item Type:	Article
Date Deposited:	12 Oct 2016 00:45
Last Modified:	12 Oct 2016 00:45
URI:	https://oak.novartis.com/id/eprint/8483