Development of Pharmaceutical QSAR Models for Regulatory Purposes

Kevin Cross and Glenn Myatt, Leadscope Inc., USA
Robert Daniel Benz, US FDA, USA

The draft regulatory guidance that provides for using QSAR models to predict the toxicity of drug impurities as an alternative to Salmonella mutagenicity testing represents a breakthrough in the acceptance of QSAR by the U.S. Food and Drug Administration's Center for Drug Evaluation and Research (FDA/CDER). Such use requires validation of QSAR models for predicting the toxicities of drugs and their impurities that may or may not be drug-like. In particular, it requires the external validation of the reliability of QSAR models used for genetic toxicology predictions. This presentation focuses on our experience in developing QSAR models for regulatory assessment of drugs and drug impurities by FDA/CDER, and how this process differs from the development of QSAR models for drug discovery and development by and for the pharmaceutical industry.