The predictivity of the toxicity of pharmaceuticals in humans from animal data--an interim assessment.

This project was undertaken by the International Life Sciences Institute's Health and Environmental Sciences Institute (ILSI-HESI) to develop an improved understanding of the extent to which various types of human toxicities (HTs) manifested during clinical trials could be predicted from standard toxicology studies. A multi-company database of 131 pharmaceutical agents to-date was based on compounds with one or more demonstrated HTs identified during clinical development. These interim results support a true positive prediction rate of animal models for human toxicity of 69%, and also that study results from non-rodent (dog, primate) species have good potential to identify HTs from many therapeutic classes. The continuing assessment of a larger database may have impact on the identification of new toxicology methodologies, and may lead to optimization of non-clinical study designs and improved assessments.