A critical assessment of modeling safety-related drug attrition

Abstract

This paper challenges the general desire to find simple rules and guidelines to reduce attrition due to toxicity and clinical safety. We present an analysis of 150 AstraZeneca development compounds and evaluate some of the published guidelines and their ability to identify compounds with safety liabilities. Interestingly, none of the current guidelines were able to discriminate compounds that successfully reached Phase II. The analysis was extended to recently approved drugs (2009–2011) and we found that a large portion did not comply and would never have reached patients if such guidelines had been applied at an early stage.