Elsevier and FDA to work on AI algorithm for drug-induced liver injury

Analytics company Elsevier has announced it is working with the FDA on a predictive algorithm to improve the clinical outcomes of drugs in development.

Elsevier is working with the National Center for Toxicological Research (NCTR) at the FDA on a predictive drug-induced liver injury (DILI) algorithm. The company started the project by integrating the rule-of-two model and preclinical data from the literature and the summary basis of FDA and European Medicines Agency (EMA) approval documents collected in Elsevier’s PharmaPendium database.

DILI is one of the two most frequent causes for drug withdrawals and is a leading cause of attrition of compounds in drug development. More so, an analysis of over 500 withdrawals shows that DILI is the leading cause for toxicity-related drug withdrawals. In the United States DILI is also a leading reason for regulatory actions involving investigational and approved medications and a leading cause of acute liver failure.

A major problem in drug development is the frequency of adverse hepatic reactions induced by new molecular entities (NMEs). Mitigating DILI risk as early as possible during drug development can lower R&D costs and more importantly, improve patient safety, particularly in oncology.

Elsevier intends to use its algorithm to assess DILI in the preclinical phase of drug development and it could be utilised by the FDA when toxicology issues arise during different stages of the regulatory review process.  

“The predictive toxicology space is becoming a growing area of interest as we seek to understand how to improve clinical outcomes without risking the safety of patients,” said Guenther Kurapkat, senior vice president, Life Science Solutions, Elsevier. “This (Collaborative Research and Development Agreement) CRADA will strengthen the collaborative relationship between Elsevier and the FDA and paves a path forward for future projects that seek to address key pain points for the industry, which continues to be challenged with rising R&D costs and low success rates.”