Abstract

Drug-induced liver injury (DILI) is one of the primary reasons why a new drug candidate may fail during development. To address this challenge, a mathematical representation, DILIsym®, has been developed as a result of an ongoing public-private partnership involving scientists from industry, academia, and the FDA. DILIsym employs mathematical representations of mechanistic interactions and events from drug administration through the progression of liver injury and regeneration to the release of traditional and novel serum biomarkers. The model parameters are varied to recreate population variation in DILI susceptibility. Using in vitro data to represent potential mitochondrial dysfunction, bile acid transporter inhibition, and/or reactive oxygen species generation, DILIsym has been able to predict the in vivo liver safety profile in individuals and in simulated populations for a growing list of drugs. DILIsym is being increasingly used to assist in decision making throughout the development pipeline, from predicting interspecies differences and their hepatotoxicity potential to aiding in the design of dosing regimens to minimize hepatotoxicity when this liability is identified. Furthermore, DILIsym’s incorporation of the release and clearance kinetics of traditional and emerging serum biomarkers can improve interpretation of potential liver safety signals. This chapter outlines the interactions and toxicity mechanisms included in DILIsym and the process of representing a compound in the software. Examples of toxicity profile predictions and biomarker interpretations are included along with future directions for the software.

By Christina Battista, Brett A. Howell, Scott Q. Siler, Paul B. Watkins