Drug-induced liver injury (DILI) is rare in clinical practice but when it occurs it can lead to acute liver failure and death. Drug developers and regulators undertake a series of steps to identify the DILI potential of a medication before it is approved for marketing. Preclinical testing by drug developers typically involves a multitude of in vitro assays and in vivo animal experiments before a compound is moved into first-in-human phase 1 testing. Over the last two decades, there have been a number of advances in preclinical screening for DILI potential of a new chemical entity, but these approaches tend to be overly sensitive with insufficient positive predictive value. Once in clinical trials, the DILI potential of an investigational agent and risks to a participant are carefully managed through patient selection, DILI monitoring paradigms, and drug interruption and discontinuation criteria, in close concert with the regulators. Recent developments in Quantitative Systems Toxicology offer promising and complementary in silico approaches to predict the compound’s risk for DILI via multifaceted systems biology. When a drug developer submits a New Drug Application (NDA) for marketing approval, regulators review the preclinical and clinical trial data in a structured fashion to assess the DILI risk. While these approaches have been successful in dramatically reducing the marketing approval of medications eventually associated with hepatotoxicity, many challenges remain in identifying the risk for DILI during preclinical and early-to-late clinical development stages for genetic medicines, biological agents, and immunotherapies. In this review, we discuss current preclinical, in-silico, and clinical development approaches to screen for DILI potential of an investigational agent and provide a high-level description of regulators’ approach for assessing DILI risk in an NDA.

By  Naga Chalasani, Paul H Hayashi, Debra Luffer-Atlas, Arie Regev, Paul B. Watkins