Drug-induced liver injury (DILI) is a major challenge for the pharmaceutical industry, regulatory authorities, and clinicians. It is usually categorized into 'intrinsic' and 'idiosyncratic', but DILI caused by most drugs is of an idiosyncratic nature and usually cannot be predicted from the regulatory required animal toxicity studies. Unfortunately, some individuals exposed to therapeutic dose will develop idiosyncratic DILI that might involve severe clinical outcome, and no biomarker is available to identify the susceptible patients prior to drug treatment. In this editorial, we summarized the recent advances in predicting idiosyncratic DILI and provided the perspectives to improve the prediction.
Keywords: HLA; data integration; drug safety; idiosyncratic drug-induced liver injury; predictive model.