Abstract

Drug-induced liver injury (DILI) is complicated and difficult to predict. It has been observed that drugs with extensive hepatic metabolism have the higher likelihood of causing DILI. Cytochrome P450 (CYP) enzymes are primarily involved in hepatic metabolism. Identifying the associations of DILI with drugs which are CYP substrates, inhibitors, or inducers will be extremely helpful to a clinician in decision-making process when dealing with a patient suspected of having DILI. We collected the metabolism data on CYP enzymes for 254 orally administered drugs in Liver Toxicity Knowledge Base Benchmark Dataset with a known daily dose, and applied logistic regression to identify these associations. We revealed that drugs which are the substrates of CYP enzymes would have the higher likelihood of causing DILI (odds ratio (95% confidence interval) [OR (95% CI)]: 3.99 (2.07-7.67); p < 0.0001), which is dose-independent, and drugs which are the CYP inhibitors would have the higher likelihood of generating DILI only when they are administered at a high daily dose (OR (95% CI): 6.03 (1.32-27.5); p = 0.0098). However, drugs which are the CYP inducers are not observed to be associated with DILI (OR (95% CI): 1.55 (0.65-3.68); p = 0.3246). Our findings will be not only useful to identify suspect medication as a cause of liver injury in clinical settings where the patients are treated with comedications, but also for personalized medicines to understand the individual susceptibility of DILI.