A drug with a dose of 100 mg or greater and high lipophilicity is significantly more likely to lead to drug-induced liver injury, according to a recent study.
Researchers evaluated the daily doses and lipophilicity of 164 FDA-approved, orally administered drugs from the liver toxicity knowledge base benchmark data set. The analysis included medications considered most (n=116) and least (n=48) likely to cause drug-induced liver injury (DILI). Drugs most likely to cause DILI included a black-box warning or had been withdrawn from the market, while those least likely had no DILI description in their labeling.
Hepatotoxicity was significantly more common among drugs administered at daily doses of 100 mg or more and with octanol-water partition coefficients of 3 or higher (OR=14.05; 95% CI, 3.42-126.85 vs. OR=6.92; 95% CI, 3.1-15.63 for high daily dose alone). These criteria defined a “rule-of-two” that the researchers then applied to an independent set of 179 orally administered medications, including 115 hepatotoxic positives and 64 negatives.
Within this data set, 85% of drugs that fulfilled the rule-of-two were hepatotoxic, compared with 59% of rule-of-two negatives (OR=3.89; 95% CI, 1.44-11.1). Application of the rule resulted in a lower number of false positives than the criteria of a 100-mg dose or greater alone (six vs. 30), but a higher number of false negatives (71% vs. 35%).
When all evaluated medications were divided by therapeutic categories within the WHO Anatomical Therapeutic Chemical classification system, applying the rule resulted in a correct classification rate of 0.6 or better in seven categories.
Investigators also applied the rule to five pairings of clean and toxic drugs of similar chemical structure and primary target activity. The rule identified the toxic drug in two of the pairings, both of which belonged to high-confidence therapeutic categories. Finally, in six evaluated case studies, only drugs that fulfilled the rule resulted in severe liver injury.
“Predicting a complex clinical endpoint such as DILI doesn’t have to use a complicated approach,” researcher Weida Tong, PhD, director of the Bioinformatics and Biostatistics division of the National Center for Toxicological Research/FDA, told Healio.com. “The simplicity of the proposed ‘rule-of-two’ can be readily applied in drug development and guide clinical practice.”