Might isoniazid plasma exposure be a valuable predictor of drug-related hepatotoxicity risk among adult patients with TB?

Objectives: To investigate the relationship between isoniazid plasma exposure and the likelihood of elevation of ALT (≥51 IU/L) among adult patients with TB. Methods: A retrospective observational study was conducted in patients who underwent periodic monitoring of hepatic function and in whom pharmacokinetic data were collected. Monte Carlo simulation was performed with the intent of identifying the probability of achieving an AUC24 greater than the identified threshold of hepatotoxicity with different dosing regimens (2.5, 5.0 and 7.5 mg/kg/day). Results: Forty-one out of 185 evaluable patients (22.2%) had an ALT elevation. A mild correlation between isoniazid AUC0-24and ALT increase was observed (Spearman's ρ=0.34, P<0.001). Patients with ALT ≥51 IU/L showed significantly higher isoniazid exposure than those with ALT <51 IU/L (mean AUC24 of 58.33 versus 31.28 mg.h/L, P<0.001). The probabilities of ALT elevation were 0.82 and 0.12 for isoniazid AUC24 ≥55.0 and <55.0 mg.h/L, respectively. Use of a logistic regression model estimated a likelihood of developing hepatotoxicity of 0.5 and 0.9 when in the presence of an isoniazid AUC24 of 53.7 and 70.0 mg.h/L, respectively. Simulation showed that the standard isoniazid 5 mg/kg daily dose gave a probability of ALT increase of 0.46 for slow acetylators and 0.03 for rapid acetylators. Conclusions: Plasma isoniazid exposure might be a valuable predictor of drug-related hepatotoxicity. Early assessment of isoniazid exposure at the beginning of treatment might allow prompt dosage reduction among those patients who are experiencing drug overexposure, thus containing the risk of hepatotoxicity occurrence.