Unexpected Clinical Drug-Drug Interaction of Fexinidazole on Midazolam Pharmacokinetics: Insights into Underlying Mechanisms from Clinical Phase I Study Results and Supporting In Vivo/Vitro Evidence.

Abstract

Background: Fexinidazole, a nitroimidazole antiparasitic, has been approved to treat human African trypanosomiasis (HAT) worldwide. In vitro studies have shown that fexinidazole inhibits and weakly induces CYP3A4/5. In silico predictions indicated that fexinidazole, which has a significant intestinal and liver first-pass metabolism, could increase the exposure of a sensitive probe substrate of CYP3A4 by two-fold. Therefore, this study investigated the potential clinical drug-drug interactions (DDIs) of fexinidazole with CYP3A4 substrates.

Objective: To assess the effect of fexinidazole on the pharmacokinetics of midazolam, a well-recognised sensitive CYP3A4 substrate (and its metabolites 1-hydroxy-midazolam and N-glucuronide-midazolam) in humans and to elucidate the underlying mechanism of the in vivo DDI.

Methods: This was a phase I, open-label, single-centre, non-randomised, single-sequence, two-period, two-treatment crossover study. The study population consisted of 12 healthy male and female participants. The two treatment periods included Period 1, wherein a single midazolam dose was administered on Day 1, and Period 2, wherein fexinidazole was administered once daily from Day 1 to Day 5, with a single midazolam dose co-administered on Day 4. Key pharmacokinetic parameters of midazolam and its main metabolites, including the maximum plasma concentration (C_max), area under the curve (AUC), and elimination half-life (t_1/2z), were evaluated. Additionally, in vitro assessments (protein-binding and CYP enzyme induction studies) were conducted to investigate potential mechanisms contributing to the observed interaction.

Results: Contrary to the in vitro predictions, fexinidazole significantly reduced midazolam exposure in vivo, resulting in a reduction of 39% in C_max, 57% in AUC, and 33% in t_1/2z, without significant changes in t_max. Mechanistic studies ruled out reduced absorption and plasma protein displacement as potential causes. At clinically relevant concentrations, fexinidazole and M1 exhibited weak induction potential on CYP3A4/5 and no significant induction on other enzymes. Further, in vivo investigations on midazolam metabolites confirmed that CYP3A4/5 induction by fexinidazole was the primary mechanism, increasing the first-pass metabolism and clearance of midazolam. The metabolic ratios of 1-hydroxy-midazolam and N-glucuronide-midazolam were increased by 1.63-fold and 1.24-fold, respectively. Steady-state exposures of fexinidazole and its metabolites M1 and M2 were consistent with those previously assessed in other clinical studies.

Conclusion: While in vitro studies showed weak induction by fexinidazole and its metabolite M1, the clinical pharmacokinetic data provided stronger evidence, supporting the conclusion that fexinidazole is a moderate inducer of CYP3A4/5 in vivo. Thus, it is suggested to update the product information by including the potential for CYP3A4/5 induction in vitro, removing the risk of CYP3A4 inhibition in vivo, and adding clinical interaction data highlighting the risk of induction on drugs predominantly metabolised by CYP3A4/5.

Trial registration: EudraCT No.: 2021-004533-36.