Ethionamide population pharmacokinetics/pharmacodynamics and therapeutic implications in South African adult patients with drug-resistant tuberculosis.

Introduction: Ethionamide is part of the drug-resistant tuberculosis regimen whose pharmacokinetic (PK) and pharmacodynamic (PD) information is limited. The aim of the study was to describe the PK and simulate doses to assess PD attainment.

Methods: This was an observational population PK study of patients admitted for drug-resistant tuberculosis at a hospital in South Africa. Nonlinear mixed-effects modelling implemented in Monolix 2019R2 was used to estimate population pharmacokinetic parameters. We performed Monte Carlo simulations to assess and optimise the dose regimen. The target C range was 2.5-5 μg/mL, which is within the minimum inhibitory concentration (MIC) range. The target AUC was 140.5 μg*h/mL, which corresponds to the PK/PD target ratio AUC /MIC of 56.2.

Results: A one-compartment pharmacokinetic model with a lag-time, first-order absorption and elimination best described the PK of ethionamide. The lag-time, absorption rate constant (ka), volume of distribution (V/F) and clearance (Cl/F) were 0.66 hours, 0.434 h , 180 L and 99.5 L/h, respectively, for a typical individual weighing 52.6 kg. Between-subject variability in lag-time, ka, V/F and Cl/F were 38%, 92%, 168% and 120%, respectively. Simulation of the recommended doses of 15-20 mg/kg, 500 mg, 750 mg and 1000 mg for patients in the weight bands <33, 33-50, 51-70 and >70 kg resulted in <17% and 3% of the patients achieving the target C and AUC , respectively.

Conclusion: There is high variability in ethionamide PK and very few patients attain the desired target exposure at standard or optimised doses. We propose individualised dose regimen optimisation.