Population pharmacokinetics and dose evaluations of linezolid in the treatment of multidrug-resistant tuberculosis

Abstract

Background: The pharmacokinetic/pharmacodynamics (PK/PD) target derived from the hollow-fiber system model for linezolid for treatment of the multidrug-resistant tuberculosis (MDR-TB) requires clinical validation. Therefore, this study aimed to develop a population PK model for linezolid when administered as part of a standardized treatment regimen, to identify the PK/PD threshold associated with successful treatment outcomes and to evaluate currently recommended linezolid doses. Method: This prospective multi-center cohort study of participants with laboratory-confirmed MDR-TB was conducted in five TB designated hospitals. The population PK model for linezolid was built using nonlinear mixed-effects modeling using data from 168 participants. Boosted classification and regression tree analyses (CART) were used to identify the ratio of 0- to 24-h area under the concentration-time curve (AUC0-24h) to the minimal inhibitory concentration (MIC) threshold using the BACTEC MGIT 960 method associated with successful treatment outcome and validated in multivariate analysis using data from a different and prospective cohort of 159 participants with MDR-TB. Furthermore, based on the identified thresholds, the recommended doses were evaluated by the probability of target attainment (PTA) analysis. Result: Linezolid plasma concentrations (1008 samples) from 168 subjects treated with linezolid, were best described by a 2-compartment model with first-order absorption and elimination. An AUC0–24h/MIC > 125 was identified as a threshold for successful treatment outcome. Median time to sputum culture conversion between the group with AUC0-24h/MIC above and below 125 was 2 versus 24 months; adjusted hazard ratio (aHR), 21.7; 95% confidence interval (CI), (6.4, 72.8). The boosted CART-derived threshold and its relevance to the final treatment outcome was comparable to the previously suggested target of AUC0–24h/MIC (119) using MGIT MICs in a hollow fiber infection model. Based on the threshold from the present study, at a standard linezolid dose of 600 mg daily, PTA was simulated to achieve 100% at MGIT MICs of ≤.25 mg which included the majority (81.1%) of isolates in the study. Conclusion: We validated an AUC0–24h/MIC threshold which may serve as a target for dose adjustment to improve efficacy of linezolid in a bedaquiline-containing treatment. Linezolid exposures with the WHO-recommended dose (600 mg daily) was sufficient for all the M. tb isolates with MIC ≤.25 mg/L.