Association Between Increased Linezolid Plasma Concentrations and the Development of Severe Toxicity in Multidrug-Resistant Tuberculosis Treatment
Johannes Eimer, M. Jachym, Damien Le Dû, Éric Caumes, Najoua El-Helali, Dhiba Marigot-Outtandy, F. Méchaï, Gilles Peytavin, Valérie Pourcher, C. Rioux, Yazdan Yazdanpanah, J. Robert, Lorenzo Guglielmetti, for the LZDM group, Alexandra Aubry, Isabelle Bonnet, F. Morel, Nicolas Véziris, Emmanuel Lecorché, Faïza Mougari, Claire Andréjak, A. Bourgarit, Elise Klement-Frutos, Bénédicte Rivoire, Guillaume Thouvenin, Simone Tunesi, Marie Wicky, Marie Jaspard, Corentine Alauzet, L. Escaut, Sophie Ellis-Corbet, Christine Bernard, Anne-Laure Roux
Abstract
BACKGROUND: Treatment of multidrug-resistant (MDR) tuberculosis with linezolid is characterized by high rates of adverse events. Evidence on therapeutic drug monitoring to predict drug toxicity is scarce. This study aimed to evaluate the association of linezolid trough concentrations with severe toxicity. METHODS: We retrospectively assessed consecutive patients started on linezolid for MDR tuberculosis between 2011 and 2017. The primary outcome was severe mitochondrial toxicity (SMT) due to linezolid, defined as neurotoxicity or myelotoxicity leading to drug discontinuation. The impact of plasma linezolid trough concentrations >2 mg/L was assessed in multivariate Cox proportional hazards models including time-varying covariates. RESULTS: SMT occurred in 57 of 146 included patients (39%) at an incidence rate of 0.38 per person-year (95% confidence interval, .30-.49). A maximum linezolid trough concentration >2 mg/L was detected in 52 patients (35.6%), while the mean trough concentration was >2 mg/L in 22 (15%). The adjusted hazard ratio for SMT was 2.35 (95% confidence interval, 1.26-4.38; P = .01) in patients with a mean trough concentration >2 mg/L and 2.63 (1.55-4.47; P < .01) for SMT after the first detection of a trough concentration >2 mg/L. In an exploratory analysis, higher maximum trough concentrations were dose-dependently associated with toxicity, while lowering elevated trough concentrations did not restore baseline risk. CONCLUSIONS: Linezolid trough concentrations >2 mg/L are strongly associated with the development of severe treatment-emergent toxicity in patients treated for MDR tuberculosis. Pending further prospective evidence, an individual risk-benefit assessment on the continuation of linezolid treatment is warranted in any patient with trough concentrations >2 mg/L.