Diagnostic accuracy of a liquid chromatography-tandem mass spectrometry assay in small hair samples for rifampin-resistant tuberculosis drug concentrations in a routine care setting

Abstract Background Treatment monitoring of drug-resistant tuberculosis (DR-TB) in resource-limited settings is challenging. We developed a multi-analyte assay for eleven anti-TB drugs small hair samples as an objective metric drug exposure. Methods Small were collected from participants at various timepoints during directly observed RR-TB treatment inpatient tertiary referral facility South Africa (DR-TB cohort). assessed qualitative determination (i.e., detection above limit detection) bedaquiline, linezolid, clofazimine, pretomanid, levofloxacin, moxifloxacin, pyrazinamide, isoniazid, ethambutol, ethionamide, and prothionamide LC-MS/MS index panel against reference standard records. Because regimens prior to hospitalization not available, we also analyzed specificity (for all except isoniazid) using external cohort HIV-positive patients treated latent TB infection with daily isoniazid (HIV/LTBI cohort) Uganda. Results Among the 57 DR-TB (58% pre-XDR/XDR-TB; 70% HIV-positive) contributing analyzable samples, sensitivity investigational was 94% or higher ethionamide (58.5, 95% confidence interval [CI], 40.7–99.9). Assay low across tested analytes within cohort; conversely, 100% HIV/LTBI cohort. Conclusions Hair concentrations reflect long-term exposure, multiple successive commonly employed may result apparent false-positive falsely elevated quantitative levels when histories growth window are known.