Evaluating culture-free targeted next-generation sequencing for diagnosing drug-resistant tuberculosis: a multicentre clinical study of two end-to-end commercial workflows.

Autor:	Colman RE; FIND, Geneva, Switzerland; Division of Pulmonary, Critical Care, Sleep Medicine, and Physiology, University of San Diego, San Diego, CA, USA. Electronic address: rcolman@health.ucsd.edu., Seifert M; FIND, Geneva, Switzerland; Division of Pulmonary, Critical Care, Sleep Medicine, and Physiology, University of San Diego, San Diego, CA, USA., De la Rossa A; FIND, Geneva, Switzerland., Georghiou SB; FIND, Geneva, Switzerland., Hoogland C; FIND, Geneva, Switzerland., Uplekar S; FIND, Geneva, Switzerland., Laurent S; FIND, Geneva, Switzerland., Rodrigues C; Hinduja Hospital and Medical Research Centre, Mumbai, India., Kambli P; Hinduja Hospital and Medical Research Centre, Mumbai, India., Tukvadze N; National Center for Tuberculosis and Lung Diseases, Tbilisi, Georgia., Maghradze N; National Center for Tuberculosis and Lung Diseases, Tbilisi, Georgia., Omar SV; Centre for Tuberculosis, National & WHO Supranational TB Reference Laboratory, National Institute for Communicable Diseases, a division of the National Health Laboratory Service, Johannesburg, South Africa., Joseph L; Centre for Tuberculosis, National & WHO Supranational TB Reference Laboratory, National Institute for Communicable Diseases, a division of the National Health Laboratory Service, Johannesburg, South Africa., Suresh A; FIND, Geneva, Switzerland., Rodwell TC; FIND, Geneva, Switzerland; Division of Pulmonary, Critical Care, Sleep Medicine, and Physiology, University of San Diego, San Diego, CA, USA.
Jazyk:	angličtina
Zdroj:	The Lancet. Infectious diseases [Lancet Infect Dis] 2024 Oct 29. Date of Electronic Publication: 2024 Oct 29.
DOI:	10.1016/S1473-3099(24)00586-3
Abstrakt:	Background: Drug-resistant tuberculosis remains a major obstacle in ending the global tuberculosis epidemic. Deployment of molecular tools for comprehensive drug resistance profiling is imperative for successful detection and characterisation of tuberculosis drug resistance. We aimed to assess the diagnostic accuracy of a new class of molecular diagnostics for drug-resistant tuberculosis. Methods: We conducted a prospective, cross-sectional, multicentre clinical evaluation of the performance of two targeted next-generation sequencing (tNGS) assays for drug-resistant tuberculosis at reference laboratories in three countries (Georgia, India, and South Africa) to assess diagnostic accuracy and index test failure rates. Eligible participants were aged 18 years or older, with molecularly confirmed pulmonary tuberculosis, and at risk for rifampicin-resistant tuberculosis. Sensitivity and specificity for both tNGS index tests (GenoScreen Deeplex Myc-TB and Oxford Nanopore Technologies [ONT] Tuberculosis Drug Resistance Test) were calculated for rifampicin, isoniazid, fluoroquinolones (moxifloxacin, levofloxacin), second line-injectables (amikacin, kanamycin, capreomycin), pyrazinamide, bedaquiline, linezolid, clofazimine, ethambutol, and streptomycin against a composite reference standard of phenotypic drug susceptibility testing and whole-genome sequencing. Findings: Between April 1, 2021, and June 30, 2022, 832 individuals were invited to participate in the study, of whom 720 were included in the final analysis (212, 376, and 132 participants in Georgia, India, and South Africa, respectively). Of 720 clinical sediment samples evaluated, 658 (91%) and 684 (95%) produced complete or partial results on the GenoScreen and ONT tNGS workflows, respectively, with 593 (96%) and 603 (98%) of 616 smear-positive samples producing tNGS sequence data. Both workflows had sensitivities and specificities of more than 95% for rifampicin and isoniazid, and high accuracy for fluoroquinolones (sensitivity approximately ≥94%) and second line-injectables (sensitivity 80%) compared with the composite reference standard. Importantly, these assays also detected mutations associated with resistance to critical new and repurposed drugs (bedaquiline, linezolid) not currently detectable by any other WHO-recommended rapid diagnostics on the market. We note that the current format of assays have low sensitivity (≤50%) for linezolid and more work on mutations associated with drug resistance is needed. Interpretation: This multicentre evaluation demonstrates that culture-free tNGS can provide accurate sequencing results for detection and characterisation of drug resistance from Mycobacterium tuberculosis clinical sediment samples for timely, comprehensive profiling of drug-resistant tuberculosis. Funding: Unitaid. Competing Interests: Declaration of interests TCR, MS, and REC received salary support from FIND through a service contract to UC San Diego. TCR and REC received grant funding from the US National Institutes of Health to develop and evaluate a tNGS solution for drug-resistant tuberculosis (R01AI176401). TCR and REC are co-inventors on a patent associated with the processing of tuberculosis sequencing data (European Patent Application number 14840432.0 and USSN 14/912,918). Both TCR and REC have transferred all rights and present and future interest in and rights to royalties from this patent to UC San Diego and the Translational Genomics Research Institute, respectively. TCR is a co-founder, board member, and unpaid shareholder of Verus Diagnostics, a company that was founded with the intent of developing diagnostic assays. Verus Diagnostics is not pursuing any drug-resistant tuberculosis diagnostics nor any diagnostics related to the technology or approaches discussed or mentioned in this manuscript. Verus Diagnostics was not involved in any way with data collection, analysis, or publication of the results of this manuscript. TCR has not received any financial support from Verus Diagnostics. CR has received honoraria payments from Becton Dickinson and she is on the scientific advisory board for Cepheid and bioMérieux. All other authors declare no competing interests. (Copyright © 2024 Elsevier Ltd. All rights reserved, including those for text and data mining, AI training, and similar technologies.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu