Abstract
Tuberculosis (TB) returns to be the leading infectious killer globally after coronavirus disease 2019. The World Health Organization (WHO) formally included targeted next-generation sequencing (tNGS) in its list of recommendations for Mycobacterium tuberculosis (MTB) and drug resistance (DR). In this study, we explored the application of various clinical sample types for TB diagnosis and DR profiles. In comparison to the composite reference standard, the overall sensitivity values of culture, Xpert, metagenomic next-generation sequencing (mNGS), and tNGS were 0.458, 0.614, 0.772, and 0.760, respectively. tNGS had sensitivity similar to mNGS, which had advantages over culture and Xpert, respectively, despite different classification of sample types. In comparison to the microbiological reference standard, the overall sensitivity values of culture, Xpert, mNGS, and tNGS were 0.606, 0.811, 0.856, and 0.884, respectively. Suprisingly, in extrapulmonary tissue and serous effusion, mNGS and tNGS had advantages over Xpert. DR-related mutations were detected in 15 cases (13.2%). There were 51 (44.7%) applicable for all DR genes, with 22 (19.3%) not applicable for DR genes. DR genes were partially applicable in 41 (36.0%) samples. However, in culture-negative TB cases, tNGS can additionally provide 52.7% first-line DR profiles. Sanger sequencing was performed on 14 samples to confirm gene mutation identified by tNGS, and the results were entirely consistent. It was concluded that the tNGS assay was a promising approach in the initial diagnostic test of MTB and DR-related genes in different clinical samples, even for the smear- and culture-negative paucibacillary samples.IMPORTANCEtNGS combines gene-specific amplification with next-generation sequencing to detect MTB and drug-resistant genes by amplifying numerous loci directly from clinical samples. The WHO implemented tNGS for the purpose of monitoring respiratory specimens for MTB detection and DR-TB due to its high sensitivity and specificity, culture independence, and ability to report heterogeneous/silent mutations. The sensitivity outperformed both culture and Xpert, and the turnaround time was significantly less than that of culture-based assays. The tNGS assay used in this study costs USD 96 and has a 12 hour turnaround time. Nonetheless, tNGS has a great deal of promise for enhancing TB detection while also addressing DR strains.