ABSTRACT
Tuberculosis (TB) management in endemic regions often grapples with resource constraints, including the scarcity of
Mycobacterium tuberculosis
(
M.tb
) culture laboratories. The emergence of
M.tb
strains with complex drug resistance profiles necessitates rapid and comprehensive drug susceptibility testing (DST) to guide patient treatment. However, traditional phenotypic DST (pDST) for
M.tb
is costly and time-consuming. In this study, we retrospectively enrolled 829 participants from six specialized TB treatment hospitals in China from September 2022 to July 2023. The diagnostic performance of TBseq test, a targeted-nanopore sequencing assay, was compared head-to-head with
M.tb
culture, acid-fast bacillus smear, quantitative polymerase chain reaction, and Xpert MTB/RIF Ultra by using clinical diagnosis as the reference standard. Subsequently, pDST for seven anti-TB drugs (rifampicin, isoniazid, ethambutol, streptomycin, levofloxacin, amikacin, and capreomycin) was performed. The resistance predictions provided by the TBseq test were compared with pDST results, which were used as a reference standard. The performance estimates of TBseq test were quantified through sensitivity, specificity, positive predictive value, negative predictive value, and the area under the receiver operating characteristic curve (AUC), providing a comprehensive assessment of its diagnostic accuracy. We found that TBseq test demonstrated significantly superior diagnostic performance for TB compared to other methods, achieving a sensitivity of 90.9% (95% CI: 88.9%–93.0%), specificity of 93.0% (95% CI: 97.2%–99.5%), and an AUC of 0.92 (95% CI: 0.876–0.963). TBseq test also exhibited robust predictive capabilities for drug resistance to the seven anti-TB drugs, with sensitivity and specificity consistently above 90% for all drugs. The AUC values ranged from 0.919 to 0.998, indicative of high diagnostic accuracy in forecasting drug resistance.
IMPORTANCE
Our results show that TBseq test offers excellent identification performance for tuberculosis (TB), significantly outperforming
Mycobacterium tuberculosis
(
M.tb)
culture, acid-fast bacillus (AFB) smear, qPCR, and Xpert MTB/RIF. Its diagnostic accuracy for anti-TB drug resistance is also superior, with sensitivity and specificity above 90% for all drugs tested. This method can be integrated into routine clinical diagnostic workflows, enabling early diagnosis and reporting of drug resistance simultaneously.