Comparison of targeted next-generation sequencing and the Xpert MTB/RIF assay for detection of Mycobacterium tuberculosis in clinical isolates and sputum specimens

ABSTRACT Targeted next-generation sequencing (tNGS) can be used to perform Mycobacterium tuberculosis (MTB) complex-specific amplification or target capture directly from sputum samples, yielding simultaneous coverage of many genes and DNA regions associated with antimicrobial resistance (AMR). Performance comparisons of tNGS and another molecular testing tool, Xpert MTB/rifampicin (RIF), have been empirical. Here, using a dilution series of a RIF-resistant clinical isolate of MTB, we found that tNGS had a slightly lower limit of bacterial detection (10 2 CFU/mL) compared with Xpert MTB/RIF (10 3 CFU/mL) in culture medium. However, the minimum detection limit of the rpoB S450L mutation in this isolate was significantly lower with tNGS (10 2 CFU/mL) than with Xpert MTB/RIF (10 6 CFU/mL). Sputum samples collected from 129 suspected pulmonary tuberculosis patients were also prospectively studied with the clinical diagnosis as a reference, revealing that the sensitivity of tNGS (48.6%) was higher than those of culture (46.8%), Xpert MTB/RIF (39.4%), and smear microscopy (34.9%) testing. Notably, AMR analysis of 56 MTB-positive samples as determined by tNGS revealed high mutation frequencies of 96.4%, 35.7%, 26.8%, and 19.6% in the following AMR-associated genes: rrs , rpoB , katG , and pncA , respectively. The findings of this study provide theoretical support for the differential clinical application of tNGS and Xpert MTB/RIF and suggest that tNGS has greater application value in tuberculosis drug resistance monitoring and prevention. IMPORTANCE Targeted next-generation sequencing (tNGS) can be used to perform Mycobacterium tuberculosis (MTB) complex-specific amplification or target capture directly from sputum samples, yielding simultaneous coverage of genes and DNA regions associated with antimicrobial resistance (AMR). Performance comparisons of tNGS and Xpert MTB/rifampicin (RIF) have been empirical. The Xpert MTB/RIF assay is a commercial system that uses the nucleic acid amplification detection method for rapid (2 hours) diagnosis of tuberculosis (TB). The cost of the tNGS and Xpert MTB/RIF assays in this study was similar, at USD 98 and USD 70–104 per sample, respectively, but the time required for tNGS (3 days) was much longer than that required for the Xpert MTB/RIF assay. However, tNGS yielded more accurate results and a larger number of AMR-associated gene mutations, which compensated for the extra time and highlighted the greater application value of tNGS in TB drug resistance monitoring and prevention.