Diagnostic utility of GenoType MTBDRsl assay for the detection of moxifloxacin-resistant mycobacterium tuberculosis, as compared to phenotypic method and whole-genome sequencing.

Abstract

Recently, moxifloxacin (MFX)-resistant results of Mycobacterium tuberculosis (Mtb) obtained by GenoType MTBDRsl (second-line line probe assay [SL-LPA]) have been stratified to determine their resistance level; however, its accuracy has not been well studied. Therefore, the study aimed to evaluate the diagnostic accuracy of SL-LPA, with phenotypic drug susceptibility testing (pDST) and whole-genome sequencing (WGS) for the detection of MFX-resistant Mtb and their resistance level. A total of 111 sputum samples were subjected to SL-LPA according to the diagnostic algorithm of the National Tuberculosis Elimination Program. Results were compared with pDST of MFX (at critical concentration [CC, 0.25 μg/ml] and clinical breakpoint [CB, 1.0 μg/ml] using BACTEC mycobacterial growth indicator tube-960), and WGS. At CC, SL-LPA and pDST yielded concordant results of MFX for 104 of 111 (94%). However, at CB, 23 of 30 (77%) isolates carrying gyrA mutation known to confer low-level resistance to MFX were scored as susceptible by pDST. Among 46 Mtb isolates carrying gyrA mutations known to confer high-level resistance to MFX, 36 (78%) isolates yielded concordant results, while 10 (22%) isolates were scored as susceptible at CB by pDST. WGS identified gyrA mutations in all isolates suggested by SL-LPA. It is concluded that the stratification of MFX-resistant results by SL-LPA/genotypic method is not very well correlated with pDST (at CB), and hence, pDST may not be completely replaced by SL-LPA. gyrA D94G and gyrAA90V are the most prevalent mutations in MFX-resistant Mtb.