Diagnostic Accuracy of LiquidArray MTB-XDR VER 1.0 for the Detection of Mycobacterium tuberculosis Complex, Fluoroquinolone, Amikacin, Ethambutol, and Linezolid Susceptibility.

Abstract

Drug susceptibility testing (DST) is essential for starting people on effective tuberculosis (TB) regimens. No published data exist for the high-throughput LiquidArray MTB-XDR (LA-XDR) test, which detects Mycobacterium tuberculosis complex (MTBC) and fluoroquinolone, amikacin, ethambutol, and linezolid susceptibility (the latter 2 have no rapid DSTs available). We enrolled people (N = 720) with presumptive TB who provided 2 sputum samples for an Xpert MTB/RIF Ultra assay and a culture (MTBC reference standard). Phenotypic DST and Sanger sequencing were the composite reference standards. LA-XDR on the manual FluoroLyse and automated GenoXtract fleXT (fleXT) DNA extraction methods were compared. For MTBC, LA-XDR had similar sensitivities (85%-87%) and specificities (99%) using each extraction method. Drug susceptibility sensitivities (95% confidence interval) varied: 94% (86%-98%) for fluoroquinolones, 64% (45%-80%) for amikacin, and 88% (79%-93%) for ethambutol (specificities of 97%-100%). Six of 7 (86%) resistant linezolid isolates were detected. LA-XDR with fleXT had indeterminate proportions of 21/251 (8%), 2/251 (1%), 63/251 (25%), and 93/251 (37%) for fluoroquinolones, ethambutol, amikacin, and linezolid, respectively (amikacin and linezolid indeterminates were higher with FluoroLyse-extracted DNA). In a hypothetical population of 100 smear-negative people with fluoroquinolone-resistant TB undergoing fleXT DNA extraction, 25 (25%) would be missed (1 extraction error, 2 invalid results, 15 MTBC-negative, 6 fluoroquinolone-indeterminate, and 1 false-susceptible). LA-XDR met the minimum World Health Organization target product profile for a next-generation, sputum-based, moderate-complexity DST with high fluoroquinolone and ethambutol resistance sensitivity, moderate amikacin resistance sensitivity, and promise for linezolid resistance, for which more data are needed. Improved LA-XDR MTBC detection would reduce missed resistance.