Performance of the GenoType MTBDRsl assay for the detection second-line anti-tuberculosis drug resistance

The rapid detection of drug-resistant tuberculosis (TB) is important to improve treatment outcomes and prevent disease transmission. The GenoType MTBDRsl assay (MTBDRsl assay) was developed to detect fluoroquinolone (FQ) and second-line injectable drug (SLID) resistance. The aim of this study was to evaluate the performance and clinical utility of MTBDRsl assay. We retrospectively reviewed patient medical records with MTBDRsl assay data between December 2011 and February 2017. MTBDRsl assay results were compared with that of phenotypic drug susceptibility testing. In addition, treatment outcomes were analyzed to evaluate the clinical utility of the MTBDRsl assay. Among 107 clinical isolates (84 cultured isolates and 23 sputum specimens), 85 (79.4%) were multidrug-resistant TB and 9 (8.4%) were extensively drug-resistant TB (XDR-TB). The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of MTBDRsl assay for detecting FQ resistance was 87.5%, 94.7%, 87.5%, 94.7%, and 92.5%, respectively. The sensitivity, specificity, PPV, NPV, and accuracy of MTBDRsl assay for detecting SLID resistance was 88.9%, 98.9%, 94.1%, 97.8%, and 97.2%, respectively. Novel drugs such as bedaquiline and linezolid were more commonly used in patients with FQ or SLID resistance detected by the MTBDRsl assay and, probably therefore, the treatment outcome was favorable irrespective of FQ or SLID resistance. The MTBDRsl assay could be used as a rule-in test to detect FQ and SLID resistance. By detecting FQ- and SLID-drug resistance rapidly, novel or repurposed drugs could be initiated earlier, suggesting that better treatment outcomes would be expected in patients with pre-XDR- and XDR-TB.