Evaluation of Whole-Genome Sequence Method to Diagnose Resistance of 13 Anti-tuberculosis Drugs and Characterize Resistance Genes in Clinical Multi-Drug Resistance Mycobacterium tuberculosis Isolates From China

Background: Whole-genome sequencing (WGS) is a viable and financially feasible tool for timely and comprehensive diagnosis of drug resistance in developed countries. With the increase in the incidence of multidrug-resistant tuberculosis (MDR-TB), second-line anti-TB drugs are gaining importance. However, genetic resistance to second-line anti-TB drugs based on WGS has not been fully studied. Methods: We randomly selected 100 MDR-TB and 10 non-MDR-TB isolates from a hospital in Zhejiang Province, China. Drug susceptibility tests against 13 anti-TB drugs were performed, and 34 drug resistance-related genes were analyzed using WGS in all isolates. For each drug, the accuracy, sensitivity, specificity, and positive and negative predictive values of WGS were compared with those of the conventional drug susceptibility test. Results: The overall sensitivity and specificity for WGS were respectively, 99.0 and 100.0% for isoniazid (INH), 99.0 and 100.0% for rifampicin (RIF), 94.8 and 65.3% for ethambutol (EMB), 86.2 and 84.4% for pyrazinamide (PZA), 95.6 and 95.6% for levofloxacin (LFX), 89.5 and 65.3% for moxifloxacin (MFX), 91.3 and 95.1% for streptomycin (SM), 90.9 and 99.0% for kanamycin, 90.9 and 100.0% for amikacin, 88.9 and 98.0% for capreomycin, 87.0 and 85.1% for prothionamide (PTO), 85.7 and 99.0% for para-aminosalicylic acid (PAS), and 66.7 and 95.9% for clofazimine (CLO). Conclusions: WGS is a promising approach to predict resistance to INH, RIF, PZA, LFX, SM, second-line injectable drugs (SLIDs), and PTO with satisfactory accuracy, sensitivity, and specificity of over 85.0%. The specificity of WGS in diagnosing resistance to EMB, and high-level resistance to MFX (2.0 mg/L) needs to be improved.