Discrepancies in isoniazid susceptibility profiles: Bactec MGIT 960-resistant but GenoType MTBDRplus-susceptible Mycobacterium tuberculosis strains in Hunan, China.

Discordant drug susceptibility testing (DST) results between the Bactec MGIT 960 system (MGIT) and the GenoType MTBDRplus assay (MTBDRplus) for isoniazid (INH) complicate clinical decision-making. In this study, we performed minimum inhibitory concentration (MIC) assays and whole-genome sequencing (WGS) on 53 Mycobacterium tuberculosis strains identified as INH-resistant by MGIT but INH-susceptible by MTBDRplus. The variants conferring INH resistance were evaluated by the WHO mutation catalogue. Our results showed that only five strains carried variants classified as "associated with resistance" (Group 1/2), including katG Trp39STOP, katG Ser315Asn, inhA -154G>A, and inhA Ser94Ala. In addition, 44 strains carried 70 variants classified as "Group 3: Uncertain significance" across nine genes, including katG, ahpC, inhA, Rv0010c, Rv1129c, Rv2752c, mshA, dnaA, and Rv1258c. The remaining four strains carried no variants (Groups 1-3) linked to INH resistance. No significant difference in the prevalence of high-level INH resistance was observed between lineage 2 and lineage 4 strains (χ² = 0.232, P = 0.630). Our findings indicate that the variants classified as "uncertain significance" may be the main genetic determinants causing discordant results, highlighting their associations with INH resistance that need to be further investigated.

Importance: This study addresses a critical challenge in drug susceptibility testing (DST): the discrepancies in DST results for isoniazid (INH) between the Bactec MGIT 960 system and the GenoType MTBDRplus assay. These discordant results significantly complicate treatment decisions, potentially leading to suboptimal patient outcomes. Using MIC assays and WGS on 53 clinical Mycobacterium tuberculosis strains, we provide valuable insights into the genetic basis of INH resistance. Our findings showed that only a small fraction of strains carried variants definitively linked to INH resistance, while a larger number harbored variants of uncertain significance across multiple genes, underscoring the complexity of INH resistance mechanisms. This study highlights the urgent need to refine our understanding of these "Group 3: uncertain significance" variants, as they appear to be a primary driver of the discrepancies. Additionally, this study emphasizes the importance of integrating advanced sequencing tools into DST to improve the accuracy of INH resistance detection.