Introduction
Multidrug-resistant tuberculosis (MDR-TB)—defined as resistance to at least isoniazid (INH) and rifampicin (RIF)—remains a formidable public health issue globally and notably in China. The World Health Organization estimates that drug-resistant TB accounts for a substantial proportion of TB mortality, with challenges compounded by heterogeneous resistance mechanisms that complicate treatment. Hainan Island, the largest tropical island of China, presents unique socio-geographical and epidemiological contexts that influence the transmission dynamics and resistance profiles of Mycobacterium tuberculosis (MTB). However, prior to this study, comprehensive characterization of MDR-TB strains from this region, particularly integrative genotype-to-phenotype analyses, had been limited. This investigative work by Wang et al. addresses this gap by leveraging whole-genome sequencing (WGS) combined with phenotypic drug susceptibility testing (DST) to delineate resistance patterns and quantitatively link resistance-conferring mutations with drug minimum inhibitory concentration (MIC) levels, thereby informing region-specific therapeutic strategies.
Study Design
The study prospectively collected 209 MDR-MTB strains isolated from patients on Hainan Island between 2019 and 2021. A phenotypic drug susceptibility approach was employed using broth microdilution (BMD) to determine MICs against a comprehensive panel of 15 anti-TB drugs, encompassing first-line agents—INH, RIF, ethambutol (EMB), and pyrazinamide (PZA)—and key second-line drugs, including injectable aminoglycosides (amikacin [AMK], capreomycin [CPM], kanamycin [KM]), fluoroquinolones such as moxifloxacin (MOX), and others. Concurrently, whole-genome sequencing was performed utilizing Illumina NovaSeq 6000 platforms, and genotypic resistance predictions were analyzed via the TB-Profiler bioinformatics pipeline. The primary endpoint was assessing correlations between specific resistance mutations and MIC values, alongside evaluating the sensitivity and specificity of WGS in predicting phenotypic resistance.
Key Findings
The 209 MDR-MTB strains predominantly belonged to lineage 2.2 (East Asian lineage), which exhibited significantly elevated EMB resistance compared to non-lineage 2 strains (P 95%) in predicting resistance to AMK, CPM, and KM. Among fluoroquinolones, only moxifloxacin showed high sensitivity (94.4%), while other second-line agents lagged behind 80% sensitivity, underscoring limitations in current mutation catalogs for predicting phenotypic resistance.
Crucially, the study mapped specific mutations to resistance magnitude: katG_S315T mutations correlated strongly with high-level INH resistance, rpoB_S450L with severe RIF resistance, and gyrA_D94G with pronounced fluoroquinolone resistance. Conversely, mutations in promoters of fabG1, ahpC, embA and gyrA codon 90 mutations correlated with lower MICs, suggesting these may confer intermediate or low-level resistance. This stratification provides quantitative granularity potentially useful for MIC-guided dose modifications and refined treatment regimens.
Detailed MIC distribution data correlated with mutation profiles illuminated the heterogeneous phenotypic landscape within MDR-MTB, highlighting the complexity of tailoring anti-TB therapy in endemic regions with diverse genetic backgrounds.
Expert Commentary
Wang et al.’s study advances the understanding of regional MDR-TB genotypic diversity and its practical implications for treatment optimization in geographically distinct settings like Hainan Island. The high sensitivity of WGS for detecting resistance to first-line drugs affirms its utility as a rapid diagnostic adjunct, although lower specificities for drugs like EMB and PZA caution against unqualified reliance on genetic data alone. The strong correlations between specific mutations and resistance levels provide mechanistic insights and support the rationale for integrating MIC determination with molecular diagnostics to individualize dosing.
However, the study’s limitations include potential sampling bias inherent to the isolates collected and the need for functional validation of novel or less-characterized mutations. Moreover, the lower predictive accuracy for some second-line drugs emphasizes the necessity for ongoing expansion and refinement of mutation databases. Generalizability beyond Hainan may be restrained by unique regional strain distributions impacting resistance mechanisms.
Current WHO guidelines advocate for the use of molecular diagnostics complemented by phenotypic DST for MDR-TB management; findings here reinforce this paradigm and encourage regional customization of molecular assays and treatment protocols.
Conclusion
This pioneering investigation into MDR-TB strains from Hainan Island distinctly characterizes the genotypic and phenotypic resistance spectrum, establishing important links between specific mutations and resistance magnitudes. Such insights foster the development of MIC-based dose-adjustment strategies, promising to enhance therapeutic efficacy and mitigate treatment failure in a regionally tailored manner. Future research incorporating longitudinal clinical outcomes linked to resistance profiles will be critical to validate and operationalize these genotype-phenotype correlations in routine clinical practice.
Funding and Trial Registration
This study is registered under MR-46-23-020530 as of 2023-07-03. Funding details were not explicitly mentioned in the manuscript.
References
Wang J, Dong J, Chen Z, Xu Y, Qiu W, Chen S, Pei H, Zhong Y. Surveillance and analysis of drug resistance and drug resistance levels in multidrug resistant tuberculosis on the tropical islands of China. BMC Infect Dis. 2025 Aug 1;25(1):973. doi: 10.1186/s12879-025-11312-8. PMID: 40750855; PMCID: PMC12317541.
Additional references supporting MDR-TB genomics and therapeutics can be found in recent WHO technical reports and reviews on molecular diagnostics for tuberculosis.