Abstract
BACKGROUND: Typing Mycobacterium tuberculosis (MTB) isolates is important for identifying clusters and guiding infection control measures. Although whole-genome sequencing (WGS) and 24-loci mycobacterial interspersed repetitive units-variable-number tandem repeat (MIRU-VNTR) are widely used for molecular typing, they may not capture phenotypic or metabolic variation. Alternative approaches, such as mass spectrometric-based profiling, could provide complementary insights. METHODS: A total of 247 clinical MTB isolates were analyzed by thermal desorption-electrospray ionization mass spectrometry (TD-ESI/MS). The resulting mass spectral profiles were evaluated using principal component analysis (PCA) and hierarchical clustering analysis (HCA). Results were compared with lineage classifications based on WGS and MIRU-VNTR to assess concordance. RESULTS: While WGS and MIRU-VNTR were highly concordant for lineage classification (98.8%; 244/247), TD-ESI/MS revealed diverse spectral profiles that did not align consistently with genetic lineages. However, HCA showed isolate-level clustering, including among genetically similar strains, suggesting that TD-ESI/MS may detect metabolic or lipidomic differences not captured by genome-based methods. CONCLUSION: Although TD-ESI/MS does not generate similar results from MTB molecular typing, it shows potential for identifying specific spectral signatures. The technique may be useful in future investigations into phenotypic diversity and other clinically relevant features not captured by genotyping alone.