Integrated pan-cancer multi-omics profiling and experimental validation identify LSM12 as a prognostic biomarker and candidate therapeutic target in human cancers
摘要
RNA metabolism regulators remain insufficiently characterized across tumor types, despite the substantial expansion of multi-omics datasets. LSM12, a member of the Like-Sm (LSM) protein family involved in RNA processing, has not been systematically evaluated in a pan-cancer context. Accordingly, this study aimed to investigate the oncogenic and immunomodulatory roles of LSM12 across multiple cancer types through an integrated multi-omics framework.
MethodsWe conducted an integrative pan-cancer analysis using publicly available multi-omics platforms, including TIMER2, GEPIA2, TISIDB, cBioPortal, and CPTAC (accessed via UALCAN), to assess LSM12 expression patterns, genomic alterations, epigenetic regulation, and clinical relevance. Immune infiltration was evaluated using partial Spearman correlation adjusted for tumor purity. Structure-based virtual screening and Density Functional Theory (DFT) calculations were performed to assess the druggability of LSM12. In vitro functional validation was conducted in HepG2 and TE-8 cancer cell lines using siRNA-mediated knockdown.
ResultsLSM12 was significantly upregulated in multiple malignancies at both mRNA and protein levels and was positively correlated with advanced tumor stage and grade. Increased expression was associated with promoter hypomethylation and copy-number amplification, and was predictive of unfavorable overall survival (OS) and disease-free survival (DFS) in univariate analyses. Immune infiltration analyses indicated that elevated LSM12 expression corresponded to an immunosuppressive tumor immune microenvironment (TIME), characterized by enrichment of myeloid-derived suppressor cell (MDSC) signatures and diminished natural killer (NK) cell infiltration. In vitro siRNA-mediated LSM12 silencing suppressed proliferation, migration, colony formation, and cell-cycle progression in cancer cells. Structure-based virtual screening coupled with DFT identified an isoindolinone derivative (Compound 1) as a potential LSM12 binder with favorable in silico drug-like properties.
ConclusionsThese hypothesis-generating findings position LSM12 as a candidate prognostic biomarker and potential therapeutic target in pan-cancer precision oncology, warranting further experimental and clinical validation.
Graphical abstract