Integrated Analysis of Glycolytic and Cholesterogenic Genes Identifies Prognostic Metabolic Subgroup of IDH Wild‐Type Glioblastoma
Zheng Hu, Jiajie Yuan, Yu Zeng, Renhui Yi, Xizhao WangABSTRACT
Glioblastoma (GBM) is a markedly heterogeneous intracranial tumor characterized by poor prognosis. While distinctive glycolysis and cholesterol synthesis capacities exist within its microenvironment, comprehensive profiling of these metabolic traits remains lacking. Using genomic, transcriptomic, and clinical data from 364 IDH wild‐type GBM cases across The Cancer Genome Atlas (TCGA) and Chinese Glioma Genome Atlas (CGGA) cohorts—supplemented by the Cancer Cell Line Encyclopedia (CCLE) and TCGA cancer cell line data—we performed consensus clustering using glycolytic and cholesterogenic genes. This identified four metabolic subgroups: cholesterogenic, glycolytic, mixed, and quiescent. The mixed subgroup demonstrated the poorest prognosis, whereas the cholesterogenic subgroup exhibited the most favorable outcome. Notably, high CD8+ T cell proportions correlated with worse prognosis in cholesterogenic and quiescent groups, while elevated CD4+ naïve T cells and mast cells associated with poorer outcomes in glycolytic and mixed groups, respectively. The glycolytic subgroup showed lowest expression of mitochondrial pyruvate carriers (MPC1/MPC2), and the mixed subgroup exhibited minimal methylation. Functionally, pharmacological inhibition of MPC with UK‐5099 increased glycolytic flux and lactate production, most prominently in cholesterogenic‐pattern GBM cells, and conditioned media from MPC‐inhibited or glycolytic GBM cells promoted microglial polarization toward an immunosuppressive phenotype, indicating that MPC1/2 may remodel the glycolysis–cholesterol metabolic axis to shape the microglial immune phenotype. Metabolism stratification by glycolytic/cholesterogenic profiles provides novel insights into IDH wild‐type GBM subtypes, supplements existing prognostic frameworks, and may inform the rational design of future metabolism‐directed strategies, although direct therapeutic validation will be required.