Glioma, IDH-mutant, Metabolism, Transcriptome, Gene-expression signature, Neoplasms. Tumors. Oncology. Including cancer and carcinogens, RC254-282

Abstract Background Although IDH-mutant glioma generally has a better prognosis than their IDH-wildtype counterparts, considerable prognostic heterogeneity persists among patients with the same IDH mutation. Current study has primarily focused on the different IDH statuses or grades, while the metabolic heterogeneity within IDH-mutant gliomas remains insufficiently characterized. This study aims to identify transcriptomic metabolic subtypes and associated immune microenvironment differences to better understand survival variability and potential therapeutic targets in IDH-mutant glioma. Methods Patients with IDH-mutant gliomas were included from four public datasets (TCGA, n = 373; CGGA325, n = 167; CGGA693, n = 333; GLASS, n = 100), supplemented by 22 cases from Beijing Tiantan Hospital as an independent cohort. Consensus clustering was used to define novel metabolic subtypes. Clinical features were assessed using chi-square tests and Kaplan–Meier analysis. Metabolic profiles were characterized through enrichment analysis and GSVA; immune infiltration was analyzed using CIBERSORTx and ESTIMATE. Tumor samples from the independent cohort underwent untargeted metabolomics for validation. LASSO regression was applied to select metabolic signatures, and the CGP2014 drug library was used for drug screening. Results Three metabolic subtypes (C1–C3) with distinct prognoses (p