TMET-13. Metabolomic and lipidomic profiling reveals distinct subtypes of IDH-wildtype glioblastoma and shared metabolic features with IDH-mutant astrocytoma grade 4

Abstract BACKGROUND Glioblastoma is a metabolically diverse and aggressive brain tumor. We used untargeted metabolomics and lipidomics with supervised analysis to define metabolic changes within IDH-wildtype glioblastoma and IDH-mutant astrocytoma, grade 4, as well as to identify potential subclusters of metabolically distinct tumor types. METHODS Brain tissue from IDH-wildtype glioblastoma (n = 38), IDH-mutant astrocytoma, Grade 4 (n = 5), and non-neoplastic controls (n = 20) underwent untargeted metabolomic and lipidomic profiling. Partial Least Squares Discriminant Analysis was followed by k-means clustering. ANOVA was used to identify significantly altered metabolites, with significance defined as fold change (FC) of greater than 2 and p-value correction based on a false discovery rate of 0.05. Heatmaps visualized differences, and pathway enrichment was performed using MetaboAnalyst 6.0. RESULTS Glioblastoma exhibited numerous significantly altered metabolites and lipids compared to controls, including increases in hypotaurine (FC 5.5, p = 9x10-5) and sorbitol (FC 3.9, p = 0.003), and decreases in ribonic acid (FC -3.3, p = 1.6x10-13) and arabitol (FC -2.9, p = 4.1x10-20). K-means clustering revealed three distinct metabolic subtypes within IDH-wildtype glioblastoma. Cluster 1 showed increased galactose (p=0.02) and sucrose metabolism (p=0.009). Cluster 2 was characterized by increased glutathione metabolism (p=0.002). Cluster 3 demonstrated increased arginine biosynthesis (p=7.5x10-4). IDH-mutant tumors exhibited 130-fold elevated levels of 2-hydroxyglutarate (2HG). Lipidomic profiling demonstrated numerous altered lipids in both tumor types compared to controls, including increases in sphingomyelin 40:1 (FC 157, p=4.4x10-5) and decreases in phosphatidylcholine 39:7 (FC -2.8, p=3.0x10-19), but overall similar lipidomic profiles between the two tumor types. CONCLUSIONS Metabolomic profiling identified three distinct subtypes of IDH-wildtype glioblastoma with unique metabolic signatures and revealed shared metabolic features between glioblastoma and IDH-mutant astrocytoma grade 4. These results highlight the potential for metabolic classification to improve understanding of tumor biology and guide personalized treatment strategies for patients.