Abstract 7502: Radiation-induced transcriptomic reprogramming of mesenchymal glioblastoma stem cells reveals key pathways and potential targets

Abstract Background: Glioblastoma (GBM) is a highly aggressive primary brain tumor with a median survival of less than two years. Despite standard therapy of surgery, radiation, and chemotherapy, GBMs recur. These recurrences may partially be due to glioblastoma stem cells (GSCs). Various subpopulations of GSCs exist, likely attributing to different resistant niches. The mesenchymal (MES) subpopulation may drive radiation (RT) resistance. The transcriptomic shifts that define GSC adaptation to RT remain unknown. This study explores the transcriptomic reprogramming of MES GSCs following RT to identify resistance mechanisms. Methods: Using dataset GSE162931 from the GEO database and MES gene signatures from PMID:31327527, we analyzed two patient-derived GBM cell lines pre- and post-10 Gy RT. Data for the two cell lines were preprocessed and integrated with Seurat, followed by clustering and UMAP visualization with 2D kernel module-score weighted density plots. Cluster proportion changes between conditions were statistically examined using Chi-squared or Fisher’s tests with adjusted p-values. Differential expression analysis revealed conserved and unique gene markers distinguishing treated from naïve conditions, visualized in volcano plots and heatmaps. A final score integrating module scores and cluster proportions quantified MES traits at the single-cell level. All analyses were conducted in R studio version 4.2.3. Results: UMAP, density plots, and cluster proportion changes indicate significant shifts in MES GSC distributions post-RT, suggesting RT selection for specific transcriptomic profiles. Highly MES subclusters exhibit distinctive RT-responsive gene expression patterns. The final scoring revealed enhanced MES traits in post-RT clusters. Distinct clustering patterns in MES GSCs highlight diverse adaptive strategies. Differential expression analysis reveals upregulation of DNA repair genes (SAT1, NDRG1, GADD45A), enhancing genomic stability post-RT, and stemness markers (NES, GDF15), emphasizing self-renewal capacity. Genes supporting oxidative stress management (SQSTM1, DDIT4) were upregulated, while downregulated metabolic genes (PGK1, LDHA) indicate a shift towards oxidative phosphorylation. Gene Set Enrichment Analysis identified enriched pathways related to DNA repair, cell migration, and differentiation regulation. Conclusion: These findings underline the transcriptomic plasticity in MES GSCs that drive RT resistance. Potential targets like AKAP12 and S100A4, essential for maintaining stem-like traits and MES phenotypes, present new therapeutic targets that could be used to combat resistant GSC populations, aiming to reduce tumor recurrence and enhance GBM patient outcomes. Citation Format: Grace Rosner, Abigail Goen, Godwin Peasah-Darkwah, Jonathan Rodgers Gochicoa, Alos Diallo, Gokul Srinivasan, Elisa Bu Sha, Joshua Levy, Kelli B. Pointer. Radiation-induced transcriptomic reprogramming of mesenchymal glioblastoma stem cells reveals key pathways and potential targets [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7502.