Identifying key molecular drivers of survival and therapeutic targets in glioblastoma through integrated transcriptomic analysis.

2080 Background: Glioblastoma multiforme (GBM) remains one of the most aggressive brain tumors, characterized by poor survival rates and limited therapeutic success. Advancing treatment requires identifying molecular drivers of tumor progression and therapy resistance. Integrated transcriptomic analyses offer a powerful means to uncover pathogenic pathways, therapeutic targets, and refine prognostic tools. This study aimed to explore GBM’s molecular landscape to identify survival-relevant genes and actionable pathways, ultimately paving the way for precision therapies to improve outcomes. Methods: RNA-sequencing data from TCGA, CGGA, CPTAC, GLASS, GSE121720, and GSE147352 datasets, encompassing 783 samples, were analyzed. Only primary, untreated GBM tumors with survival data were included. Data normalization was performed using the voom function in limma, and batch effects were corrected using ComBat while preserving survival-related and gender-specific variations. Differential expression analysis (DEA) was used with thresholds of |logFC| > 1 (two-fold expression change) and adjusted p < 0.001, accounting for age, gender, race, and IDH1 mutation. Protein–protein interaction (PPI) networks were constructed using STRING, and Cox regression identified survival-related genes. DrugBank was used to link survival-associated genes to potential therapeutics. Results: Of the 783 samples, 488 met inclusion criteria (473 tumor, 15 non-tumor). DEA identified 1,453 differentially expressed genes (DEGs) with significant differences between tumor and non-tumor samples. PPI analysis highlighted 270 hub genes, of which 47 were significantly associated with survival. Notably, CDK1, RRM2, and BIRC5 showed negative prognostic effects (hazard ratio [HR] > 1.7, adjusted p < 0.05), while RPL3L, RPL21, and RPL9 exhibited protective effects (HR < 0.5, adjusted p < 0.001). DrugBank analysis identified gallium nitrate (targeting RRM2) and Alsterpaullone (inhibiting CDK1) as promising therapeutic candidates. Conclusions: This study provides a comprehensive bioinformatics analysis of GBM, integrating multiple transcriptomic datasets to identify critical survival-related genes, including CDK1, RRM2, and BIRC5, as key therapeutic targets. Notably, it highlights the protective roles of ribosomal proteins (RPL7, RPL9, RPL21), challenging the traditional view that ribosomal upregulation solely drives tumorigenesis. These findings emphasize the dual nature of ribosomal dysfunction, which has been implicated in both oncogenic and tumor-suppressive pathways. Through rigorous data normalization, batch correction, and PPI analysis, the findings offer robust insights into GBM biology and its molecular drivers. These results lay a strong foundation for future studies to validate their clinical relevance, refine prognostic models, and advance precision therapy strategies for GBM patients.