Glioblastoma multiforme (GBM) is a highly aggressive brain tumor associated with poor survival outcomes and is driven by a complex tumor microenvironment (TME) that promotes tumor progression and treatment resistance. To explore the role of the TME in GBM, we analyzed glioma-related microarray and single-cell RNA sequencing (scRNA-seq) datasets from the Gene Expression Omnibus (GEO). Functional enrichment and weighted gene coexpression network analyses revealed distinct immune profiles, metabolic alterations, and differences in chemotherapeutic drug sensitivity between the high-risk and low-risk patient groups. scRNA-seq data processed with the 'Seurat' package were used to identify differentially expressed genes in pericytes, endothelial cells, and glioma cells, particularly those involved in extracellular matrix (ECM) remodeling. A 17-gene prognostic signature developed through Cox regression and LASSO analyses revealed that key genes (COL1A1, COL4A1, and VIM) were significantly associated with survival outcomes in GBM patients. Drug sensitivity analyses using data from the Genomics of Drug Sensitivity in Cancer (GDSC) and Cancer Therapeutics Response Portal (CTRP) identified potential targeted therapies for GBM, including SB-505,124, staurosporine, and AZD8186. This integrative study underscores the critical roles of the ECM and synaptic remodeling in GBM and suggests novel therapeutic targets to improve personalized treatment strategies for GBM patients.© 2025. The Author(s).