Abstract

Despite advances, Glioblastoma (GBM) treatment remains challenging due to its rapid progression and resistance to therapies. This study aimed to investigate the role of anoikis-a mechanism by which cells evade programmed cell death upon detachment from the extracellular matrix-in GBM progression and prognosis. Utilizing single-cell sequencing and bulk-transcriptome sequencing data from TCGA, GEO, and CGGA databases, we performed comprehensive bioinformatics analyses. We identified anoikis-related genes, constructed a prognostic model using 101 machine learning algorithms, and validated its clinical utility across multiple cohorts.Finally, we also verified the expression of model genes and the function of key gene in clinical samples and cell lines. Single-cell sequencing revealed heterogeneous expression of anoikis-related genes across distinct cell populations within GBM. MES-like Malignant cells and Myeloids exhibited higher enrichment of these genes, implicating their role in anoikis resistance and tumor aggressiveness. The prognostic model, based on identified genes, effectively stratified patients into high-risk and low-risk groups, demonstrating significant differences in survival outcomes. Mutation and tumor microenvironment analyses highlighted distinct genetic landscapes and immune cell infiltration patterns associated with different risk groups. SLC43A3 emerged as a key gene, showing significant upregulation in tumor tissues and correlating with poor prognosis in GBM. This study provides insights into the molecular mechanisms of anoikis resistance in GBM, underscoring its critical role in tumor progression and patient prognosis. The developed prognostic model offers a promising tool for personalized treatment strategies and warrants further exploration of targeted therapies to improve outcomes for GBM patients.