Single-cell RNA sequencing combined with artificial intelligence to uncover immunotherapy biomarkers and decipher drug resistance mechanisms in glioblastoma

Abstract

Glioblastoma represents the most aggressive primary malignant tumor of the central nervous system, characterized by extensive intratumoral heterogeneity, profound immunosuppression, therapeutic resistance, and poor clinical outcomes. Although immunotherapy has revolutionized cancer treatment, its clinical efficacy in GBM remains limited, largely due to the complexity of the tumor ecosystem and the lack of validated predictive biomarkers. Recent advances in single-cell RNA sequencing (scRNA-seq), spatial transcriptomics, and artificial intelligence (AI)-based computational approaches have provided unprecedented opportunities to decipher tumor heterogeneity and identify clinically actionable biomarkers. This mini-review integrates studies related to scRNA-seq and its combination technology, further elaborating on the tumor plasticity, immune suppression, and metabolic-epigenetic reprogramming mechanisms of GBM. The article systematically summarizes the complete analysis process of integrating single-cell multi-omics data by AI and highlights the application value of AI in validating biomarkers. The above research ideas are expected to overcome clinical resistance problems, optimize patient stratification schemes, and enhance the efficacy of immunotherapy, providing theoretical support for developing highly stable prognostic biomarkers and achieving individualized precise immunotherapy for glioblastoma.