Integrating multi-omics data and artificial intelligence for personalized medicine in glioblastoma

Glioblastoma (GBM) is one of the most lethal primary brain tumors and is characterized by profound molecular heterogeneity, rapid progression, and limited therapeutic responsiveness. Traditional diagnostic and treatment paradigms have struggled to capture the complex, multilayered biology that drives GBM evolution. Recent advances in multi-omics technologies, including genomics, epigenomics, transcriptomics, proteomics, metabolomics, and single-cell profiling, have offered unprecedented resolution into the tumor ecosystem, revealing actionable biomarkers and mechanistic pathways. In parallel, artificial intelligence (AI) and machine learning (ML) have emerged as transformative tools capable of integrating high-dimensional omics data with clinical, radiological, and pathological information to generate predictive models for diagnosis, prognosis, and treatment response. This narrative review synthesizes current progress in multi-omics-AI integration for GBM, highlighting methodological innovations, emerging biomarker signatures and applications in precision oncology. We discuss challenges related to data harmonization, interpretability, model generalizability, and clinical translation and outline future directions for building robust, ethically grounded, and clinically deployable AI-driven personalized medicine frameworks. Together, multi-omics and AI have the potential to redefine GBM management by enabling individualized therapeutic strategies and accelerating the discovery of novel targets.