Advanced Machine Learning and Multiomics Reveal Tumor Dynamics and Therapeutic Opportunities in Glioblastoma Multiforme

Gliomas, the most prevalent primary malignant brain tumors, pose a significant clinical challenge, particularly glioblastoma (GBM), the most aggressive subtype with a median survival of just 12-15 months post-diagnosis. Molecular profiling has revolutionized GBM classification, revealing distinct subtypes-Proneural, Neural, Classical, and Mesenchymal-each associated with unique genetic and epigenetic signatures. Despite these advances, the heterogeneity within GBM demands more comprehensive approaches for effective stratification and treatment. In this study, we integrated multi-omics data, including mRNA, lncRNA, somatic mutations, and DNA methylation profiles, using consensus clustering across ten algorithms. This approach led to the identification of novel GBM subtypes and prognostic biomarkers, enhancing the precision of molecular classification. Our findings underscore the critical role of multi-omics integration in refining GBM subtypes, offering new avenues for personalized therapeutic strategies in combating this lethal malignancy.