Optimizing CAR T-cell therapy for glioblastoma: AI, dose selection, and future directions

Abstract Chimeric Antigen Receptor (CAR) T-cell therapy has emerged as a groundbreaking approach to treating glioblastoma (GBM), the most aggressive type of brain tumor with poor survival rates. CAR T-cells are genetically engineered to attack tumor-specific markers such as IL-13R α 2, EGFRvIII, and HER2. Despite promising preclinical results, clinical trials have shown limited success due to challenges like tumor heterogeneity, antigen escape, and the brain’s immunosuppressive environment. This paper explores the potential of CAR T-cell therapy in GBM treatment, integrating artificial intelligence (AI) for predictive modeling and patient stratification, identifying key challenges, cur- rent research gaps, and innovative methodologies for dose selection and trial design, such as adaptive Bayesian approaches and personalized strategies. Future directions focus on leveraging AI-driven innovations, combination therapies, biomarkers, and advanced mathematical modeling to optimize therapeutic outcomes