An Integrated Immuno-Computational Framework for Glioblastoma Therapy: Modeling Synergies in Focused Ultrasound, Nanoparticle-Mediated Immunomodulation, and CAR-Engineered NK/T-Cell Interventions

Glioblastoma multiforme (GBM) represents a paradigmatic challenge in neuro-oncology, characterized by the blood-brain barrier's impermeability to therapeutic agents, profound intratumoral heterogeneity, and a profoundly immunosuppressive tumor microenvironment. We delineate a multimodal therapeutic paradigm that integrates magnetic resonance imaging-guided focused ultrasound for precise blood-brain barrier disruption, ligand-conjugated nanoparticles for antigen-specific immunomodulation and crosspresentation, and chimeric antigen receptor-engineered natural killer (NK) or T cells for targeted cytolytic activity. This synergistic strategy is posited to enhance pharmacodynamic penetration, facilitate epitope spreading, and circumvent adaptive resistance mechanisms. The conceptual framework is underpinned by a stochastic partial integro-differential equation model embedded within a quantitative systems pharmacology architecture, enabling rigorous simulation of emergent interactions. Personalization is operationalized through hierarchical Bayesian inference, with asymptotic stability assessed via Floquet-Riccati spectral analysis and global sensitivity quantified using Sobol indices complemented by Morris elementary effects. A fully reproducible Python implementation, incorporating Crank-Nicolson finite-difference schemes for stochastic partial differential equation resolution and No-U-Turn samplers for posterior estimation, guarantees computational veracity. This contribution furnishes a theoretically robust and empirically actionable scaffold for preclinical corroboration and clinical translation in GBM therapeutics.