Introduction: Glioblastoma (GBM) is the most common and lethal adult brain tumor, with a median survival of just 8–15 months, underscoring the urgent need for novel therapies. One promising approach is Chimeric Antigen Receptor (CAR) T-cell therapy, where patient T-cells are engineered to target tumor-specific antigens, enabling selective killing. While some studies show promise in GBM, resistance and relapse remain major challenges. Established resistance mechanisms include target antigen loss and intrinsic CAR-T dysfunction, but additional mechanisms remain poorly understood. Understanding these could improve CAR-T efficacy through combination therapies or targeted CAR-T modifications.
Methods: CRISPR/Cas9 screening enables high-throughput, unbiased investigation of therapeutic resistance, with in vivo systems able to uncover insights overlooked in vitro. To explore GBM CAR-T resistance, I conducted an in vivo genome-wide CRISPR/Cas9 screen in an orthotopic, immunocompetent GBM mouse model using a custom, modular small-pooled sgRNA library. Library-infected tumor cells were intracranially injected into mice, followed by direct intratumoral CAR-T administration. Top hits informed a validation library screened in parallel in vitro and in vivo. Genetic and pharmacologic hit validation in murine and human models is ongoing. Additionally, single-cell RNA sequencing (scRNA-seq) was performed on CAR-T-treated tumors.
Results: Notably, there was no overlap between in vivo and in vitro hits. Enrichment analyses identified in vivo resistance mechanisms linked to cell migration and motility, where cells lacking key genes in these pathways were more sensitive to CAR-T killing. While target antigen loss emerged as a major resistance mechanism in vitro, it was not significant in vivo.
Conclusion : This screen uncovers potentially clinically relevant tumor-intrinsic CAR-T resistance mechanisms in GBM beyond antigen loss, which could provide insights to enhance CAR-T efficacy. Additionally, this study—the first GBM screen of its kind—establishes a robust model and iterative screening workflow for investigating therapeutic resistance across different agents and contexts, offering potential for improved treatments.
.jpg)
