Astrocytes reprogram radiation-resistant glioblastoma to reveal new therapeutic vulnerabilities

Glioblastoma (GBM) is a highly aggressive and therapy-resistant brain tumor, with limited treatment options and poor patient prognosis. One major barrier to effective treatment is the tumor's interaction with the surrounding brain microenvironment, particularly astrocytes, which can modulate therapeutic responses and contribute to resistance. To investigate how astrocytes influence resistance and uncover exploitable vulnerabilities, we developed direct and transwell co-cultures using patient-derived (HW1) GBM cells and primary human astrocytes plated across multiple ratios (10:90, 50:50, and 90:10). This novel co-culture system enabled us to model both contact-dependent and paracrine astrocyte signaling. We tested whether astrocytes at different densities were able to modulate tumor response to a panel of clinical and pre-clinical drugs, in the post-radiation setting. GBM-astrocyte cocultures were exposed to a clinically relevant radiation protocol (three cycles of 2 Gy daily x 5 days) and screened against 12 drugs, encompassing DNA-damaging agents, multi-kinase inhibitors, and nuclear transport inhibitors. Notably in direct co-cultures, even low concentrations of astrocytes (as few as 10%) significantly altered GBM sensitivity to multiple compounds. Additionally, radiation induced a therapeutic vulnerability in GBM cells when co-cultured with astrocytes, not present in GBM monocultures, to selinexor, afatinib, altiratinib, and crenolanib, with strongest effects at intermediate astrocyte:GBM ratios (10:90, 50:50). Transwell experiments demonstrated similar shifts in sensitivity, supporting the hypothesis that astrocyte-secreted paracrine factors alone are sufficient to reprogram GBM drug response. These observations imply that astrocytes reprogram resistance-associated pathways in GBM cell lines post-radiation treatment, potentially hijacking saturated DNA repair mechanisms, cell cycle checkpoints, stress signaling, and transporter activity, all of which invites further mechanistic investigation. Building upon these findings, ongoing studies include genetic validation of candidate pathways, expanded transwell assays to further separate contact-dependent and paracrine mechanisms, and live-cell imaging to define the temporal resistance dynamics. In conclusion, our study demonstrates that the presence of astrocytes alone significantly influences GBM response to therapy, and radiation therapy can further expose collateral sensitivities driven by astrocyte signaling. Modeling and mechanistically dissecting these microenvironmental interactions is essential for identifying new avenues for combination therapies that may improve outcomes in a cancer type where innovation is urgently needed.