Answers are grounded in this paper and cite their sources.
Chimeric antigen receptor (CAR) T-cell therapy has shown clear biological activity in brain tumors, yet durable responses remain uncommon. Most studies have focused on CAR design, target selection and route of delivery, whereas the biological consequences of treatments administered before CAR-T infusion have received less attention. In neuro-oncology, however, CAR-T cells rarely encounter an untreated tumor. Surgery, radiotherapy, temozolomide, corticosteroids, anti-VEGF therapy and disease progression can reshape antigen expression, vascular access, myeloid composition, lymphocyte fitness and systemic immune competence. These effects may either create transient therapeutic opportunities or generate additional barriers to cellular therapy. Evidence from glioblastoma illustrates how surgery, chemoradiotherapy and recurrence progressively remodel the tumor-host ecosystem, while emerging pediatric data extend this concept beyond glioma. In diffuse midline glioma, early-phase trials have positioned CAR-T therapy after upfront radiotherapy and before or around progression. In medulloblastoma, craniospinal irradiation can produce profound lymphopenia, FLASH radiotherapy can modify macrophage states and sensitize tumors to GD2-directed CAR-T cells, and pharmacological modulation of GD2 expression can alter CAR susceptibility. Preclinical and early clinical studies in medulloblastoma and ependymoma further support locoregional delivery and histology-specific target selection. We therefore propose that treatment history should be considered a biological variable in CAR-T trial design rather than a descriptive baseline characteristic. Prospective studies should reconstruct prior treatment exposure, reassess target expression, measure immune recovery and define histology-specific windows in which tumor burden, antigen availability, tissue access and T-cell fitness are most favorable for CAR-T activity. This framework may help explain heterogeneous outcomes across current brain-tumor trials.
No takes yet. Share an insight, caveat, or question.
Baldo et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: