GPC2-CAR T-cells have potent preclinical activity against orthotopic medulloblastoma xenografts

Medulloblastoma (MB) is the most common malignant brain tumor in children. Patients with high-risk features, such as MYCN aberrations, and relapsed/refractory disease have poor clinical outcomes. This underscores the urgent need for new therapies. Glypican-2 (GPC2) is a MYCN-regulated and recently discovered oncofetal antigen. Given that GPC2 is also expressed in brain tumors, we evaluated the preclinical activity of our GPC2-CAR (CT3-CD28HTM-BB?) against MB and compared it to two existing CARs targeting GD2 and B7-H3. A newly generated patient-derived xenograft (PDX) MAF1433, with MYCN aberrations, were used to test CAR T-cells in vivo. Single-cell RNA-sequencing were used for mechanistic studies. MB patient samples express intermediate to high levels of GPC2 and can be targeted with a GPC2-CAR, leading to significant in vivo tumor regression in orthotopic tumor models. GPC2-CAR T-cells had equivalent activity to the B7-H3-CAR and enhanced activity compared to the GD2-CAR in vivo. T-cell kinetic studies revealed that GPC2-CAR T-cells home to the TME, expand, and upregulate genes critical for cytotoxicity and T-cell homing to induce MB cell death. CT3-CD28HTM-BB? GPC2-CAR can regress GPC2+ MB with MYCN aberrations in preclinical studies, providing a preclinical rationale for including children with GPC2+ MB in our upcoming clinical GPC2-CAR T-cell trial. Overall design: To investigate the difference in tumor microenvironment status between MB PDX treated with GPC2-CAR T cells, GD2-CAR T cells, and B7-H3-CAR T cells, we inoculated mice orthotopically with a new MB PDX; MAF1433. Around 2 weeks after the tumor injection, mice underwent IVIS bioluminescence imaging (BLI) and were randomized to receive either 5 × 10^6 CAR+ T-cells or matched total numbers of untransduced mock T-cells or PBS. On day 10 post CAR injection, the cerebella were extracted and dissociated. The cells were analyzed using scRNA seq.