Abstract 661: Rapid CAR T testing in an ex vivo platform that maintains passage-zero patient tumor tissues and their native immunosuppressive microenvironment.

Abstract Adoptive T cell therapies, including chimeric antigen receptor (CAR) T cells, have shown limited success in solid tumors, largely due to widespread preclinical models that fail to capture patient tumor heterogeneity and the immunosuppressive tumor microenvironment (TME). Conventional systems, including cell lines, xenografts, and organoids, often lack critical TME features, leading to overestimation of therapeutic potency. Progress in next-generation CAR T therapies requires more biologically representative models. We have therefore developed Screening Live Cancer Explants (SLiCE), a New Approach Methodology platform that engrafts living, passage-zero patient tumors onto substrates of living tissue. This design preserves tumor heterogeneity better than in vitro culture and enables investigation of how patient tumor-associated cells influence T cell activity. As proof of concept, we tested B7-H3-targeted CAR T cells from three healthy donors against cell lines and passage-zero adult and pediatric brain tumors (n = 3 GBM, 2 medulloblastoma, 1 ATRT). B7-H3 expression was quantified by IHC (H-score) and flow cytometry. CAR T or control T cells were added to SLiCE-engrafted tumors at effector-to-target ratios from 1:3 to 3:1. Tumor kill was measured by live tumor cell bioluminescence within four days of seeding. Killing of B7H3+ tumors occurred only in the presence of CAR+ T cells and not non-transduced T cells; killing among multiple tumor specimens collected from the same patient positively correlated with B7H3 expression (assessed by flow cytometry). Notably, the ATRT tumor showed minimal killing despite high B7-H3 expression (H-score = 152). Flow/CyTOF analyses revealed inhibitory myeloid cells, regulatory T cells (∼30% CD4+CD25+FoxP3+), and >60% PD-L1+ tumor cells, features likely suppressing CAR T activity. These findings indicate SLiCE maintains key immunosuppressive TME elements and can dissect mechanisms of immune evasion. Ongoing studies aim to overcome these barriers. Together, these data highlight SLiCE as a model that can uniquely leverage passage-zero patient tumor and tumor-associated cells to rapidly predict outcomes and potential challenges for CAR T cells and other cell-based therapies. Citation Format: Xiaopei Zhang, Breanna Mann, Adebimpe Adefolaju, Alain Valdivia, Rajaneekar Dasari, Caroline Stockwell, Noah Bell, Ashley Geiger, Tracy A. Withers, Xin Zhou, Alexa Rodriguez, Kerry Fitzgerald, Jon Williams, Hardik Parikh, Shuang Gao, Jie An, Taylor Jensen, Shakti Ramkissoon, Dominique Higgins, Yasmeen Rauf, Scott Elton, Kimberly Hamilton, Albert Baldwin, Barbara Savoldo, Shawn Hingtgen, David Kram, Andrew Satterlee. Rapid CAR T testing in an ex vivo platform that maintains passage-zero patient tumor tissues and their native immunosuppressive microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 661.