P-08.07 Evaluation of CAR T cell function in advanced 3D heterospheroid models reveals monocyte-dependent protection against cancer cell killing

<h3>Background</h3> CAR T cell therapy has improved outcomes for hematologic malignancies but remains largely ineffective in solid tumors.¹ Despite promising preclinical data, most clinical trials have failed due to barriers such as immune escape and the immunosuppressive tumor microenvironment, which limit CAR T cell infiltration and function.¹ These complexities are poorly mimicked in traditional preclinical models, limiting their ability to predict clinical responses for solid tumor treatments. Advanced 3D heterospheroid models, incorporating multiple cell types like fibroblasts and monocytes, provide a more physiologically relevant platform for evaluating CAR T cell therapy as well as other immunotherapies. <h3>Materials and Methods</h3> 3D heterospheroids were generated using human fibroblasts and either HT-29 or MDA-MB-231 reporter cancer cells expressing a luciferase/mCherry construct. After three days of spheroid formation, monocytes isolated from healthy donors were added and allowed to infiltrate the spheroids for four days. Subsequently, exhausted or non-exhausted EGFR-targeting CAR T cells were introduced, along with immune checkpoint inhibitors. Cancer cell lysis was assessed 72 hours post-treatment using a luciferase assay, while CAR T cell activation and exhaustion were evaluated by Luminex and flow cytometry. Continuous killing dynamics were tracked by mCherry fluorescence using the xCELLigence platform. <h3>Results</h3> EGFR-targeting CAR T cells significantly killed MDA-MB-231 and HT-29 heterospheroids compared to untransduced T cells, though differences between the two cell lines were observed. In HT-29 spheroids, CAR T cells exhibited higher expression of activation markers CD69 and CD137, while CTLA-4 expression was more elevated in MDA-MB-231 spheroids. IFN-γ secretion was highest in MDA-MB-231 spheroids, which correlated with cancer cell killing within 15 hours, whereas HT-29 spheroids required nearly 45 hours before killing was observed. The addition of monocytes significantly reduced CAR T cell-mediated killing, with a more pronounced effect in HT-29 spheroids. However, checkpoint marker expression and cytokine secretion were similar in the presence or absence of monocytes, suggesting no direct modulation of CAR T cell activation. Exhausted CAR T cells showed reduced killing capacity, which was partially restored by the addition of an anti-LAG-3 antibody. <h3>Conclusions</h3> In this study, we successfully mimicked solid tumor conditions, demonstrating that advanced heterospheroids are effective models for understanding tumor-immune interactions and evaluating immunotherapies. We showed that monocytes within 3D heterospheroids impaired CAR T cell efficacy, highlighting the immunosuppressive role of monocytes and reflecting the immunosuppressive tumor microenvironment. Additionally, we found that exhausted CAR T cells exhibited reduced killing, which was partially restored with an anti-LAG-3 antibody, replicating T cell exhaustion in vivo. <h3>Reference</h3> Yu WL, Hua ZC. Chimeric antigen receptor T-cell (CAR T) therapy for hematologic and solid malignancies: efficacy and safety-a systematic review with meta-analysis. <i>Cancers (Basel)</i> 2019;<b>11</b>(1). N.H. Madsen: None. B.S. Nielsen: None. I. Skandorff: None. C. Rodriguez-Pardo: None. S.R. Hadrup: None. M. Ormhøj: None. A.L. Blomberg: None. K. Holmstrøm: None. J. Larsen: None. M. Gad: None.