Abstract 4162: Dissecting personalized PD-1 inhibitor efficacy using patient-derived 3D spheroids

Abstract

Abstract Agents that invoke an immune response against tumors, such as immune checkpoint inhibitors (ICIs), have revolutionized cancer treatment. Understanding the efficacy of these agents involves more than just the characterization of patient-specific tumor cells, but also the supporting cells, such as immune cells, that encompass the tumor microenvironment (TME). Typically, selection of cancer therapy involves evaluation of the genomic landscape and expression profiling of the tumor cells alone. However, the composition of the TME has been shown to be a critical determinant in patient response to immune modulators. Thus, ex vivo profiling of ICI response for functional precision medicine should include a platform that provides a working replica of the patient’s TME. We have developed a model utilizing all cells resident in a patient’s tumor to engineer 3D spheroids for ICI response profiling. To highlight the importance of the TME, we sought to detect PD-1 inhibitor efficacy using tumor tissue from a cancer patient whose immune cells express PD-L1, but tumor cells are negative for PD-L1 and positive for MHC class I and II. To determine the role of immune and stromal cells in response to PD-1 inhibitor, we treated spheroids that were depleted of EpCAM+ cells. While PD-1 inhibitor increased cytosolic interferon gamma (IFNγ) and granzyme B in both CD4+ and CD8+ T-cells, the degree of response was decreased in the CD8+ T-cells suggesting CD4+ T-cell response is tumor cell independent but still regulated via the PD-1/PD-L1 signaling axis. To test the effect of MHC class I and class II in response to PD-1 inhibitor, MHC class I and class II blockade experiments were conducted. Only MHC class II blockade had a negative effect on CD4+ T-cell granzyme B expression, but that effect was abrogated by treatment with PD-1 inhibitor. MHC class I and II blockade individually blocked any impact of PD-1 inhibitor upon granzyme B expression in CD8+ T-cells. We then measured T-cell activation and cytokine secretion in relation to tumor cell killing following PD-1 inhibitor treatment. We found substantially increased CD25 on T-cells over the course of 96 hours but only modestly increased CD69 at 24 hours and 48 hours. Furthermore, we observed a concentration dependent increase in granzyme B and IFNγ secretion that peaked at 72 hours. We found treatment decreased EpCAM+ tumor cells at timepoints after 24 hours. These data demonstrate response from PD-1/PD-L1 blockade can be elicited in a tumor cell independent manner indicating a role for the TME in ICI efficacy. This work provides evidence that ex vivo recreation of the TME is critical to detect and accurately predict patient response to ICI therapies. Citation Format: Kathryn M. Appleton, Katy A. Lassahn, Ashley K. Elrod, Teresa M. DesRochers. Dissecting personalized PD-1 inhibitor efficacy using patient-derived 3D spheroids [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 4162.