Abstract PR008: Overcoming PDAC T cell therapy barriers with CD47-targeted costimulatory fusion proteins

Abstract

Abstract T cell therapy is a promising immunotherapy treatment option that uses genetically modified immune cells (T cells) to eliminate tumors. This approach uses the patients’ own immune cells to generate a “living drug” and can avoid the toxic side effects of other common therapies, such as radiation or chemotherapies. However, the pancreatic ductal adenocarcinoma (PDAC) tumor microenvironment (TME) can establish several obstacles to protect the tumor from T cells, including delivery of inhibitory and death signals that shut down T cells, usurping metabolic nutrients, and recruiting and/or converting immune cells to inhibitory phenotypes that block the T cell response. CD47 is upregulated in PDAC tumors, and patients with CD47hi tumors exhibit worse survival. CD47 serves as a “don’t eat me” signal by binding to the SIRPα receptor on macrophages to block phagocytosis. CD47 expression also inhibits T cell-mediated tumor rejection by blocking cross-presentation of tumor antigens by SIRPα+ dendritic cells. Reagents that interfere with SIRPα-CD47 signaling, such as monoclonal antibodies, are currently in clinical trials and have been shown to synergize with chemotherapy in PDA models, but native CD47 expression can produce erythrocyte and platelet depletion. Costimulatory signals delivered by surface-bound receptors can initiate gene expression programs that address multiple issues in the PDAC TME by mechanisms such as lowering the threshold of activation, altering metabolic programming, and reducing exhaustion. We develop engineering strategies to deliver costimulatory signals to engineered T cells. Immunomodulatory fusion proteins (IFPs) combine the ectodomain of an inhibitory T cell receptor with an intracellular costimulatory signaling domain and we have shown in proof-of-concept studies that this effectively “replaces a brake with an accelerator” in CD8 T cells. We have found that generating IFPs is not as simple as indiscriminately fusing any two proteins together and we have developed critical design concepts that inform the selection of the fusion site, ectodomain and costimulatory signaling domain pairing, and immunological synapse localization to enhance signaling. To develop a CD47-targeted IFP, we combined the SIRPa ectodomain with the CD28 signaling endodomain and tested this technology with our mesothelin-targeted TCR-T cell platform. In in vitrostudies, SIRPa-CD28 T cells exhibited enhanced proliferation, accumulation, and tumor cytotoxicity. SIRPa IFP-T cells exhibited increased intratumoral accumulation and therapeutic efficacy in the KrasLSL-G12D/+; Trp53LSL-R172H/+;p48Cre/+ (KPC) model, without toxicity or erythrocyte depletion. Here we describe how a thoughtfully engineered fusion protein can “armor” T cells against multiple obstacles and significantly improve therapeutic efficacy against PDAC. Citation Format: Shannon Oda, Leah Schmidt, Ashley Thelen, Cody Jenkins, Edison Chiu, Aitong Ruan, Philip Greenberg. Overcoming PDAC T cell therapy barriers with CD47-targeted costimulatory fusion proteins [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer; 2022 Sep 13-16; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2022;82(22 Suppl):Abstract nr PR008.