Background: Nowadays, CAR-T cell therapy still faces limited therapeutic efficacy for solid tumors. SIRPγ is known as a ligand for a well-known immune checkpoint molecule CD47. It is distinguished from other members of its family in that it has a very short intracytoplasmic tail and is incapable of transducing signals on its own. Of note, the role of SIRPγ still largely unknown to date. Preliminary data indicated SIRPγ plays a key role in T-cell transendothelial migration and promotes antigen-specific T-cell proliferation. We thus designed a novel SIRPγ-CD28 chimeric receptor comprising the extracellular part of SIRPγ, the transmembrane and intracellular domains of CD28 and used as the co-receptor for a CEA-targeting CAR. Here, we report on the activity of this "armed" CAR-T with the SIRPγ-CD28 co-receptor in colorectal tumor (CRC) and its anti-tumor efficacy in mice xenograft model. Methods:In vitro luciferase-based cytotoxicity assay were used by coculturing CAR-T and DLD1-CEA CRC cells to assess the activities of CAR-T w/wo the co-receptor. The incubation supernatants were collected for detecting the release of cytokines. In vivo evaluation of the anti-tumor efficacy of CAR-T with the SIRPγ-CD28 co-receptor versus control T cell, and control CAR-T was performed in NOG mice CRC xenograft model. DLD1-CEA cells expressing luciferase reporter were s.c. implanted and T cells were i.v. injected. Anti-tumor efficacy was assessed by in vivo system. Results: CAR-T with the SIRPγ-CD28 co-receptor exhibited specific cytotoxicity only to CEA⁺ CRC cells, and the efficacies of CAR-T w/wo the co-receptor were comparable (88.6% vs 87.8%, p > 0.05). However, cytokines were significantly higher in the CAR-T group with the co-receptor. The levels of IFNγ, IL2 and TNFα in groups of CAR-T w/wo the co-receptor were 72,793.33 vs 19,013.33 pg/ml (p < 0.05), 1834.67 vs 25.15 pg/ml (p < 0.05) and 79.65 vs 0 pg/ml (p < 0.05), respectively. In mice CRC xenografts model, the tumors were totally eliminated in the CAR-T group with the co-receptor at 21 days and lasted until the end of experiment, while the tumor growth was suppressed in the control CAR-T group at first and out-of-control in the end (Table). Conclusions: These preclinical studies demonstrated the potential of SIRPγ-CD28 co-receptor as a novel T cellular activation signals for CAR-T. The "armed" CAR-T exhibits a fabulous anti-tumor efficacy in CRC and has great potential for applications in other solid tumors.

^aBioluminescent signal (photons/sec) values (mean ± SD, n = 5). b. One-way ANOVA test, significant difference (*).