MSLN and CLDN18.2 dual-target CAR-T therapy: Enhancing cancer treatment with improved efficacy and safety.

Abstract

e16381 Background: While CAR-T cell therapy has been highly successful in treating hematologic malignancies, its application to solid tumors remains challenging due to the unique complexities of these cancers. MSLN and CLDN18.2, which show limited expression in normal tissues but are over-expressed in several tumor tissues, represent promising targets for CAR-T therapy. However, targeting CLDN18.2 carries the risk of gastrointestinal toxicity due to off-target effects on normal tissues. Notably, MSLN and CLDN18.2 are not co-expressed in healthy tissues but are frequently overexpressed together in malignancies such as gastric and pancreatic cancers，both notoriously difficult to treat and associated with poor prognoses. Methods: Given that CLDN18.2-BBζcaused gastrorrhagia toxicity in our preclinical mouse models, we developed OriC613 utilizing an "AND" logic gate design to enhance safety while maintaining anti-tumor efficacy.. OriC613 incorporates an anti-CLDN18.2 scFv with high binding strength and an anti-MSLN V H H with moderate binding affinity, ensuring full activation exclusively in tumor cells co-expressing both MSLN and CLDN18.2. Results: In both in vitro and in vivo models, OriC613 shows more potent cytotoxicity against MSLN & CLDN18.2 double-positive cells than MSLN-CD3ζ, and the secretion of IFN-γ and IL-2 is positively correlated with the expression levels of MSLN and CLDN18.2. Notably, OriC613 exhibited no cytotoxic effect on cells expressing CLDN18.2 alone, either in vitro or in vivo , indicating that it mitigates toxicity toward CLDN18.2-positive normal stomach cells. Similarly, OriC613 did not induce dose-gradient-dependent secretion of IFN-γ and IL-2 cytokines targeting BEAS-2B human lung bronchial epithelial cells (MSLN very low & CLDN18.2 neg ), suggesting minimal off-target toxicity to normal human lung tissue. OriC613 demonstrated efficacy, prolonged survival, and no gastrointestinal (GI) toxicity in the tumor stress model (NUGC-4 cells, MSLN low &CLDN18.2 med ), even against relatively large tumors (976 mm³), underscoring its potential effectiveness in patients with a high tumor burden. In a tumor re-stress model for pancreatic cancer (AsPC-1-hCLDN18.2 cells, MSLN low &CLDN18.2 high ), OriC613 exhibited efficacy, boosted IFN-γ production, and induced T cell expansion. Notably, T cells migrated to the bone marrow and spleen by Day 115, the study endpoint, demonstrating long-term persistence. Conclusions: The remarkable anti-tumor effects and favorable safety of OriC613 stress the necessity of further clinical trials in patients suffering from pancreatic and gastric cancers.