Abstract 2544: Anti-MUC1* CAR T for solid tumors

Abstract Purpose: The purpose is to develop a CAR T therapeutic that will effectively treat solid tumors by targeting cleavage product MUC1*, which is the growth factor receptor, rather than full-length MUC1, and by including a novel element that is only expressed by activated CAR T cells that overcomes problems of tumor heterogeneity and microenvironment. This will be a first-in-human trial since every other MUC1-targeting therapeutic tested in humans targeted full-length MUC1. Studies show that MUC1 cleavage and release from tumor surface increase as tumor stage increases. MUC1* has no sites for O-glycosylation. Therapeutics that target aberrant O-linked glycans, such as 5E5, only enrich for the tumor-promoting MUC1* form. Experimental Procedures: Anti-MUC1* CARs were generated using standard lentiviral vector procedures and transduced into human T cells. 60 novel CARs were tested in vitro to confirm cytokine release and specific cell killing upon exposure to MUC1*-positive cancer cells, but not to cells expressing full-length MUC1 or MUC1-negative cells. Co-culture killing was measured by video, IF, FACS and xCELLigence. In vivo, 500K MUC1*-positive tumors were implanted s.c. or i.p. Human anti-MUC1* CAR T cells were injected into tumor or tail vein. Tumors were measured weekly by IVIS for 60 days. Post sacrifice, tumors and blood were analyzed for presence of CAR T cells, target cells and apoptotic markers. New Data: Human tissue studies for IND have been completed (n>3,000) and indicate our anti-MUC1* CAR T, huMNC2-CAR44, should be safe and effective. huMNC2-scFv-Fc robustly stained over 90% of breast, 83% ovarian, 78% pancreatic and 71% lung tumors, yet showed no staining on most normal tissues, including critical organs. huMNC2-CAR44 was selected from 60 novel CARs that we developed and tested. Selection was based on extensive human tissue studies, in vitro testing against a wide range of human cancer cells and in vivo testing. In vivo huMNC2-CAR44 T cells were very effective in inhibiting tumor growth. Kaplan-Meier curves show that at Day 61, there were no tumor-related deaths for HCT-MUC1* mice in the CAR T treated group, whereas untransduced T cells and PBS control animals were all dead by Day 20. At Day 18 the average IVIS measurement for CAR T treated group was 0.5 e7 compared to 87 e7 for untransduced T-cell treated group. At present (Day 56), ovarian cancer control group animals have been ordered sacrificed. Average IVIS measurement for the control group is 743 e7 compared to 38 e7 for the CAR T treated group, with 2 mice with no detectable tumor and 2 more under 6 e7. Conclusions: Subtle differences between MUC1* expressed on healthy stem-like cells and MUC1* expressed on cancerous cells have allowed development of cancer-specific MUC1* antibodies, especially suited for cancer immunotherapies. Tissue studies indicate huMNC2-CAR44 T cells should be safe and effective for treatment of breast and ovarian cancers. Conditional on IND approval, human testing will begin Q2/Q3 2018. Citation Format: Cynthia C. Bamdad, Andrew K. Stewart, Benoit J. Smagghe, Pengyu Huang, Luke T. Deary, Nelson D. Glennie. Anti-MUC1* CAR T for solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2544.