First-in-human autologous CAR-T for metastatic breast cancers to target growth factor receptor MUC1*.

Abstract

TPS2683 Background: MUC1 has been one of the most important therapeutic targets for solid tumor cancers for the past 20+ years. However, no MUC1 targeted therapeutic has yet succeeded or has been granted FDA approval. Although MUC1 is overexpressed on cancer cells, it is also widely expressed on normal epithelial cells. We previously reported that a MUC1 transmembrane cleavage product, MUC1* (muk 1 star), is a Class I growth factor receptor activated by ligand-induced dimerization of its truncated extra cellular domain. We generated monoclonal antibodies that recognize the specific conformation that is created when MUC1 is cleaved by specific enzymes expressed in the tumor microenvironment. 93% of human breast cancer tissues are recognized by one of these antibodies, huMNC2, which importantly does not bind to normal MUC1. We have incorporated this antibody into two CARs, wherein one bears the “1XX” mutations in CD3z that greatly increase CAR-T cell persistence, inhibit exhaustion and enable the killing of low antigen expressing cancer cells. Methods: Phase I for huMNC2-CAR44, an autologous frozen product, opened January, 2020, but as a 1st-in-human trial was paused during COVID. All patients receive 300 mg/m2 cyclophosphamide and 30 mg/m2 fludarabine for 3 days prior to CAR-T treatment. Dose escalation from 3.3x105 – 1x107 CAR+ cells/kg follows a standard 3 + 3 design. In addition to standard organ function inclusion criteria, the trial was first open to breast cancer patients whose cancer had progressed after at least 2 or 3 prior therapies while metastatic, yet there was no limit on the number of prior therapies. A recent patient biopsy had to be at least 30% positive for MUC1* in a CLIA validated IHC assay. Importantly, there were no limits on the vein-to-vein time. Dose levels 1 and 2 for huMNC2-CAR44 were completed without DLTs. Phase I for huMNC2-CAR22, which bore the 1XX mutations in CD3z to increase persistence and increase the killing of low antigen expressing cells opened September, 2023. Based on responses of initial huMNC2-CAR44 patients, inclusion criteria were amended to an Enrichment Trial Design. The number of prior therapies was limited to less than or equal to 10. Anticipated survival at the time of product infusion should be at least 3 months. Vein-to-vein time, which historically had been 15-16 days at one site and as high as 83 days at another site, would be targeted to 16- 22 days. The tumors of eligible patients need to have a MUC1* membrane positive H-score equal to or greater than 120 out of a possible 300, defined as high MUC1* positivity. Based on IHC analysis of human breast cancer TMAs, the H-score restriction would include roughly 40% of breast cancers. The increased ability of huMNC2-CAR22 to kill low antigen expressing cancer cells may support future studies of patients expressing low levels of MUC1*. Clinical trial information: NCT04020575 .