Abstract 1521: Combining a PRAME-specific TCR showing potent in vitro and in vivo anti-tumor reactivity and a favorable preclinical safety profile with a PD1-41BB switch receptor results in highly efficient T cells

The development of effective adoptive T cell therapies to treat cancer patients has two main challenges. The first is identifying an antigen that is highly expressed in tumors with limited or no expression in normal tissues. The second is to develop a receptor that specifically binds to this antigen inducing potent anti-tumor activity without cross-reactivity to healthy cells. An additional challenge in the treatment of solid tumors is the hostile tumor microenvironment (TME) including the T cell inhibitory PD-1/PD-L1 axis. The cancer/testis antigen PRAME is an attractive target for immunotherapies of solid tumors using TCR-modified T cells as it is highly expressed in several solid tumor indications while its expression in normal tissues is mainly restricted to the testis. TCR candidates with high specificity for a PRAME-derived peptide presented on HLA-A2 were isolated from a non-tolerized T cell repertoire using our well-established high-throughput TCR generation process. Based on multi-parameter screening a lead TCR candidate was selected from more than 30 analyzed specific TCRs. T cells genetically modified to express the lead TCR (MDG1014) were characterized using a dedicated set of multi-parameter in vitro assays to evaluate specificity and activity. Favorable preclinical specificity was confirmed by analysis of the potential cross-recognition of partially homologous peptides presented on HLA-A2 and the allo-cross-recognition of common HLAs. In addition, no off-target toxicity was observed when testing a set of normal cells. Specific effector functions were confirmed by cytotoxicity and cytokine release assays using a panel of PRAME-positive tumor cell lines from various solid tumor types. Furthermore, the efficacy of MDG1014 was corroborated in a xenograft melanoma mouse model. To prevent the inhibition of T cells via the PD1-PD-L1 axis in the TME, we developed a switch receptor fusing the extracellular domain of PD1 with the intracellular signaling domain of 4-1BB. Co-expressing the switch receptor on MDG1014 in vitro led to enhanced proliferation and increased effector function against PRAME/PD-L1-positive tumor cell lines, including after repeated exposure, suggesting increased T cell fitness under chronic antigen stimulation. To confirm this effect in vivo, we have established a xenograft mouse model with tumor cells expressing PRAME and high PD-L1 levels, mimicking the TME of hard to treat solid tumors. In summary, we developed a TCR specific