Background Chimeric antigen receptor (CAR)-T cell therapy has shown incredible clinical success for hematopoietic malignancies, but for solid tumors is limited by “on-target off-tumor” toxicity to vital organs due to lack of target specificity. We have developed dual CAR-T cells consisting of a canonical activating CAR (aCAR), intended to elicit efficacy in solid tumors, and an inhibitory CAR (iCAR) designed to effectively suppress aCAR activation in normal tissues. The iCAR and aCAR bind distinct cell-surface antigens that are widely co-expressed in normal tissues. The iCAR is allele specific, and targets an antigen that is commonly lost in cancer due to chromosomal loss-of heterozygosity (LOH). Therefore, whereas healthy cells express the iCAR antigen and are protected, tumors have irreversibly lost iCAR antigen expression via LOH and are killed.

Methods In this study we developed dual CARs with an aCAR targeting Her2 and iCAR targeting HLA-A2. We screened bicistronic dual CARs in human PBMCs following lentiviral transduction against target cell lines that express normal levels of both antigens paired with targets following CRISPR KO of the HLA-A2 to mimic LOH. In-vitro assays included Luciferase-based killing assays, live-imaging killing assays, and cytokine secretion. In-vivo studies were performed in NSG-mice inoculated with either the HLA-A2+ or HLA-A2 KO targets.

Results We optimized the iCAR scFv, hinge, signaling domain, and bicistronic linker to generate dual CARs with high potency and tumor-specificity. These dual CARs were highly active against HLA-A2 KO targets (“tumor”) and inhibited against HLA-A2+ targets (“normal tissue”).

Conclusions These results show that the iCAR enables the tremendous therapeutic potential of CAR-T therapy to transition to solid tumors while maintaining safety and tumor specificity.