646. Development of a Novel Chimeric Antigen Receptor as a Therapy Against Solid Tumors

Abstract

The use of chimeric antigen receptor (CAR) T cells as a therapy for cancer is gaining traction as they potentially target tumors with high specificity and minimal off-target toxicities. Current CAR designs consist of an antigen-specific single chain antibody fragment (scFv) fused to the transmembrane and signaling domains of receptors capable of activating T cells. In CAR T cell therapy, patients’ T cells are transduced with cDNA encoding the CAR allowing CAR expressing T cells to target tumor cells. Clinical trials using anti-CD19 CAR T cells to target B cell malignancy have demonstrated the potential of this therapy with numerous patients achieving partial or full remission. In the treatment of solid tumors, however, CAR T cell therapy has not proven as effective. Evidence supporting T cell persistence is sparse, suggesting that currently used CARs may not provide optimal signals for memory cell generation. In addition, some patients have experienced life-threating side effects including cytokine release syndrome and on-target off-tumor activity, indicating a need for better modulation of CAR T cell activity. In line with these clinical observations, data from our laboratory and others demonstrate that currently used CARs can activate T cells in the absence of antigen, which promotes toxicity and T cell exhaustion while preventing differentiation into memory T cells. We propose that this antigen-independent activation is a result of high expression of CAR molecules that lack the regulatory context of the endogenous T cell receptor (TCR) complex. To more effectively mimic the temporal regulation of endogenous TCR signaling pathways, we developed a novel TCR-based CAR in which a scFv is fused to the constant region of TCRβ, thereby becoming incorporated into the native complex. We hypothesize that incorporation of our CAR into the TCR complex will mimic endogenous T cell activation and regulation by including contributions from all CD3 members (CD3Δ, CD3e, CD3γ, and CD3ζ). Furthermore, we propose that endogenous regulation of CAR signaling will prevent the constitutive T cell activation observed with currently used CAR constructs and allow for the formation of memory T cells. We have demonstrated that surface expression of our TCR CAR is modulated by endogenous CD3 expression levels and mirrors normal TCR surface expression. In contrast to both first and second generation CARs, T cells expressing our novel CAR lack constitutive activation, are specific for tumor cells expressing high levels of antigen, and retain higher proliferative potential in culture. Moreover, TCR CAR T cells are capable of signaling through the TCR complex and activating T cells in an antigen specific manner. Currently, we are examining the effector potential of our TCR CAR T cells compared to first and second generation CARs. Additionally, we are developing TCR CARs capable of signaling not only through the TCR complex but also through co-stimulatory receptors. In summary, we are designing the next generation of CARs by incorporating the tumor recognition component of a scFv into the endogenous TCR complex.