Introduction: Chimeric antigen receptor (CAR) T cells have had spectacular success in the clinic, particularly in B cell malignancies. This clinical success was built on years of pre-clinical development; despite this there remains aspects of CAR-T cell responses to antigen activation, which have not been fully explored. In this study we compared the kinetics of CAR- versus TCR-mediated activation, immune synapse formation and tumor cell killing. This study has particular clinical relevance to CAR-T cell therapy, including conventional CAR-T cells and the more recently described dual antigen receptor T cells.

Experimental procedures/strategy: To address these questions, we developed an in vitro model system where T cells could be activated via a CAR or TCR in response to cognate antigen on tumor cells. We used CD8+ T cells from a novel transgenic mouse model where both the CAR and OT-I TCR were constitutively expressed (CAR.OT-I cells), the CAR comprised an scFv directed to human HER-2 (hHER-2) and intracellular signaling domains (CD28 and CD3ζ). Effector CAR.OT-I cells were activated via their TCR using MC57 cells pulsed with SIINFEKL (MC57 OVA257), or via their CAR using MC57 cells, which expressed hHER-2. Interactions between individual effector CAR.OT-I cells and the tumor cells were monitored via time lapse live video (TLLM) microscopy and the kinetics of effector cell activation, delivery of lethal hit, tumor cell killing and effector cell detachment recorded. In addition, short term (four and eight hour Cr-release assay) and long term (50 hour xCELLigence assay) killing assays were performed to compare killing capacity and kinetics.

Key data: When effector CAR.OTI cells recognized targets via their TCR or CAR, they displayed the same kinetics for activation, delivery of lethal hit and tumor cell killing to OTI CTL. In contrast, time for CAR.OT-I cell detachment from tumor cells was significantly shorter when activated via the CAR rather than the TCR. Short term killing assays (4-8 hours) showed no difference in bulk CAR.OT-I cell killing of tumor cells. In contrast, longer term killing assays (50 hours) showed there was a significant difference in tumor cell killing rate between CAR- versus TCR-activated. The difference in killing rate was only apparent from 20-50 hours, when CAR-activated CAR.OT-I cells had a significantly reduced tumor cell killing rate compared to TCR-activated CAR.OT-I cells. Finally, we showed for the first time, that individual CAR. OT-I cells were capable of killing multiple tumor target cells (serial killing). This occurred whether the CAR.OT-I cells were activated via the CAR or TCR.

Conclusion: In conclusion, this study provides novel information regarding CAR-T cell killing of tumor cells and has practical implications for the use of CAR-T cells in the clinic, whether the therapy is using single or dual antigen receptor T cells.

Citation Format: Alex J. Davenport, Misty R. Jenkins, Ryan S. Cross, S Yong Carmen, H Miles Prince, David S. Ritchie, Joseph A. Trapani, Michael H. Kershaw, Phillip K. Darcy, Paul J. Neeson. Individual CAR-T cells, when activated via their CAR or TCR, kill tumor cells with similar kinetics and can kill multiple sequential tumor targets. [abstract]. In: Proceedings of the AACR Special Conference: Tumor Immunology and Immunotherapy: A New Chapter; December 1-4, 2014; Orlando, FL. Philadelphia (PA): AACR; Cancer Immunol Res 2015;3(10 Suppl):Abstract nr B10.