Development of a modular customizable screening platform for chimeric antigenreceptor optimization

About This Grant

Project Summary Chimeric antigen receptors (CARs) redirect T cells to eliminate antigen-expressing cells by coupling extracellular antigen binding domains to intracellular T cell signaling domains via hinge and transmembrane domains. Each domain of a CAR plays a crucial role in shaping CAR T cell phenotype and function. Despite the fundamental relationship between CAR design and T cell phenotypes, efforts to optimize CAR design have been limited and low-throughput. Consequently, most CARs in the clinic utilize the same basic design architectures, likely contributing to suboptimal performance. The overall goal of this Phase I STTR project is to develop a modular, high-throughput screening platform to optimize CAR designs for any antigen, disease, and modality. The central hypothesis is that CAR design optimization can address key challenges such as tonic signaling, antigen sensitivity, and antigen loss. To achieve the overall goal, the research proposal is structured into three specific aims focused on (1) screening large libraries of CD30 and (2) BCMA CAR designs to overcome diverse and fundamental challenges, and (3) building computational models to predict optimal CAR designs. The first aim is motivated by the profound signatures of tonic signaling of a clinically tested CD30 CAR design. Previous data suggest that multiple domains of a CAR can influence tonic signaling. To identify optimal CAR designs that reduce tonic signaling and retain antigen-specific potency, a large (~105 member) library of hinge, transmembrane, and CD3z variants will be screened using readouts of dysfunction and antigen potency. The second aim seeks to showcase the modularity of the platform by addressing a different challenge – increasing antigen sensitivity and reducing antigen loss – of a commercial BCMA CAR. The BCMA CAR exhibits impaired cytokine production in vitro and tumor control in vivo in response to tumor cells with low levels of BMCA. To identify optimal CAR designs that increase antigen sensitivity and reduce antigen loss, a large (~105 member) library of hinge, transmembrane, and CD3z variants will be screened using readouts of antigen sensitivity and antigen loss. The third aim will leverage these screening datasets to develop computational models that predict CAR activity, expanding the search space from empirical data of ~105 CARs to all possible ~1011 CAR designs. This research proposal is strongly aligned with the mission of NCATS to turn research observations into health solutions through translational science. Developing a screening platform will expedite the scale and speed at which research observations of CAR designs can happen and translate into life changing therapies. In the short- term, this proposal aims to discover optimal CD30 and BCMA CAR designs to improve CAR T cell therapy for Hodgkin Lymphoma and Multiple Myeloma. This proposal establishes the foundation of a long-term goal to engineer more effective CAR T cell therapies for indications with high unmet need.