Adaptive immunity is one of the most complex and clinically relevant aspects of human biology, where immune cells have evolved over millions of years to discriminate non-self from self, and the dangerous from the innocuous. Though cancer cells pose an intrinsic danger to an organism, their derivation from self enables them to evade immune recognition. Nevertheless, an increased understanding of interactions between immune cells and cancer cells has culminated in extraordinary cures in refractory cancers by harnessing the immune response. However, the success of these therapies remains relegated to subsets of cancers (mostly leukemias and high mutational tumors), owing largely to a lack of highly specific immunotherapy targets in the majority of cancers. In my previous work, I developed new methods to identify novel tumor antigens and engineer a new class of chimeric antigen receptors (CAR) T cells called peptide-centric (PC)-CAR T cell, which are capable of targeting low mutational solid tumors. These PC-CAR T cells completely eradicate highly aggressive tumors in all relevant preclinical models and are entering clinical trials in 2024. The process that I developed for generating PC-CARs provides a roadmap that can be applied to all tumor types, significantly expanding the population of patients that are eligible for CAR therapies (Yarmarkovich et al., Nature 2021). Having pioneered the first- generation of PC-CARs, I am uniquely familiar with the challenges of the process and have carefully devised strategies to improve this process such as to develop safer and more effective immunotherapies. The work described in my DP2 is focused on developing new technologies that address two major bottle-necks in the current CAR development process, identifying tumor-specific antigens and engineering antigen-specific receptors. To identify tumor-specific targets, I describe a comprehensive immunoproteomic search tool that I expect will uncover targets in any given tumor. I also propose new strategies for identifying logic-gated CAR targets and methods to streamline their screening in a high-throughput manner. To generate safer and more effective PC-CAR receptors, I describe new library systems that allows for screening of CAR T cells at an unprecedented scale, enabling the efficient generation of 104 antigen-specific CAR clones and a system to screen them in the most relevant contexts. I also describe an entirely new approach for developing CAR T cells through “reverse engineering” the immune response from patients who have been cured of cancer using immune checkpoint inhibitors. Finally, I describe new computational methods for de novo engineering of PC-CAR receptors and in silico cross-reactivity screening that have the potential to dramatically scale PC-CAR development. This multi-pronged strategy describes several new approaches, each of which can transform the development of immunotherapies and benefit a large population of cancer patients if successful. In alignment with the NIH Director's New Innovator Award, this work directly builds on my previous experiences and describes high-risk, high-reward research that has significant potential to impact the lives of cancer patients.

PROJECT NARRATIVE Immunotherapies have revolutionized the treatment of cancer, but only a small fraction of patients benefit from these therapies. In my recent work, I pioneered a new class of chimeric antigen receptor (CAR) T cells, peptide- centric (PC)-CARs, paving the way for a clinical trial slated to open in 2024 and offering a roadmap for developing immunotherapies for the lethal cancers that don't benefit from first-generation CAR T cells. Here I introduce a multi-pronged approach to build on these methods, offering innovative new strategies for developing safer and more effective therapeutic strategies that can to lead to new therapies for the most challenging cancers.