ABSTRACT Modern immunotherapies have had a broad impact in clinical oncology. CAR T cell therapy has been dramatically successful against certain hematologic malignancies, and ‘immune checkpoint blockade’ targeting the PD-1/PD- L1 pathway or other T cell inhibitory pathways has proven to be an effective treatment for advanced solid tumors in a subset of patients. On the other hand, CAR T cell therapy has had only limited success against solid tumors, and many patients treated with immune checkpoint blockade either do not respond or experience tumor recurrence. One shared factor limiting the ability of these treatments to control cancer is that tumor-infiltrating T cells become ‘exhausted’. We need to understand immune cell exhaustion at a molecular level, both in mouse models and in humans, in order to design therapies that will be more effective at eradicating the original tumor and in fostering the development of tumor-specific memory cells that will prevent a recurrence. We discovered some years ago that the transcription factor NFAT— classically a main driver of T cell effector responses— also initiates T cell exhaustion. T cell receptor stimulation paired with effective costimulation activates NFAT and its transcriptional partner AP1, and drives the effector response. However, NFAT simultaneously activates a separate cell-intrinsic program that damps down T cell responses, with the relative strength of the two transcriptional programs depending on the context. In tumors, the hyporesponsiveness program is favored, leading to T cell exhaustion. We recently identified TOX- and NR4A-family transcription factors as genes induced by NFAT that cooperate with NFAT to further the exhaustion program. It has been thought that exhaustion is a gradual process, but in fact we have demonstrated that TOX mRNA and TOX protein— along with other markers of exhaustion— are sharply induced in tumor antigen-specific CD8+ T cells almost immediately when they encounter a tumor. This means that CAR T cells or expanded tumor antigen- specific T cells are immediately at risk when they are infused into a patient, perhaps explaining the requirement for transferring large numbers of cells, and accounting in part for the limited success of the therapies. Here, we plan to address the fundamental transcriptional mechanisms controlling the onset of exhaustion, as a background for eventually circumventing this problem in the clinic. In Aim 1, we will define the ensemble of transcription factors and, by extension, the cellular signalling pathways that drive TOX expression and lead to exhaustion. In Aim 2, we will probe how TOX then furthers the exhaustion program by directly controlling key exhaustion-specific genes. In Aim 3, we will address the very practical question whether TOX protein levels in tumor-infiltrating lymphocytes can be titrated to prevent or reverse exhaustion, taking an approach that could in principle be developed further for clinical use. Our results will ...