Identification of genes that limit CAR T persistence in solid tumors

Project Summary Chimeric Antigen Receptor T (CAR T) cell therapy has been highly effective in a subset of hematological malignancies. However, this efficacy has not translated to solid tumors for a variety of reasons, including limited persistence in an immunosuppressive and metabolically challenging tumor microenvironment. Unbiased techniques such as genome-scale CRISPR screens have identified several pathways to enhance therapeutic efficacy. However, these screens have mainly utilized reductionist in vitro approaches, and do not recapture a solid tumor microenvironment. I have performed a genome-wide in vivo CRISPR screen of human CAR T persistence in solid tumors. The top two hits were ATXN7L3 and USP22, components of the Spt-Ada-Gcn5 acetyltransferase (SAGA) complex deubiquitylation module (DUB). The SAGA DUB is primarily known for its role in histone H2B deubiquitination, but its role in T cell persistence and antitumor function remains to be described. I show knockout of USP22 or ATXN7L3 (USPKO/ATXKO) increased T cell memory markers and enhanced CAR T expansion and persistence in vitro. Paradoxically, USPKO/ATXKO CAR T exhibit a defect in IL- 2 secretion and fail to enhance antitumor immunity or persistence in vivo. The data indicate that USP22 and ATXN7L3 are fundamental regulators of CAR T cell function, yet the mechanisms through which they influence CAR T biology remain unclear. I hypothesize that knockout of USP22 or ATXN7L3 reprograms the CAR T transcriptome through changes in histone ubiquitination and augments CAR T persistence in an IL-2 dependent manner. In Aim 1, I propose to determine the relationship between cell-extrinsic IL-2 and USPKO/ATXKO CAR T persistence in solid tumors. I postulate that cell-extrinsic IL-2 is sufficient to rescue USPKO/ATXKO persistence and enhance antitumor activity. As USPKO/ATXKO CAR T persist in a pooled in vivo setting, I will pool wildtype CAR T cells with USPKO/ATXKO CAR T at different ratios and infuse into tumor-bearing mice to assess persistence and antitumor activity. I will also determine the effects of exogenous IL-2 supplementation on in vivo USPKO/ATXKO CAR T persistence through serial recombinant IL-2 infusions. In Aim 2, I propose to investigate the impact of USPKO/ATXKO and histone deubiquitination on CAR T transcriptional programs. I hypothesize that H2B deubiquitination through SAGA regulates gene expression programs that promote CAR T persistence. To evaluate this, I will perform bulk RNA-sequencing and histone ubiquitin CUT&RUN-sequencing on USPKO/ATXKO to determine the extent to which histone ubiquitin changes affect these programs. Together, the findings of this work will describe a novel role for the SAGA complex and histone ubiquitination in CAR T persistence and has potential for direct clinical translation for treatment of solid tumors.