The chance integration of a lentiviral chimeric antigen receptor (CAR) vector in a TET2 allele led to the expansion of a dominant CAR T cell clone, which coincided with tumor clearance in a chronic lymphocytic leukemia (CLL) patient (Fraietta et. al. Nature 2018). This observation raised the possibility of treating patients with few TET2-edited CAR T cells rather than hundreds of millions of CAR T cells as is currently done in most CAR therapies. In a pre-clinical murine model of human acute lymphoblastic leukemia (ALL), we evaluated the functional implications of TET2 loss in human T cells engineered to express different CARs. We assessed the effect of TET2 disruption on anti-tumor CAR T cell efficacy by treating immune deficient mice bearing human ALL cell line, NALM6, with limiting dose of either TET2 edited (CRISPR/Cas9) or unedited CAR T cells. We employed the two most widely used second generation CD19-specific CAR encompassing the costimulatory domain of either CD28 (Rv-1928z) or 4-1BB (Rv-19BBz). TET2 editing enhanced the anti-tumor efficacy of Rv-19BBz CAR T cells but not Rv-1928z CAR T cells. This enhanced efficacy in TET2 edited Rv-19BBz was associated with increased early cell numbers and a higher fraction of central memory CAR T cells, consistent with the clinical case report by Fraietta et al. who observed a dominant clonal expansion with a lentiviral vector-encoded 4-1BB costimulated CAR. In contrast, TET2 editing in Rv-1928z CAR T cells did not enhance their proliferation or increased their memory phenotype. Rv-1928z CAR T cells have a stronger induction of effector differentiation than Rv-19BBz CAR T cells. This led us to hypothesize that TET2 editing cannot limit the induction of effector differentiation in Rv-1928z CAR T cells but enhanced the early expansion and memory phenotype of Rv-19BBz CAR T cells. We have previously shown that 1928z driven by the TRAC promoter (TRAC-1928z) and Rv-1928z co-expressing the 4-1BB ligand (Rv-1928z-41BBL) can limit the induction of effector differentiation as compared to Rv-1928z. Disruption of TET2 enhanced the anti-tumor efficacy of both these CAR T cells and promoted acquisition of early central memory phenotype. However, TET2 edited CAR T cells, over time, attain a hyper-proliferative phenotype while maintaining a central memory profile by flow cytometry phenotyping. The frequency at which T cells achieve this phenotype depends on the CAR receptor. Functional studies reveal that these cells have highly compromised effector function. Transcriptional profiling show enrichment in gene signatures related to T cell leukemia/lymphoma with sustained upregulation of cell cycle associated factors that could not be explained by frequently observed point mutations and occasional chromosomal aberrations. These observations support a model wherein signals emanating from CARs and loss of TET2 establish sustained proliferation with loss of effector function.

Citation Format: Nayan Jain, Zeguo Zhao, Archana Iyer, Michael Lopez, Judith Feucht, Richard Koche, Julie Yang, Yingqian Zhan, Michel Sadelain. Emergence of a hyper-proliferative phenotype in TET2 edited human CAR T cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr LB153.