The CD8 T Cell Landscape in Acute Myeloid Leukemia: Insights From Patient Studies and Unconditioned Mouse Models

Abstract

Acute myeloid leukemia (AML) is an aggressive hematologic malignancy with poor survival outcomes and few effective immunotherapy options. While checkpoint blockade and other immunotherapies have transformed treatment in solid tumors, they have largely failed in AML. One reason may be that our understanding of T cell dysfunction, built mostly from studying solid tumor-infiltrating lymphocytes or chronic infections, do not apply to the leukemic bone marrow.

This dissertation investigates CD8 T cell biology in AML through two complementary approaches. First, we profiled bone marrow from 46 newly diagnosed AML patients using mass cytometry and single-cell RNA and TCR sequencing on a subset. Rather than finding the canonical exhaustion signature driven by checkpoint receptors like PD1, we found that CD8 T cells in AML are skewed toward a terminally differentiated state called TEMRA. These cells are clonally expanded and express NK-like genes, suggesting they may be activated through pathways other than classical antigen recognition. Despite their terminally differentiated phenotype, they retained the ability to produce cytokines when stimulated ex vivo. Finally, we found that viral-specific CD8 T cells could not explain the most clonally expanded T cells, suggesting the possibility for leukemia-reactive clones.

Second, we established three unconditioned, immunocompetent mouse models of AML driven by distinct molecular drivers. This approach avoids the immune disruption caused by irradiation typically used in leukemia transplant experiments. Each model produced a phenotypically distinct disease with a different bone marrow immune microenvironment. Only in the most aggressive model, FLT3-NPM1, did canonical exhaustion markers emerge, and even then, cytokine production was preserved.

Taken together, these findings suggest that terminal differentiation is the dominant form of CD8 T cell dysfunction in AML. Our studies provide insight to how CD8 T cells change in the leukemic marrow, what antigens they recognize, and how the AML shapes their fate. These studies define the AML immune microenvironment across species and molecular subtypes and lay the groundwork for rational CD8 T cell-directed immunotherapy.