Multimodal delineation of a layer of effector function among exhausted CD8 T cells in tumors

The anti-tumor function of CD8 T cells is limited through well-established pathways of T cell exhaustion (TEX). Strategies to capture emergent functional states amongst this dominant trajectory of dysfunction are necessary to find pathways to durable anti-tumor immunity. By leveraging transcriptional reporting (by the fluorescent protein TFP) of the T cell activation marker Cd69, related to upstream AP-1 transcription factors, we define a classifier for potent versus sub-optimal CD69+ activation states arising from T cell stimulation. In tumors, this delineation acts an additional functional readout along the TEX differentiation trajectory, within and across TEX subsets, marked by enhanced effector cytokine and granzyme B production. The more potent state remains differentially prominent in a T cell-mediated tumor clearance model, where they also show increased engagement in the microenvironment and are superior in tumor cell killing. Employing multimodal CITE-Seq in human head and neck tumors enables a similar strategy to identify Cd69RNAhiCD69+ cells that also have enhanced functional features in comparison to Cd69RNAloCD69+ cells, again within and across intratumoral CD8 T cell subsets. Refining the contours of the T cell functional landscape in tumors in this way paves the way for the identification of rare exceptional effectors, with imminent relevance to cancer treatment. Overall design: To identify patterns of gene expression within six broadly defined cell populations, we also performed cell sorting for bulk RNA-sequencing from freshly resected primary human tumors including compartments denoted: 1. “Live”: All viable cells at the time of sorting, 2. “Tconv”: sorted conventional CD4+ and CD8+ T cells, 3. “Treg”: CD25+ CD4+ (enriched for regulatory) T cells, 4. “Myeloid”: Lymphocyte-negative HLA-DR+ (enriched for myeloid) cells, 5. “Stromal”: CD45- CD44+Thy1+ cells, and 6. “Tumor”: all other CD45- cells. We use the bulk RNASeq data from the 'Tconv' compartment for this study *************************************************************** The table below lists GEO accessions reused/reanalyzed for this study. ***************************************************************