Back

Predicting developmental relationships of tumor resident and circulating T cells in ovarian cancer

dos Santos Carneiro, M.; Bareche, Y.; Zhao, C.; Thebault, P.; Rahimi, K.; Provenche, D.; Samouelian, V.; Cormier, B.; Cailhier, J.-F.; Mes-Masson, A.-M.; Petropoulos, S.; Stagg, J.; Lapointe, R.

2023-03-01 immunology
10.1101/2023.02.28.530488 bioRxiv
Show abstract

Characterizing T cell populations and understanding their developmental relationships may help design more effective cancer immunotherapies. We coupled single-cell transcriptomics and T cell receptor (TCR) {beta} profiling of intratumoral and peripheral T cells in ovarian cancer patients to identify transcriptional programs and infer their relationship by trajectory and TCR overlap analyses. We proposed a model of differentiation pathway from an intermediate GZMH-expressing CD8 T cell subset found in the blood and tumor that progressively reinforces the exhaustion and tissue residency programs from a CCL4-expressing cluster towards XCL1- and CXCL13-expressing terminally exhausted cells. Inferred cell communication analysis suggests that interaction with CXCL13-expressing CD4 T cells, which we refer to as Tfh-like cells, sustains the effector function of this intermediate GZMH-expressing CD8 T cell subset. Moreover, our results suggest that Tfh-like cells attract cells expressing GPR183 through the production of its ligand 7,25 dihydroxycholesterol (7,25-HC). Finally, we demonstrated that GPR183 is highly expressed in a subset of pre-effector GZMK-expressing CD8 T cells and plasmacytoid dendritic cells. Collectively, our results suggest that Tfh-like cells expressing IL-21 help promote antitumor immunity against ovarian tumors by coordinating the action of immune cells responsive to 7,25-HC.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.