Back

Immune Checkpoint Blockade induces peripheral cytotoxicity and persistence of large effector CD8+ T cell clones

Watson, R. A.; Tong, O.; Cooper, R. A.; Taylor, C. A.; Verge de los Aires, A.; Mahe, E.; Nassiri, I.; Middleton, M. R.; Fairfax, B. P.

2020-11-16 immunology
10.1101/2020.11.15.383786 bioRxiv
Show abstract

Immune checkpoint blockers (ICB) exert their anti-cancer effects via CD8+ T cells, although how responses vary over sub-populations and across clones is incompletely understood. We performed single-cell RNA-sequencing of CD8+ T cells and their receptors pre- and post-ICB across eight patients, integrating results with bulk-sequencing data (n=209). We identify seven subsets with divergent responses to ICB, finding the effector cluster demonstrates the most pronounced changes. Likewise, transcriptomic response to ICB relates to clone size, with large clones demonstrating increased numbers of regulated genes of higher immunological pertinence. Cytotoxic effector clones were more likely to persist long-term following ICB and overlapped with public tumour-infiltrating lymphocyte clonotypes. Notably, pre-treatment CD8+ cytotoxicity associated with progression-free survival, highlighting the importance of the baseline CD8+ immune landscape in long-term response. This work further advances understanding of the molecular determinants of ICB response and assists in the search for peripheral prognostic biomarkers. ONE SENTENCE SUMMARYUsing single-cell and bulk RNA sequencing we explore checkpoint immunotherapy activity on peripheral CD8+ T cells in metastatic melanoma; demonstrating that cell subset and clone size determine gene expression responses to treatment, and that pre-treatment cytotoxicity and clonality of peripheral CD8+ T cells is clinically prognostic.

Matching journals

The top 9 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.