Back

Clinical and molecular correlates of CD39+ CD8 T cells in patients with lung cancer

Chow, A.; Uddin, F. Z.; Mangarin, L.; Rizvi, H.; Dobrin, A.; Tischfield, S. E.; Quintanal-Villalonga, A.; Chan, J. M.; Shah, N. S.; Allaj, V.; Manoj, P.; Mattar, M.; Meneses, M.; Liu, M.; Landau, R.; Ward, M.; Kulick, A.; Kwong, C.; Wierzbicki, M.; Yavner, J.; Chavan, S.; Farillas, A.; Holland, A.; Sridhar, H.; Ciampricotti, M.; Hirschhorn-Cymerman, D.; Richards, A.; Donoghue, M. T. A.; Heller, G.; Klebanoff, C. A.; Hellmann, M. D.; de Stanchina, E.; Sen, T.; Wolchok, J. D.; Merghoub, T.; Rudin, C. M.

2022-05-12 immunology
10.1101/2022.01.24.477554 bioRxiv
Show abstract

The repertoire of tumor-infiltrating lymphocytes (TILs) can be vast, and many of these TILs are not endowed with tumor reactivity. While a number of reports have shown that tumor-reactive TILs express CD39, few reports have demonstrated that conversely, CD39 can be leveraged to serve as a proxy of tumor-reactive CD8 T cells. Using single-cell CITE/RNA/TCRseq, we show that CD39+ CD8 T cells in human lung cancers demonstrate transcriptional and proteomic features of exhaustion, tumor reactivity, and clonal expansion. Moreover, TCR cloning revealed that CD39 enriched for tumor-reactive CD8 T cell clones. Flow cytometry of 440 lung cancer specimens revealed that CD39 level on CD8 T cells is only weakly correlated with tumoral features that currently guide lung cancer therapy, such as histology, driver mutation, PD-L1 and tumor mutation burden. PD-1 axis blockade, but not cytotoxic chemotherapy, increased intratumoral CD39+ CD8 T cells. CD39 correlated with PD-1 expression on CD8 T cells and high pre-treatment/early-on-treatment levels were associated with improved clinical outcomes, but not immune-related adverse events, from immune checkpoint blockade therapy. This comprehensive profiling of the clinical, pathological and molecular features highlights the utility of CD39 as a proxy for tumor-reactive CD8 T cells in human lung cancer.

Matching journals

The top 11 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.