Dataset Information

Checkpoint blockade induced CD8+ T cell differentiation in responders in head and neck cancers

ABSTRACT: Background: Outcomes for locally advanced or recurrent/ metastatic head and neck squamous cell carcinomas (HNSCCs) remain unfavorable despite recent advances with immune checkpoint blockade (ICB). Preclinical studies using model antigens identified a critical role of CD8+TCF7+PD1+ T cells in anti-PD1 response. Studies on the heterogeneity and clonal dynamics of global tumor infiltrating T lymphocytes (TILs) in the HNSCC anti-PD1 and anti-CTLA4 response have not been explored. Results: In this study, we generated isogenic HNSCC anti-PD1 sensitive/resistant models that bore distinct cancer cell intrinsic transcriptomic programs. Tumor microenvironment characterization using mass cytometry and targeted depletion revealed the contribution of Tregs and M2-like macrophages in anti-PD1 resistance. Paired single-cell RNA and TCR sequencing on tumor infiltrating immune cells from ICB responsive and resistant HNSCC models identified a spectrum of CD8+ TIL subsets including TCF7+PD1- (naïve/memory-like), TCF7+PD1+ (progenitor exhausted), and TCF7-PD1+ (terminally exhausted). Mapping TCR shared fractions between these subsets identified that successful anti-PD1 or anti-CTLA4 therapy induced higher post-treatment T cell lineage transitions. A TIL differentiation gene signature was associated with better responses in multiple ICB clinical trials. Conclusions: Together, analyses integrating scRNAseq and TCRseq demonstrate distinct differentiation dynamics of CD8+ TILs in novel ICB responsive and resistant HNSCC models, highlighting critical aspects of CD8+ TIL differentiation in response to ICB.

ORGANISM(S): Mus musculus

PROVIDER: GSE153383 | GEO | 2021/12/01

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:Checkpoint blockade therapy using antibodies targeting CTLA4, PD1 or PDL1 have changed the way cancer patients with advanced disease are being treated, as evident by their FDA approval in a wide variety of malignancies, and their ability to induce high response rates when compared to conventional therapies (e.g. chemotherapy). However, even in melanoma, despite the high response rate, most patients are refractory to therapy or acquire resistance in 10-12 months. To identify key immunological elements coupled with response or resistance to checkpoint immunotherapy, we performed an unbiased analysis on 16,291 CD45+ immune cells from 32 patients (48 samples) treated with checkpoint therapy (anti-PD1=37; anti-PD1/CTLA4=11), using single cell transcriptomics. Initial unsupervised clustering of all CD45+ cells identified 11 clusters. When examining the association of these clusters with clinical outcome, we found that two clusters (B-cells, p=0.005; memory T-cells, p=0.03) were significantly enriched in responder lesions while 4 clusters (monocytes/macrophages, p=0.003; dendritic cells, p=0.015; exhausted CD8+ T-cells, p=0.005; and exhausted/cell-cycle lymphocytes, 1.3x10-5) are enriched in non-responder lesions. Next, by leveraging our unbiased single cell approach, we identified not only clusters but also specific markers associated with either responder lesions (PLAC8, LTB, LY9, SELL, TCF7, IGKC and CCR7) or non-responder ones (CCL3, CD38, HAVCR2, ENTPD1 and WARS), many of which were not previously reported to be associated with clinical outcome to checkpoint therapy. Due to the importance of CD8+ T-cells in controlling tumors, the significant association of T-cell states with clinical outcome, and their high abundance in melanoma tumors, we next focused our analysis on CD8+ T-cells and identified 2 main clusters CD8_G (with increased expression of genes linked to memory) and CD8_B (enriched for genes linked to cell dysfunction), that were significantly enriched in responder (CD8_G, p=1.4x10-6) and non-responder (CD8_B, p=0.005) lesions, respectively. Since cells with both states coexist in each of the responder and non-responder lesions, we decided to calculate the ratio between the number of cells in these 2 clusters and observed a significant separation between responders (CD8_G/CD8_B>1) and non-responders (CD8_G/CD8_B<1) when looking at all samples, baseline or post samples separately. Similar results were detected when looking at the expression of a single transcription factor, TCF7, in CD8 T-cells in an independent cohort (n=43) using a simple immunofluorescence assay. Additionally, we validated the identity and function of some of the newly identified cell states as well as their epigenetic landscape, and found that cells expressing the inhibitory molecules CD39 and TIM3, on top of being enriched in non-responder lesions, are likely to play a crucial role in T-cell exhaustion in melanoma. Collectively, our study provides extensive unbiased data in human tumors for discovery of predictors and therapeutic targets to checkpoint immunotherapy.

Dataset Information

Checkpoint blockade induced CD8+ T cell differentiation in responders in head and neck cancers

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure