Single-cell RNA-sequencing of HNSCC-infiltrating T cells

This study examines the T cell landscape in head and neck squamous cell carcinoma (HNSCC) using single-cell RNA sequencing combined with T cell receptor and protein profiling. By analyzing tumor-infiltrating T cells from 28 patients, we identified diverse T cell subsets, characterized their transcriptional states, and mapped their clonal dynamics. These findings provide a detailed view of the immune microenvironment in HNSCC and offer insights into T cell-mediated immunity and potential targets for immunotherapy and are patient-matched to PBMC TCR sequencing and Xenium spatial transcriptomics in separate GEO submissions. Overall design: T cells (CD45+CD3+) were sorted from 28 patients with head and neck and squamous cell carcinoma (HNSCC) using a Cytek Aurora Cell Sorter. Cells originating from each patient were labeled with one Totalseq-C human hashtag antibody (BioLegend). Four patient samples with distinct hashtags were pooled and loaded into one well of a GEM-X microfluidics chip (10x Genomics). 16 GEM-X reactions were performed, run on two independent GEM-X microfluidics chips. Pools 1-7 were replicated on both chips, and two pools (pool 8 and pool 16) were each run on one chip. Hashes used for each patient and the patient composition of each pool are described in the patient_matrix.txt supplementary file. Hashtag sequences are as follows: Hashing antibody sequences: hashtag 1 - GTCAACTCTTTAGCG; hashtag 2 - TGATGGCCTATTGGG; hashtag 3 - TTCCGCCTCTCTTTG; hashtag 4 - AGTAAGTTCAGCGTA. Gene expression (GEX), T cell receptor (TCR), and CITE-seq libraries were generated for each pool according to the manufacturer's instructions. For gene expression libraries, processed files include the aggregated gene expression and CITE-seq matrix from CellRanger. aggregation.csv lists the order of aggregation. A cloupe file of processed gene expression data for use in 10x Genomics' Loupe Browser software is also included. For T cell receptor (TCR) libraries, processed files for each sample are the folder output by cellranger vdj for each sample. After completion of the study, one patient (patient #10) received a final diagnosis of ameloblastoma.