Profiling the immune infiltrate in tumor samples at single cell resolution

Abstract Profiling the complex interactions of immune infiltrate with tumor cells in a tumor microenvironment is critical for advancing our understanding of tumor biology for developing personalized cancer therapies. Using a droplet-based single cell RNA sequencing (scRNA-seq) platform, we profiled the transcriptome and immune repertoire of gastric, kidney, and lung cancer cells that are primarily immune cells from three different donors. scRNA-seq analysis also identified similar fractions of CD45+ immune cells, CD4+ and CD8+ T cells, CD19+ B cells, and myeloid cells compared to flow. Targeted scRNA-seq was also used to identify paired, full length B-cell (BCR) and T-cell (TCR) receptors. In the gastric adenocarcinoma tumor cells, gene expression analysis identified a large population of B cells, but with no clonal expansion. T cells expressing CD8A constituted about 10% of cells with the top clonotype being 1% of all clones. The clear cell RCC sample had a modest fraction of infiltrating T cells with the top clonotype representing 7.4% of all clones, and no B cell infiltrate. The NSCLC cells were mainly T and B cells but limited clonal expansion was observed; the top TCR clonotype was present on 2.2% of all T cells, no expansion was seen in any of the B cell clonotypes. These findings highlight the value of profiling of tumor immune cells holistically and not relying on the presence of B or T cells alone to understand the immune dynamics of the tumor microenvironment. The presence of tumor-infiltrating lymphocytes is associated with favorable clinical outcomes in some cancers, but understanding their cellular subtype and clonality by high resolution profiling is key in the development of immune-based cancer treatments