ISENICS: a model for identifying senescent immune cells and samples and characterization of their roles in tumor microenvironment.

Abstract

Senescent immune cells secrete varied inflammatory factors that weaken the systemic anti-tumor ability and promote the proliferation and metastasis of tumor cells. Tumor cells could also accelerate the immune cellular senescence through diverse mechanisms. However, there has been a lack of indicators to quantify the senescence levels of different immune cell types. A model for Identifying Senescent Immune Cells and Samples was developed to explore the role of senescent immune cells in the tumor immune microenvironment (TIME). By integrating bulk and single-cell RNA-seq data, we constructed immune cell gene expression profiles for 23 cancer types using a deconvolution algorithm. By calculating the cellular senescence scores, we found that tumor samples exhibited higher senescence levels than normal samples. Monocytes/macrophages were prone to co-senescence with other cell subtypes. Differentially expressed genes in the high- and low-immune cellular senescence scores groups were enriched in the senescence pathway. Patients with higher levels of immunosenescence were associated with better prognosis. At the single-cell level, the number and strength of cell-to-cell interactions increased following immune cellular senescence in most cancers. Samples with senescent immune cells exhibited poorer immunotherapy response. Our study advances our understanding of senescent immune cells in the TIME, provides insights into cancer-specific relationships between immune cellular senescence and immune characteristics, and offers a model for identifying these senescent immune cells.