136 Dissecting intratumoral heterogeneity and microenvironment interactions in SCC through single-cell RNA-sequencing

Abstract

Diverse communities of malignant and non-malignant cells enable tumorigenesis in cutaneous squamous cell carcinoma (SCC) and other human cancers. Single-cell studies have identified substantial intratumoral heterogeneity, including tumor subpopulations that mediate clinical aggressiveness and therapeutic resistance. In addition to identifying tumor subpopulations that drive malignant behavior, single-cell RNA-sequencing (scRNA-seq) may reveal the degree to which normal gene expression programs, such as differentiation, are subverted across subpopulations, thereby providing unique insights into normal tissue homeostasis. Equally important, the tumor microenvironment (TME) plays a considerable role in tumorigenesis and therapeutic approaches, and delineation of crosstalk among diverse cell types in the TME can be greatly facilitated by single-cell analyses. We therefore undertook scRNA-seq on ∼47,000 cells from a series of 10 primary SCCs from unrelated human subjects, along with patient- and site-matched normal skin controls. Our analyses reveal a dysregulated differentiation hierarchy in SCC that allows for maintenance of an undifferentiated cell state and yields a tumor-specific subpopulation (TSS). The TSS resembles a potentially invasive subpopulation and is enriched for genes encoding ligands and receptors which mediate communication to various cell types in the TME. This network also denotes potential therapeutic vulnerabilities. A TSS expression signature correlates with poor overall survival across several epithelial cancer types, indicating that specific tumor subpopulations may disproportionately influence prognosis and warrant more specific therapeutic targeting. Therefore, scRNA-seq of human patient samples can not only provide insight into the complex cellular ecosystem required for tumor maintenance, but also reveal therapeutic vulnerabilities of the utmost necessity to help prioritize therapeutic targeting.