Breast cancers are a complex ‘ecosystem’ of diverse cell types, whose heterotypic interactions between malignant epithelial, stromal and immune populations are central in defining the aetiology of the disease and response to therapy. Despite advances in other carcinomas, combinational therapies that target the supporting microenvironment have made little clinical progress in breast cancer. Such therapies have been largely impeded by a poor understanding of the cellular heterogeneity in breast cancers, which is completely masked by conventional bulk sequencing studies.
Single-cell RNA sequencing has emerged as a remarkable tool for studying diverse cellular populations amongst complex tissues. In this study, we apply this to sequence more than 200,000 cells sampled from over 30 primary and metastatic breast cancers. At single-cell resolution, we describe the neoplastic, immune and stromal landscape, providing powerful new tissue- and disease-specific cell type signatures. In particular, we focus on heterogeneous cancer-associated fibroblast (CAF) populations, and comprehensively profile distinct gene expression between subsets, cell signaling interactions with microenvironment and polarised gene-regulatory states. We show these subsets highly contribute to worse overall survival and immune dysfunction in large clinical cohorts, narrowing down on several candidate therapeutic targets. To deeply phenotype the microenvironment, we have optimised the reported CITE-SEQ method to validate novel cell-surface markers, allowing for the prospective isolation and functional investigation of our CAF subsets. Altogether, our study proposes that diverse subsets of CAFs contribute to multi-faceted roles in regulating malignancy and immune-suppression, revealing new strategies for CAF-directed therapies in breast cancer.
This is by far the largest single cell genomic study in any cancer to date. Our study highlights the power of single cell genomics to unravel the complexities of the tumour microenvironment and identify novel mechanisms underlying carcinogenesis. Such insights will guide the next-generation of therapies, which will likely be based upon an integrated understanding of the neoplastic, stromal and immune states that define a tumour and inform treatment response.