Decoupling epithelial-to-mesenchymal transitions from stromal profiles by integrative analysis

Epithelial-to-mesenchymal transition (EMT) is the most commonly cited mechanism for cancer metastasis, but it is difficult to distinguish from profiles of normal stromal cells in the tumour microenvironment. In this study we used published single cell RNA-seq data to directly compare expression of mesenchymal signature genes in cancer and stromal cells. Informed by these comparisons, we developed a method to deconvolve the mesenchymal signature in bulk RNA-seq data into `true` EMT and stromal components. This offers an opportunity for a pan-cancer analysis of EMT using the wealth of readily available bulk tumour transcriptome data, enabling a characterisation of EMT across hundreds of tumours and an examination of its association with metastasis and other clinical features. We show that classical EMT marker genes often primarily reflect stromal content, while our inferred cancer-cell-specific EMT signatures usually do not correlate with metastasis. This study demonstrates the importance of distinguishing ‘true’ EMT from stromal contributions in order to elucidate the therapeutic relevance of EMT in cancer.