The epithelial–mesenchymal transition (EMT) program is critical to metastatic cancer progression. EMT results in the expression of mesenchymal proteins and enhances migratory and invasive capabilities. In a small percentage of cells, EMT results in the expression of stemness-associated genes that provide a metastatic advantage. Although EMT had been viewed as a binary event, it has recently become clear that the program leads to a spectrum of phenotypes, including hybrid epithelial/mesenchymal (E/M) cells that have significantly greater metastatic capability than cells on the epithelial or mesenchymal ends of the spectrum. As hybrid E/M cells are rarely observed in physiological, non-diseased states in the adult human body, these cells are potential biomarkers and drug targets. Hybrid E/M cells are distinguished by the co-expression of epithelial and mesenchymal proteins, such as the intermediate filament proteins cytokeratin (CK; epithelial) and vimentin (VIM; mesenchymal). Although these intermediate filaments have been extensively used for pathological characterization and detection of aggressive carcinomas, little is known regarding the interactions between CK and VIM when co-expressed in hybrid E/M cells. This review describes the characteristics of hybrid E/M cells with a focus on the unique co-expression of VIM and CK. We will discuss the structures and functions of these two intermediate filament proteins and how they may interact when co-expressed in hybrid E/M cells. Additionally, we review what is known about cell-surface expression of these intermediate filament proteins and discuss their potential as predictive biomarkers and therapeutic targets.

Background: Aggressive forms of breast cancer, such as triple-negative breast cancer (TNBC), are associated with an increase in cancer cells that exhibit stem cell properties. The activation of the epithelial–mesenchymal transition (EMT) program, mediated by the transcription factor FOXC2, generates these stem-like cells. FOXC2 is linked to poor prognoses across various cancer types and is notably upregulated in TNBC, where it establishes and sustains these stem-like cells within the tumor population. Methods: Here, we decode the pathways regulating FOXC2 activation using EMT-enriched cell line models. Stemness was assessed using mammosphere assays and mesenchymal markers by western blot. Expression correlations with clinical data was examined using the EMTome. Results: We demonstrate that β-catenin serves as a critical mediator of mesenchymal and stemness characteristics through FOXC2 upregulation. By disrupting β-catenin, we find that FOXC2 expression, mesenchymal properties, and stemness are reduced; however, the introduction of exogenous FOXC2 expression in β-catenin deficient cells is enough to restore the mesenchymal and stemness phenotype. These findings support the idea that FOXC2 acts as the downstream regulator of β-catenin and influences both mesenchymal and stemness properties. Moreover, there is a positive correlation between the expression of β-catenin and FOXC2 in various cancer subtypes observed in clinical patient samples. Conclusions: Our study clarifies the role of the β-catenin/FOXC2 signaling axis in maintaining stemness properties, suggesting potential targets for TNBC and other cancers driven by EMT-related mesenchymal and stemness characteristics.