Data-Driven Sustainable In Vitro Campaigns to Decipher Invasive Breast Cancer Features

The intrinsic complexity of biological processes often hides the role of dynamic microenvironmental cues in the development of pathological states. Microphysiological systems (MPSs) are emerging technological platforms that model in vitro dynamics of tissue-specific microenvironments, enabling a holistic understanding of pathophysiology. In our previous works, we engineered and used breast tumor MPS differing in matrix stiffness, pH, and fluid flow mimicking normal and tumor breast tissue. High-dimensional data using two distinctive human breast cell lines (i.e., MDA-MB-231, MCF-7), investigating cell proliferation, epithelial-to-mesenchymal transition (EMT), and breast cancer stem cell markers (B-CSC), were obtained from breast-specific microenvironments. Recognizing that the widespread adoption of MPS requires tailoring its complexity to application demands, we herein report an innovative machine-learning (ML)-based approach to analyze MPS data. This approach uses unsupervised k-means clustering and feature extraction to inform on key markers and specific microenvironments that distinguish invasive from non-invasive breast cell phenotypes. This data-driven approach streamlines future experimental design and emphasizes the translational potential of integrating MPS-derived insights with ML to refine prognostic tools and personalize therapeutic strategies.