Summary
Published in Science Advances (2024), this peer-reviewed study used 3D Petri Dish® micro-molds to form scaffold-free 3D microtissues. Full citation: Goldstein, Yoel, et al. Particle uptake in cancer cells can predict malignancy and drug resistance using machine learning
Particle uptake in cancer cells can predict malignancy and drug resistance using machine learning
Research Overview
Tumor heterogeneity is a primary cause of treatment failure, so tools that classify cancer cells by function could meaningfully extend patient survival. This study exploited the link between cell biomechanics and cancer cell function, classifying cells through mechanical measurement — specifically their pattern of particle uptake.
Three pairs of human cancer cell subpopulations differing in drug resistance or malignancy were exposed to fluorescently labeled polystyrene particles ranging from 0.04 to 3.36 µm, and machine learning classified cells from their single-cell uptake patterns — a label-free functional fingerprint of aggressive versus sensitive tumor cells.
Key Discoveries
- Machine learning classified cancer subpopulations from single-cell particle uptake patterns
- Particles spanning 0.04 to 3.36 µm probed biomechanical differences between cells
- Distinguished pairs differing in drug resistance or malignancy without molecular labels
