Micro- and Nanotechnologies

SLAS2018 Innovation Award Finalist: Label-free prediction of cancer cell invasion by single-cell physical phenotyping

• AR

  Amy Rowat, BSc, MSc, PhD

  Associate Professor
  UCLA

The physical properties of cells, such as cell deformability, are promising label-free biomarkers for cancer diagnosis and prognosis. Here we determine the physical phenotypes that best distinguish human cancer cell lines, and their relationship to cell invasion. We use the high throughput, single-cell microfluidic method, quantitative deformability cytometry (q-DC), to measure six physical phenotypes including elastic modulus, cell fluidity, transit time, creep time, cell size, and maximum strain at rates of 102 cells/s. By training a simple k-nearest neighbor machine learning algorithm, we demonstrate that multiparameter analysis of physical phenotypes enhances the accuracy of classifying pancreatic cancer cell lines compared to single parameters alone. We also discover a set of four physical phenotypes that predict invasion; using these four parameters, we generate the physical phenotype model of invasion by training a machine learning algorithm with experimental data from a set of human ovarian cancer (HEYA8) cells that overexpress a panel of tumor suppressor microRNAs. We validate the model using breast and ovarian human cancer cell lines with both genetic and pharmacologic perturbations. Our results reveal that the physical phenotype model correctly predicts the invasion of five cancer cell samples. We also identify a context where our model does not accurately predict invasion, which incites deeper investigation into the role of additional physical phenotypes in cancer cell invasion. Taken together, our results highlight how physical phenotyping of single cells coupled with machine learning provide a complementary biomarker to predict the invasion of cancer cells.