Stiffness based enrichment of leukemia cells using microfluidics

Muhymin Islam; Abhishek Raj; Brynn McFarland; Hannah Maxine Brink; Jordan Ciciliano; Meredith Fay; David Richard Myers; Christopher Flowers; Edmund K Waller; Wilbur Lam; Alexander Alexeev; Todd Sulchek

doi:10.1063/1.5143436

Stiffness based enrichment of leukemia cells using microfluidics

APL Bioeng. 2020 Jul 1;4(3):036101. doi: 10.1063/1.5143436. eCollection 2020 Sep.

Affiliations

¹ George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, 801 Ferst Drive, Atlanta, Georgia 30332-0405, USA.
² The School of Biological Sciences, Georgia Institute of Technology, 310 Ferst Drive, Atlanta, Georgia 30332-0535, USA.
³ Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology, 313 Ferst Drive, Atlanta, Georgia 30332-0535, USA.
⁴ Winship Cancer Institute, Emory School of Medicine, 1365 Clifton NE Rd.: Atlanta, Georgia 30322, USA.

Abstract

To improve the survival rate of cancer patients, new diagnosis strategies are necessary to detect lower levels of cancer cells before and after treatment regimens. The scarcity of diseased cells, particularly in residual disease after treatment, demands highly sensitive detection approaches or the ability to enrich the diseased cells in relation to normal cells. We report a label-free microfluidic approach to enrich leukemia cells from healthy cells using inherent differences in cell biophysical properties. The microfluidic device consists of a channel with an array of diagonal ridges that recurrently compress and translate flowing cells in proportion to cell stiffness. Using devices optimized for acute T cell leukemia model Jurkat, the stiffer white blood cells were translated orthogonally to the channel length, while softer leukemia cells followed hydrodynamic flow. The device enriched Jurkat leukemia cells from white blood cells with an enrichment factor of over 760. The sensitivity, specificity, and accuracy of the device were found to be $> 0.8$ . The values of sensitivity and specificity could be adjusted by selecting one or multiple outlets for analysis. We demonstrate that low levels of Jurkat leukemia cells (1 in $10^{4}$ white blood cells) could be more quickly detected using flow cytometry by using the stiffness sorting pre-enrichment. In a second mode of operation, the device was implemented to sort resistive leukemia cells from both drug-sensitive leukemia cells and normal white blood cells. Therefore, microfluidic biomechanical sorting can be a useful tool to enrich leukemia cells that may improve downstream analyses.