Trapping and sorting of circulating tumor cells in a microfluidic filter array

Abstract

The presence and enumeration of Circulating Tumor Cells (CTCs) in peripheral blood has been identified as an indicator for early detection and treatment efficacy of various forms of cancer. Physical property-based isolation techniques of CTCs by microfluidic filters have emerged as a method capable of providing viable cells suitable for further study. Numerical simulations of a cell passing through microfil- ters provide insight into cell deformation and filter device operating characteristics. In this study, a microfluidic filter array device is proposed to sort the CTCs ac- cording to their physical properties along with isolation. The array device consists of three conical filter funnels having decreasing pore size connected in series. Cell passage through the filter funnels was computationally simulated by two-phase fluid flow with the cell modeled as a Newtonian droplet having constant cortical tension and viscosity. The CTCs were considered to be suspended in the blood plasma medium. Cell trapping and sorting phenomena in the filter array device based on cell size and cortical tension were investigated for different filter inlet driving pres- sures. The properties were selected to reflect the physical properties of RT4 bladder cancer cell, nontumorigenic breast cell MCF-10A and breast cancer cells MDA-MB- 231 and MDA-MB-468. CTCs were found to be trapped and sorted in the filter funnels according to their physical properties. Stiffer and larger cells were trapped early in funnels within a larger pore radius. Whereas softer and smaller cells can travel forward to be trapped in funnels with smaller pore sizes. An increase in the driving pressure resulted in a different sorting pattern, with all the cells being able to move further towards smaller funnels. Some of the cells passed through all the funnels to the exit chamber. Cell deformation in funnels was also quantified to show larger deformation in funnels with smaller pore radius. To characterize the filter array device, the threshold pressure required for passing through filters of different pore sizes was also investigated. The threshold pressure showed variation with filter pore size and cell physical properties. Smaller-sized filter funnels were found to require higher pressure to pass. The threshold pressure was also high for cells with large size and cortical tension. Based on the threshold passing pressures, filter char- acteristics were established to show that for a given filter geometry, different inlet driving pressure can be used to sort CTCs by different ranges of physical properties. The results provide a framework to customize microfluidic filter array device design and operation for selective detection and physical property-based sorting of CTCs, which have significant clinical importance in cancer diagnosis and treatment.