Direct isolation of multiple types of circulating tumor cells from undiluted human blood using cascaded viscoelastic microfluidics.

Circulating tumor cells (CTCs) are rare cells detaching from metastasis or primary tumor and flowing into bloodstream, which has been a promising biomarker for non-invasive cancer monitoring and diagnosis. Although CTC separating techniques have been widely reported, few platforms practically realize high-throughput and precise isolation of multiple types of CTCs directly from undiluted human blood due to complex flow behavior and interference from high concentration of blood cells. Herein, we present a rapid and label-free cascaded viscoelastic microfluidics (CVEM) composed of blood cells depletion module (BCDM) and CTCs isolation module (CIM). BCDM is used to filter out blood cells from samples, and remaining two types of CTCs (A549 and MCF-7) are isolated in CIM. To systematically reveal the separation mechanism, particle trajectory prediction model including criterion of crossing interface and particle dynamics equation is constructed, and the theoretical results are verified by polystyrene (PS) particle separation experiments. Further, the performance of CVEM is tested with four PS particles mixture solution and human blood spiked with A549 and MCF-7, respectively. 1 mL undiluted blood can be processed within 40 min, and both types of CTCs have high purity (≥81.27%) and recovery rate (≥82.15%). CVEM represents a powerful platform for sorting multiple types of CTCs from directly collected blood, demonstrating significant clinical application potential. Furthermore, particle trajectory prediction model provides a solid theoretical basis for optimizing viscoelastic microfluidics (VEM), and permits rapid iterative design according to different separating requirements, thereby realizing the flexible extension of VEM to other fields.