Two-Step Acoustic Cell Separation Based on Cell Size and Acoustic Impedance─toward Isolation of Viable Circulating Tumor Cells

Abstract

Isolation and characterization of circulating tumor cells (CTCs) present a noninvasive alternative to monitor disease progression in individual patients. However, the heterogeneous lineage specificity of CTCs makes it difficult to isolate and identify possible CTCs by a liquid biopsy. Better label-free methods for the isolation of viable CTCs are needed. Our solution is a combined approach that is inherently epitope independent. Cells are separated by size-sensitive acoustophoresis using an ultrasonic standing wave field, followed by size-insensitive, acoustic barrier-medium focusing, which enables the enrichment of viable cancer cells in blood. With standard acoustophoresis in homogeneous medium, lymphocytes and monocytes were efficiently removed, while removal of granulocytes from the target MCF7 breast cancer cells was not possible due to overlapping acoustic migration velocities for viable cells. Remaining granulocytes were removed by a second separation step with an acoustic impedance barrier-medium selectively blocking the transport of MCF7 cells to generate a clean cancer cell fraction. For two series of 500 mL samples containing 5 × 10⁵ white blood cells, spiked with 2 × 10⁴ or 1 × 10³ MCF7 cells, the recovery of MCF7 cells was 77.3% with a 99.9% depletion of white blood cells in the final cancer cell fraction. The most abundant contaminating cell type was granulocytes (85.9% of remaining cells). Nearly all lymphocytes (99.996%) and monocytes (99.995%) were depleted. A two-step acoustic cell separation based on cell size and acoustic impedance is well suited to generate a purified cancer cell fraction as a preparatory step for downstream single-cell analysis.

Figure 1

Renderings of the acoustofluidic platforms and cell separation principles. (A) Acoustophoresis chip with prefocusing and a subsequent separation channel. Scale bar: ∼1 cm. (B) Isoacoustic barrier separation chip. Scale bar: ∼1 cm. (C) Schematic of the prefocusing and the subsequent acoustophoresis principle. (D) Schematic of isoacoustic barrier separation based on acoustic impedance (Z).

Figure 2

Separation of paraformaldehyde fixed cells versus nonfixed cell separation. Flow cytometer forward/side scatter dot plot of (A) PFA fixed WBC and MCF7 cells and (B) non-PFA fixed cells. (C) Comparison of separation performance by acoustophoresis of PFA-fixed WBCs and MCF7 cells versus nonfixed cells. (D) Separation performance by acoustophoresis of different WBC subgroups for PFA-fixed and nonfixed cells.

Figure 3

Acoustic medium-barrier separation of white blood cells and cancer cells. (A) Cancer cell and granulocyte separation performance in media of different iodixanol concentrations. (B) Cancer cell and granulocyte separation performance at different cell concentrations.

Figure 4

Optimizing settings for the acoustic medium-barrier separation of white blood cells and cancer cells. (A) Acoustic medium-barrier separation of MCF7 cells and WBC with 16% iodixanol as central medium. (B) WBC subgroups from panel (A). (C) Effects of 6% iodixanol as separation buffer on separation of MCF7 cells and WBCs in the size-based separation setup. (D) Effect on cell viability when running a two-step separation.

Figure 5

Two-step separation of white blood cells and cancer cells. Six samples (0.5 mL) with 500,000 WBCs spiked with either (A) 20,000 MCF7 cells or (B) 1000 MCF7 cells. Samples were run through first size-based acoustophoresis followed by medium-barrier acoustic separation.