Optimization and Evaluation of a Novel Size Based Circulating Tumor Cell Isolation System.

Abstract

Isolation of circulating tumor cells (CTCs) from peripheral blood has the potential to provide a far easier “liquid biopsy” than tumor tissue biopsies, to monitor tumor cell populations during disease progression and in response to therapies. Many CTC isolation technologies have been developed. We optimized the Parsortix system, an epitope independent, size and compressibility-based platform for CTCs isolation, making it possible to harvest CTCs at the speed and sample volume comparable to standard CellSearch system. We captured more than half of cancer cells from different cancer cell lines spiked in blood samples from healthy donors using this system. Cell loss during immunostaining of cells transferred and fixed on the slides is a major problem for analyzing rare cell samples. We developed a novel cell transfer and fixation method to retain >90% of cells on the slide after the immunofluorescence process without affecting signal strength and specificity. Using this optimized method, we evaluated the Parsortix system for CTC harvest in prostate cancer patients in comparison to immunobead based CTC isolation systems IsoFlux and CellSearch. We harvested a similar number (p = 0.33) of cytokeratin (CK) positive CTCs using Parsortix and IsoFlux from 7.5 mL blood samples of 10 prostate cancer patients (an average of 33.8 and 37.6 respectively). The purity of the CTCs harvested by Parsortix at 3.1% was significantly higher than IsoFlux at 1.0% (p = 0.02). Parsortix harvested significantly more CK positive CTCs than CellSearch (p = 0.04) in seven prostate cancer patient samples, where both systems were utilized (an average of 32.1 and 10.1 respectively). We also captured CTC clusters using Parsortix. Using four-color immunofluorescence we found that 85.8% of PC3 cells expressed EpCAM, 91.7% expressed CK and 2.5% cells lacked both epithelial markers. Interestingly, 95.6% of PC3 cells expressed Vimentin, including those cells that lacked both epithelial marker expression, indicating epithelial-to-mesenchymal transition. CK-positive/Vimentin-positive/CD45-negative, and CK-negative/Vimentin-positive/CD45-negative cells were also observed in four of five prostate cancer patients but rarely in three healthy controls, indicating that Parsortix harvests CTCs with both epithelial and mesenchymal features. We also demonstrated using PC3 and DU145 spiking experiment that Parsortix harvested cells were viable for cell culture.