Abstract 2396: Key factors for filtration of tumor cells from whole blood

Abstract

Abstract Introduction: Circulating tumor cells (CTC) in patients with metastatic carcinomas are associated with poor survival. The current standard CTC enrichment method employs antibodies, which misses cells not expressing the target antigen. Several alternative methods enrich CTC by filtration, exploiting differences in size between blood cells and CTC. Between different filtration methods large variations in material, pore size, operating pressure and sample fixation exist. As their influence for CTC enrichment is poorly understood we have investigated these parameters. Methods: Track etched polycarbonate filters with pore diameters of: 5, 8 and 10 µm and microfabricated silicon nitride filters with pore diameters of 5, 6, 7, 8, 9 and 10 µm were used. 300 cells from the cell lines SKBR3, PC3-9, MDA-231, MDA-468, MCF-7, SW480, HL-60 or K-586 were spiked in 1 ml of blood from healthy volunteers. Samples were filtered while both pressure across the filter membrane and the flow speed was measured. The pressure across the filter provides the force driving a cell through a filter pore. Cells were identified on the filters by fluorescence microscopy. Results: MDA-231, a 15 µm sized epithelial cell line, can readily pass through a 5 µm pore at low pressure (< 50 mbar). We show that the number of red and white blood cells contribute most to the pressure across a filter and at least 10e5 8 µm pores are needed to process 0.5 mL of whole blood per minute at 10 mbar of pressure. The influence of buffers such as PBS or serum was found to be negligible. Fixation using paraformaldehyde increases the pressure needed to push white, red and culture cells through a pore by 25 fold and recovery with spiked samples decreases from 85% to 40% at the same flow rate. Different filter types were compared and evaluated for cell line recovery using unfixed blood. Optimal recovery of cell lines was between 80-90% with low WBC contamination for track etch filters with 8 µm pores and 5 µm pores for silicon nitride chips. Only a weak correlation was found between recovery and size for the eight different cell lines. Conclusions: There are big differences in the ease with which different culture cell lines of similar size pass a filter pore; a challenging cell line such as MDA-231 should be selected for evaluation of filtration methods. Pressure across the filter is an important parameter in determining whether a cell passes a filter pore. Enrichment without fixation is best and should be performed at low pressures (< 50 mbar). A sufficiently small pore retains most of the tumor cells, but passes most white and all red cells. The optimal pore size was different for the different filter types evaluated, possibly due to surface of the filter or filter thickness. In the optimal conditions recoveries of at least 80% were found for most cell lines across a wide range of CTC concentrations and blood volumes. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2396. doi:1538-7445.AM2012-2396