Morphologic variation of circulating tumor cells in metastatic breast cancer patients visualized by immunofluorescent stain pattern.

Abstract

e21079 Background: Isolation of circulating tumor cells (CTCs) using microfiltration is a growing utility in the field of CTC detection. The microfiltration approach can be used on peripheral blood as a non-invasive liquid biopsy for cancer detection and subtyping. We present a utilization of the CellSieve microfilter to subtype CTCs based on immunofluorescent staining pattern of cytokeratin filamentation and EpCAM surface marker expression. Our initial study on CTCs in patient blood indicates that disseminated CTC populations have high rates of phenotypic heterogeneity. Further detailed molecular analysis and patient tracking of these phenotypes may lead to individualized patient assessment based on CTC characterization. Methods: 7.5 mL of whole blood collected from metastatic breast cancer patients were diluted in a fixative solution. An 8 µm CellSieve precision microfilter was placed into a filter holder and the samples were passed through the filter (~ 90 seconds). The cells captured on the filter were fixed, permeabilized, and stained with DAPI, cytokeratin (FITC), EpCAM (PE), and CD45 (Cy5). Cells without CD45 staining were classified by their morphology, nuclear integrity and the presence of cytokeratin and EpCAM staining. Results: In our initial assessment, patient samples were found to have a number of phenotypic CTC subtypes. Cytokeratin filamentation was clearly seen on a number of CTCs while other cells presented with spotted patterns, implying CTCs in various stages of apoptosis. Later stage apoptosis, with segmented nuclear signature, was also seen in various samples. Cell clusters and cells in division were DAPI positive, while EpCAM positivity was negligible, 0-3 cells/sample, correlating with established data from the CellSearch CTC assay. Conclusions: In addition to enumeration, phenotypic variation of CTCs may be a valuable tool for the personalized care of cancer patients. We have shown that individual breast cancer patients have overlapping phenotypes of CTCs circulating in their peripheral blood. We have begun categorizing patients based on these phenotypes and plan to correlate them with overall prognosis.