Microfluidic continuum sorting of sub-populations of tumor cells via surface antibody expression levels.

The extent of inter- and intra-tumor cell heterogeneity observed in patient tumors appears to be directly associated with patient prognosis. Moreover, studies indicate that targeting distinct subpopulations of tumor cells may be more relevant to successfully managing cancer metastasis. The ability to distinguish and characterize unique tumor cell subpopulations within a given sample is thus exigent. Existing platforms separate cells binarily, based on some threshold level of phenotypic characteristics without consideration of the continuum levels of biomarker expression and the associated implications. Herein we describe how specific tumor cell groups have been immunomagnetically enriched according to a continuum of EpCAM surface marker expression levels. Even among a relatively homogenous group of cells such as the PANC-1 cell line, cells could be separated according to their EpCAM levels into low, moderate and high expression. To physiologically assess each subpopulation, a wound healing assay was performed which revealed distinct invasive potentials among each subset. Furthermore, the clinical relevance of the approach was demonstrated by isolating pancreatic cancer CTCs from the same patient sample based on their EpCAM levels. We demonstrate a robust method of isolating CTCs according to their varying protein levels, which enables extensive studies on tumor cell heterogeneity. Interestingly, 5 of 6 samples had CTCs that could be recovered at all three levels of EpCAM expression though the majority of CTCs were recovered as low expression events. Preliminary studies that compare tumor cell subpopulations in this continuum manner can potentially increase our understanding of the dynamic nature of cell heterogeneity and how it relates to patient outcomes. Ultimately further investigation may yield therapeutic targets against virulent cell subpopulations.