Introduction:Epithelial circulating tumor cells (CTCs) are detectable in patients with non-small cell lung cancer (NSCLC). However, epithelial to mesenchymal transition, a widely reported prerequisite for metastasis, may lead to an underestimation of CTC number. We compared directly an epithelial marker-dependent (CellSearch) and a marker-independent (isolation by size of epithelial tumor cells [ISET]) technology platform for the ability to identify CTCs. Molecular characteristics of CTCs were also explored.Methods:Paired peripheral blood samples were collected from 40 chemonäive, stages IIIA to IV NSCLC patients. CTCs were enumerated by Epithelial Cell Adhesion Molecule-based immunomagnetic capture (CellSearch, Veridex) and by filtration (ISET, RareCell Diagnostics). CTCs isolated by filtration were assessed by immunohistochemistry for epithelial marker expression (cytokeratins, Epithelial Cell Adhesion Molecule, epidermal growth factor receptor) and for proliferation status (Ki67).Results:CTCs were detected using ISET in 32 of 40 (80%) patients compared with 9 of 40 (23%) patients using CellSearch. A subpopulation of CTCs isolated by ISET did not express epithelial markers. Circulating tumor microemboli (CTM, clusters of ≥3 CTCs) were observed in 43% patients using ISET but were undetectable by CellSearch. Up to 62% of single CTCs were positive for the proliferation marker Ki67, whereas cells within CTM were nonproliferative.Conclusions:Both technology platforms detected NSCLC CTCs. ISET detected higher numbers of CTCs including epithelial marker negative tumor cells. ISET also isolated CTM and permitted molecular characterization. Combined with our previous CellSearch data confirming CTC number as an independent prognostic biomarker for NSCLC, we propose that this complementary dual technology approach to CTC analysis allows more complete exploration of CTCs in patients with NSCLC.

Analysis of Circulating Tumor Cells in Patients with Non-small Cell Lung Cancer Using Epithelial Marker-Dependent and -Independent Approaches