Cancer metastasis is the main cause of cancer-related death, and dissemination of tumor cells through the blood circulation is an important intermediate step that also exemplifies the switch from localized to systemic disease. Early detection and characterization of circulating tumor cells (CTCs) is therefore important as a general strategy to monitor and prevent the development of overt metastatic disease. Furthermore, sequential analysis of CTCs can provide clinically relevant information on the effectiveness and progression of systemic therapies (e.g., chemo-, hormonal, or targeted therapies with antibodies or small inhibitors). Although many advances have been made regarding the detection and molecular characterization of CTCs, several challenges still exist that limit the current use of this important diagnostic approach. In this review, we discuss the biology of tumor cell dissemination, technical advances, as well as the challenges and potential clinical implications of CTC detection and characterization.

BackgroundCirculating tumour cells (CTCs) have shown prognostic relevance in metastatic breast, prostate, colon and pancreatic cancer. For further development of CTCs as a biomarker, we compared the performance of different protocols for CTC detection in murine breast cancer xenograft models (MDA-MB-231, MDA-MB-468 and KPL-4). Blood samples were taken from tumour bearing animals (20 to 200 mm2) and analysed for CTCs using 1. an epithelial marker based enrichment method (AdnaTest), 2. an antibody independent technique, targeting human gene transcripts (qualitative PCR), and 3. an antibody-independent approach, targeting human DNA-sequences (quantitative PCR). Further, gene expression changes associated with epithelial-to-mesenchymal transition (EMT) were determined with an EMT-specific PCR assay.MethodsWe used the commercially available Adna Test, RT-PCR on human housekeeping genes and a PCR on AluJ sequences to detect CTCs in xenografts models. Phenotypic changes in CTCs were tested with the commercially available “Human Epithelial to Mesenchymal Transition RT-Profiler PCR Array”.ResultsAlthough the AdnaTest detects as few as 1 tumour cell in 1 ml of mouse blood spiking experiments, no CTCs were detectable with this approach in vivo despite visible metastasis formation. The presence of CTCs could, however, be demonstrated by PCR targeting human transcripts or DNA-sequences - without epithelial pre-enrichment. The failure of CTC detection by the AdnaTest resulted from downregulation of EpCAM, whereas mesenchymal markers like Twist and EGFR were upregulated on CTCs. Such a change in the expression profile during metastatic spread of tumour cells has already been reported and was linked to a biological program termed epithelial-mesenchymal transition (EMT).ConclusionsThe use of EpCAM-based enrichment techniques leads to the failure to detect CTC populations that have undergone EMT. Our findings may explain clinical results where low CTC numbers have been reported even in patients with late metastatic cancers. These results are a starting point for the identification of new markers for detection or capture of CTCs, including the mesenchymal-like subpopulations.

Circulating tumor cells (CTCs) were introduced as biomarkers more than 10 years ago, but capture of viable CTCs at high purity from peripheral blood of cancer patients is still a major technical challenge. Here, we report a novel microfluidic platform designed for marker independent capture of CTCs. The Parsortix™ cell separation system provides size and deformability‐based enrichment with automated staining for cell identification, and subsequent recovery (harvesting) of cells from the device. Using the Parsortix™ system, average cell capture inside the device ranged between 42% and 70%. Subsequent harvest of cells from the device ranged between 54% and 69% of cells captured. Most importantly, 99% of the isolated tumor cells were viable after processing in spiking experiments as well as after harvesting from patient samples and still functional for downstream molecular analysis as demonstrated by mRNA characterization and array‐based comparative genomic hybridization. Analyzing clinical blood samples from metastatic (n = 20) and nonmetastatic (n = 6) cancer patients in parallel with CellSearch® system, we found that there was no statistically significant difference between the quantitative behavior of the two systems in this set of twenty six paired separations. In conclusion, the epitope independent Parsortix™ system enables the isolation of viable CTCs at a very high purity. Using this system, viable tumor cells are easily accessible and ready for molecular and functional analysis. The system's ability for enumeration and molecular characterization of EpCAM‐negative CTCs will help to broaden research into the mechanisms of cancer as well as facilitating the use of CTCs as “liquid biopsies.”