Flow cytometry detection of circulating tumor cells: Achievements and limitations as prognostic parameters


  • Henning Ulrich,

    1. Departamento de Bioquímica, Instituto de Química, Universidade de São Paulo, Brasil
    Search for more papers by this author
  • Attila Tárnok

    Corresponding author
    1. Department of Pediatric Cardiology, Heart Centre Leipzig, University of Leipzig, Leipzig, Germany
    2. Translational Centre for Regenerative Medicine (TRM), University of Leipzig, Leipzig, Germany
    • Correspondence to: Prof. Attila Tárnok, Dept. Pediatric Cardiology, Heart Centre Leipzig, University of Leipzig, Strümpellstr. 39, 04289 Leipzig, Germany. E-Mail: tarnok@medizin.uni-leipzig.de

    Search for more papers by this author

  • This work was made possible by funding from the German Federal Ministry of Education and Research (BMBF: PtJ-Bio); #1315883 [to A.T.].

Detection and enumeration of circulating tumor cells (CTCs) in blood are of substantial relevance in oncology as the results obtained are strong predictors for overall survival and critical for clinical decision making. Not only that, but CTC counting could be extremely useful in therapy monitoring. Due to the extremely low frequency of CTCs of ∼100 −1,000 cells/L, there is an ongoing effort to improve sensitivity, specificity and the limit of detection of clinical analysis systems. Competing methods are image cytometry (IC) and cell enrichment followed by quantitative PCR (for an overview see 1).

To date, the sole IC system that is approved by the U.S. Food and Drug Administration (U.S.FDA), is the Cell Search System [2] which relies on some morphological criteria as well as the positivity or negativity for certain markers. Other relevant, and from the detection point, comparable IC technologies and among others are the CellTracks, TDI, and CellTracks Analyzer II [3]. In addition, high quality flow cytometry (FCM) methods have also been developed, such as the Fishman-R microfluidic FCM [4] and new markers for specific CTCs, such as epidermal growth factor receptor (EGFR) and phosphorylated EGFR, have been proposed [5].

Watanabe and coworkers from Shizouka and Tokyo, Japan (this issue, page 206) used the Fishman-R FCM. The authors report on their innovative research for detecting these rare tumor cells, which detach from their original tumors or metastases and circulate in the blood of patients. Biomarkers expressed by these cells provide prognostic tools for cancer progression and treatment. Diagnostic platforms, including the FDA-approved, semi-automated Cell SearchTM system for prediction of therapeutic outcome of metastatic breast, prostate or colorectal cancers, mostly rely on detection of epithelial cell adhesion molecule (EpCAM) by a monoclonal antibody, the absence of the pan-leukocyte marker CD45, and the expression or secretion of tumor type-specific markers [1]. However, care must be taken in such an approach in view of varying EpCAM expression levels caused by epithelial-to-mesenchymal transition (EMT) of disseminating cancer and tumor stem cells.

In view of the unresolved sensitivity issue due to down-regulation of EpCAM, Watanabe etal. applied their recently developed FISHMAN-R flow cytometry system [4], which uses a microfluidic chip for cross-contamination-free measurement and quantification of the entire sample and permits the collection of the measured sample for multi-color detection. Immunostaining was performed with anti-CD45-Alexa Fluor 700, anti-EpCAM and anti-cytokeratin FITC antibodies, together with 7-actinomycin-D for nuclear staining. Three human tumor cell lines, KATO-III gastric tumor cells, non-small cell lung cancer A549 cells and PC-14 cells known for high, intermediate and no EpCAM expression, respectively, were used for spiking human blood samples at various dilutions. Cell populations were cleaned by immunomagnetic depletion of CD45+ white blood cells, and then subjected to microfluidic chip multiplex flow cytometry. As an outcome of this study, CTCs were – independently of their EpCAM expression levels – detected at mean sensitivities of 92% EpCAM-/CK- cells. Furthermore, EMT-induced tumor cells could be detected as the matter of anti-vimentin-immunostaining, further providing evidence for the suitability of the method in detecting and quantifying rare tumor cells, as shown by Watanabe and colleagues.

However, a considerable amount of patient samples needs to be processed to affirm whether the method is suitable for clinical use and diagnostics. It is important to mention that besides the issues of marker selectivity for tumor cells and sensitivity and specificity of the detection, CTC marker kinetics accompanying disease development have not yet been elucidated. As an example, PSA levels, used as prognostic biomarkers for life expectancy of castration-resistant prostate carcinoma patients, may not follow the treatment progress induced by cytotoxic drugs. Such observations suggest that relevant prognostic biomarkers as well as the features of CTC of various cancer types need yet to be defined. This can be achieved by molecular profiling using interdisciplinary approaches of molecular and protein biology (FISH, microarray analysis, and differential protein expression profiling) along with imaging and flow cytometry analysis, FACS or immunomagnetic sorting and cell culture techniques of isolated tumor cells (see ref. [6] for a comprehensive review).

Finally, in our view, cut-off counts for CTCs that are set to >5 CTCs/L for the Cell Search System method should be critically and repeatedly revisited. With regard to the sensitivity and specificity of the assays an error rate of at least 10% can be estimated, making the cut-off questionable, especially when it is a decision between therapy or not, and whether to pass the information about (hypothetical) life expectancy on to patients and their relatives or not. In conclusion, there is an ongoing need to improve the recovery and specificity and also to ensure if the cut-offs are indeed at the same level for all tumor types tested.