Date Published: August 28, 2015
Publisher: Public Library of Science
Author(s): Jian Yin, Yi Wang, Hanlu Yin, Wenping Chen, Guangfu Jin, Hongxia Ma, Juncheng Dai, Jiaping Chen, Yue Jiang, Hui Wang, Zhian Liu, Zhibin Hu, Hongbing Shen, Huei-Wen Chen.
It has been considered that the detection methods for circulating tumor cells (CTCs) based on epithelial cell adhesion molecule (EpCAM) underestimate the number of CTCs and may miss a metastatic subpopulation with cancer stem cell (CSC) properties. Therefore, we investigated EpCAM-positive and -negative CTCs in non-small cell lung cancer (NSCLC) patients at different stages, assessed the clinical value of these CTCs and explored their capacity in the following CSC model.
CTCs were enriched by the depletion of leukocytes with bi-antibodies using a magnetic bead separation technique and then identified by the expression of EpCAM and cytokeratin 7 and 8 using multi-parameter flow cytometry. We determined the distribution of CTCs classified by the expression of EpCAM in 46 NSCLC patients with stages I to IV, assessed the diagnostic value of these CTCs by longitudinal monitoring in 4 index patients during adjuvant therapy and characterized the stemness of these CTCs by the expression of CXCR4 and CD133 in 10 patients.
EpCAM-negative (E-) CTCs were detected to be significantly higher than EpCAM-positive (E+) CTCs in stage IV (p = 0.003). The patients with the percentage of E-CTCs more than 95% (r > 95%) were detected to be significantly increased from 13.3% in stage I-II to 61.1% in stage IV (p = 0.006). Kaplan–Meier analysis indicated that the patients with r > 95% had significantly shorter survival time than those with r ≤ 0.95 (p = 0.041). Longitudinal monitoring of CTCs indicated that the patients with a high percentage of E-CTCs in the blood were not responsive to either chemotherapy or targeted therapy. Further characterization of CTCs revealed that a stem-like subpopulation of CXCR4+CD133+ CTCs were detected to be significantly more prevalent in E-CTCs than that in E+CTCs (p = 0.005).
The enrichment of CTCs by the depletion of leukocytes with bi-antibodies is a valuable method for estimating the number of CTCs, which can be potentially applied in predicting the prognosis, monitoring the therapeutic effect of NSCLC patients and further analyzing the biology of CTCs.
Lung cancer is the leading cause of cancer-related death , and approximately 85% of lung cancer cases are non-small cell lung cancer (NSCLC). Although systematic treatment has been improved, the overall 5-year survival rate is only 10–20% . The primary reason for the low survival rate is distant metastasis of tumor cells. In the metastatic cascade, circulating tumor cells (CTCs) have been considered to be key participants in the formation of distant metastases . A previous study showed that CTCs expressing epithelial cell adhesion molecule (EpCAM) are detectable in stage IV NSCLC patients and are a novel prognostic factor for this disease . However, it has been suggested that the methods based on the expression of EpCAM underestimate the number of CTCs and may miss a metastatic subpopulation of CTCs with cancer stem cell (CSC) properties [5, 6]. A recent study reported that CTCs are detected 2 times more effectively by ISET (isolation by size of epithelial tumor cells) than those by CellSearch, and that a subpopulation of CTCs, which did not express EpCAM (i.e., E-CTCs), can be detected in the blood of NSCLC patients . Another study has also shown that the enumeration of these cells is much higher than that of CTCs captured by CellSearch . Until now, the clinical value and biology of these E-CTCs has been unclear, and a recent publication has indicated that future studies should include the detection of E-CTCs .
CTCs can be used as a prognostic marker [17, 18]. However, the small number and low concentration of CTCs has limited study of the biology of CTCs [9, 19]. Due to the heterogeneity of CTCs, it is difficult to capture all types of CTCs by EpCAM-based methods. A more reasonable strategy is to deplete leukocytes. Using our improved method, the purity of CTCs was increased by 10 times in spiking experiments, and up to 50 times higher in blood samples compared with a previously reported method based on magnetic nanobeads . Using this method, the mean value of the purity of CTCs could be up to 10% in spiking experiments and 10.6% in the blood samples of patients, which also supported its further application in molecular analysis. This method is comparable for the detection of CTCs with and without the expression of EpCAM. In the results, 36.4% (for E+CTCs) and 81.8% (for E-CTCs) of patients had stage IV disease. These data are consistent with the published results that E+CTCs were detected in only 23~32% of advanced NSCLC patients by CellSearch and 80% of those by size isolation strategy [4, 7]. This method is specific for the identification of CTCs with and without expression of EpCAM. Both E+ and E-CTCs were detected to be much lower in the blood samples from healthy volunteers relative to those in patients, which was also consistent with a previously published study . Although E+ and E-CTCs were undetected in five patients, it is possible that the enumeration of CTCs was lower than the cutoff values or absent from additional specific markers to identify the heterogeneous tumor cells in NSCLC [21, 22].
Based on a technique of CTC enrichment by the depletion of leukocytes with bi-antibodies, this study assessed the clinical value of CTCs with and without EpCAM expression in NSCLC. Our data suggest that the percentage of E-CTCs significantly increased from stage I-II to IV and could be used to evaluate the distant metastasis, therapeutic effect and prognosis of patients in NSCLC. Furthermore, E-CTCs appeared to be more stem-like than E+CTCs. These results provide the first evidence for the potential clinical application of E-CTCs in NSCLC.