Enumeration and molecular characterization of circulating tumor cell using an in vivo capture system in squamous cell carcinoma of head and neck

OBJECTIVE Detection rate and isolation yield of circulating tumor cell (CTC) are low in squamous cell carcinoma of head and neck (SCCHN) with in vitro approaches due to limited sample volumes. In this study, we applied the CellCollector to capture CTC in vivo from peripheral blood. METHODS In total, the study included 22 cases with 37 times of detection. All of the patients were newly diagnosed with locally advanced or metastatic SCCHN, including laryngocarcinoma (40.9%, 9/22) and hypopharyngeal carcinoma (59.1%, 13/22). All patients received CTC analysis before treatment. Three patients received induction chemotherapy. Sixteen patients received surgical therapy, of which 13 patients received postoperative detection. Two patients received both induction chemotherapy and surgery treatment. Patients underwent two successive CellCollector applications 24 h before and 7 d after surgical therapy. Nine healthy volunteers were enrolled as the control group. Epidermal growth factor receptor variant type III (EGFRVIII) expression was analyzed with fluorescent dye labeled antibody. RESULTS With CellCollector isolation, 72.7% (16/22) of the patients were positive for ≥1 CTC (CTC; range, 1-17 cells) before treatments and 46.7% (7/15) of patients were CTC positive for ≥1 CTC (CTC; range, 1-29 cells) after surgical therapy. Moreover, the detection rate of CellCollector (82.4%, 14/17; CTC count range, 0-17) in advanced SCCHN (stage III-IV) was much higher than that in early stages (stage I-II, 40.0%, 2/5; CTC count range, 0-2) (P<0.05). EGFRVIII expression of CTC was also analyzed with fluorescence staining. One CTCEGFRVIII-positive patient was detected from six CTC-positive patients, and the positive expression of EGFRVIII was also found in the tumor tissue of this patient. CONCLUSIONS In vivo detection of CTCs had high sensitivity in SCCHN, which might improve CTC application in clinic.