Isolation and Enumeration of CTC in Colorectal Cancer Patients: Introduction of a Novel Cell Imaging Approach and Comparison to Cellular and Molecular Detection Techniques

Simple Summary Despite significant progress in screening and treatment regimens, colorectal cancer (CRC) still is a major health burden lacking profound liquid biomarkers for identifying patients at risk. Circulating tumour cells (CTC) have the potential to non-invasively improve the diagnosis. We have already established a sensitive semi-quantitative RT-qPCR against CK20 for CTC quantification in CRC patients. For clinical translation, this study aims to validate our molecular detection method by terms of cytological approaches, and implement a novel semi-automated microscopic detection after immunofluorescence labelling of CTC. Additionally, we aim to compare our PCR-based approach to a marker-independent, but size-dependent, enrichment process. We have successfully applied the validation techniques and proved their feasibility. Enumeration by size yielded the highest numbers of CTC and demonstrated to be the most reliable strategy for CTC detection in CRC patients. Future studies with larger patient cohorts will have to investigate the clinical significance and prognostic value of this approach. Abstract Circulating tumour cells (CTC) were proven to be prognostically relevant in cancer treatment, e.g., in colorectal cancer (CRC). This study validates a molecular detection technique through using a novel cell imaging approach for CTC detection and enumeration, in comparison to a size-based cellular and correlated the data to clinico-pathological characteristics. Overall, 57 CRC patients were recruited for this prospective study. Blood samples were analysed for CTCs by three methods: (1) Epithelial marker immunofluorescence staining combined with automated microscopy using the NYONE® cell imager; (2) isolation by size using membrane filtration with the ScreenCell® Cyto IS device and immunofluorescence staining; (3) detection by semi-quantitative Cytokeratin-20 RT-qPCR. Enumeration data were compared and correlated with clinic-pathological parameters. CTC were detected by either approach; however, with varying positivity rates: NYONE® 36.4%, ScreenCell® 100%, and PCR 80.5%. All methods revealed a positive correlation of CTC presence and higher tumour burden, which was most striking using the ScreenCell® device. Generally, no intercorrelation of CTC presence emerged amongst the applied techniques. Overall, enumeration of CTC after isolation by size demonstrated to be the most reliable strategy for the detection of CTC in CRC patients. Ongoing studies will have to unravel the prognostic value of this finding, and validate this approach in a larger cohort.

CA, USA) diluted 1:4 in PBS/1% FCS. After cell counting using NYONE ® , an average of 100 HT29 cells (determined by repeated counting) were then spiked into 8.2 ml of healthy donors´ blood samples and then transferred to BD Vacutainer® Mononuclear Cell Preparation Tubes (CPT) (Becton Dickinson, Heidelberg, Germany). For isolation of the mononuclear cell (MNC) fraction, the CPT tubes were centrifuged at 1650xg for 20 minutes. After transferal of the MNC to 50 ml centrifuge tubes, a fixation buffer (#14190-094, Biolegend, San Diego, CA, USA) was added and incubated for 15 minutes at room temperature after which the samples were stored at 4°C for up to four days until further analysis was possible.
For final quantification of CTC, at first, the cells were permeabilised with a Perm-/Wash-Buffer (#421002, Biolegend) for 5 minutes at room temperature and then centrifuged at 330xg for 10 minutes.
Patients´ blood samples were handled equally and as described above.

Immunofluorescence assays used to immunostain CTC on the porous membrane (ScreenCell ® ):
The colon cancer cell line HT29 was used to immunostain a panel of cytokeratins (pan-CK, Clone: AE1/AE3, #M3515; Agilent, Santa Clara, CA, USA). The HT29 cells were enriched and fixed with 2% formalin on a porous membrane named Cyto isolation support or Cyto IS using ScreenCell ® Cyto kit (ScreenCell, Sarcelles, France) from 3 ml of blood of healthy donors (Figure 1B and 2B). The Cyto IS device was dried during one hour at 37°C and stored at 4°C. Prior to the immunofluorescence staining, the Cyto IS device with HT29 cells was hydrated with Tris-Buffered Saline (TBS) (#S30001; Agilent).
The antigens were retrieved with target retrieval solution pH 6 (#S1700; Agilent) at 95-99˚C for 20 min, cooled during 20 minutes at room temperature and washed with TBS. The isolated cells were treated with a blocking solution containing 3% BSA in TBS during 30 minutes at room temperature. Then, the ScreenCell ® Cyto IS device was incubated overnight at 4°C with anti-pan-CK mouse monoclonal antibody (#M3515; Agilent) and anti-CD45 (Clone EP68) rabbit monoclonal antibody (#AC-0065A; Epitomics, Abcam, Cambridge, GB). The ScreenCell ® Cyto IS device was washed twice with TBST (#S3306, Agilent) and incubated for one hour at room temperature with secondary antibodies goat antimouse IgG Alexa Fluor488 (#A11001; Life Technologies, Carlsbad, CA, USA) and goat anti-rabbit IgG Alexa Fluor568 (Cat# A11011; Life Technologies). The ScreenCell ® Cyto IS was then washed twice and incubated with DAPI (#D1306, Life Technologies) during 10 minutes to counterstain the nucleus. After a final wash with TBST the ScreenCell ® Cyto IS device was mounted with the Vectashield ® (#H-1000; Vector Laboratories, Burlingame, CA, USA). The microscopic analysis was performed with Epifluorescent microscope (Nikon Eclipse E80i, Tokyo, Japan) equipped with NIS-Elements Viewer software (Nikon). Mx obecive (obj.) 40.
Blood samples from CRC patients were processed using ScreenCell® Cyto kit and were firstly stained with May Grunwald Giemsa (MGG) (RAL Diagnostics, Martillac, France). The CTC were analysed after microscopic scanning and counted.
The cells were than incubated in TBST to remove MGG and cells were immunostained according to the immunofluorescence protocol applied on spiked HT29 cells. The isolated atypical cells that were pan-CK-negative/CD45-negative, pan-CK-positive/CD45-negative or pan-CK-positive/CD45-positive, were