Abstract
Background/Aim: This study describes a rare cell sorter (RCS) method to detect circulating tumor cells (CTCs) and CTC clusters in whole blood without pretreatment. Patients and Methods: We collected samples from breast cancer patients at the University of Tsukuba Hospital. A total of 15 whole-blood specimens from patients with breast cancer were collected and analyzed via a microfluidics chip, fluorescence-conjugated antibody staining, and fluorescence microscopy. Of 15 total cases, eight were analyzed by RCS ver3 and seven were analyzed by RCS ver3.5 to reveal potential clinical differences in scanning methods. We then examined the HER2 status on 4 of the 15 patients using our RCS system. Results: RCS efficiently detected all subtypes of CTCs and CTC clusters from the peripheral blood of cancer patients. The concordance rate of HER2 status between tissue and CTCs in 4 tested clinical samples was 100%. Conclusion: RCS is a non-invasive method that allows for simultaneous detection of CTCs, cluster presence, and surface marker (e.g., HER2) status. Frequent sampling is, thus, possible and the large amount of data obtained will be clinically useful to predict response to therapy as well as plan adjunct support therapies in cancer patients.
Circulating tumor cells (CTCs) are malignant cells originating from a primary tumor that circulate in the bloodstream and metastasize to a secondary site (1, 2). The number of CTCs is thought to be associated with prognosis and may be involved tumor metastasis (3). Within the circulation, CTCs take on diverse forms (including singlets and cell clusters) and recent reports indicate that cluster-type CTCs are more likely than single CTCs to survive, proliferate, and act as a generative source of metastatic diseases (4-6).
The limitations of conventional technologies prevent a fully accurate analysis of highly heterogeneous populations of CTCs and CTC clusters since, overall, CTCs are comparatively rare and represent a very small minority of cells detected in peripheral circulation. CTC clusters are even rarer than single CTCs; as such, the detection and isolation of a statistically significant number of either single cells or clusters is technically challenging.
Recent studies have revealed molecular alterations in cancer cells that represent evolution during tumor progression and treatment response. Although these molecular analyses are typically performed on biopsy tissue acquired at diagnosis, the invasiveness of tissue biopsies (surgical or needle) precludes frequent sampling for disease tracking. Such limits in progression profiling are reflected in cases where metastatic tumors have different molecular alterations from the primary tumor. As an example, discrepancies in HER2 status between the primary tumor and distant metastases have been observed in 7–26% of patients with metastatic breast cancer (7-9). Instead of tissue biopsies, isolation of CTCs from peripheral blood has emerged as a reliable source of tumor cells as well as a suitable prognostic and/or predictive biomarker (10).
We hypothesized that less-invasive peripheral collection and specific antibody staining, dubbed rare cell sorting (RCS), can rapidly distinguish CTCs (11) and isolate CTC clusters (including different phenotypes and cluster mixes) to better predict prognosis, guide and monitor treatment, and design antitumor strategies. We, thus, utilized this combination approach in clinical samples of breast cancer patients to characterize clusters by HER2 status.
Patients and Methods
Cell lines and culture. Human colorectal adenocarcinoma HT-29 cells were purchased from the ATCC. HT-29 cells were maintained at 37°C under 5% CO2 with humidification in Dulbecco’s modified Eagle medium (DMEM) (044-29765, FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) supplemented with 10% fetal bovine serum (FBS, 35-079-CV, Corning, NY, USA), and 50 U/ml penicillin and 50 μg/ml streptomycin (FUJIFILM Wako Pure Chemical Corporation).
Preparation of spiked samples. Whole-blood samples (10ml) were collected from healthy volunteers at the University of Tsukuba Hospital. This study was approved by the Institutional Ethics Committee of the University of Tsukuba Hospital (approval number H30-150). All healthy volunteers provided written, informed consent. Cancer cells (300 cells) and whole-blood sample (5 ml) were mixed.
Patient sample collection and study approval. We prospectively collected samples from breast cancer patients at the University of Tsukuba Hospital. A total of 15 whole-blood specimens from patients with breast cancer were collected. Study protocols were approved by the Institutional Review Board at the University of Tsukuba Hospital. All patients provided written, informed consent (approval number H30-150).
Isolation of single and clustered CTCs for Cases 1 to 8. A total of 5 ml of whole blood was collected from patients with breast cancer in an ethylenediamine tetraacetic acid (EDTA) collection tube (Venoject II vacuum blood collection tube, Terumo, Tokyo, Japan). Sample processing and evaluation were done as described by Masuda et al. (10). Briefly, the blood sample was diluted with the same volume of PBS (EDTA-2Na solution). The blood samples were then aspirated through open-channel microfluidic chips using a syringe pump that relies on capillary forces (flow rate=10-20 ml/h). CTCs within the blood sample, including CTC clusters, were then trapped in pockets on the microfluidic chip due to capillary force associated with the meniscus of the air-liquid interface (Figure 1). Due to their relatively small size, blood cells passed through the gaps between micropillars of 7 μm and were flushed away, isolating the CTCs on the microfluidic chip. Next, the trapped cells were washed once with 600 μl PBS and fixed with 4% paraformaldehyde (PFA) in PBS for 10 min at room temperature. After washing 3 times with PBS, trapped cells were permeabilized with 0.1% Triton-X100 in PBS for 15 min. After washing 3 times with PBS, trapped cells were stained with primary anti-EpCAM antibody (clone VU1D9, GTX42071, GeneTex, Irvine, CA, USA, 1:200) and anti-human Cytokeratin antibody (clone CK3-6H5, 130-090-866, MiltenyiBio, Bergisch Gladbach, Germany, 1:100) for 30 min at room temperature. We then stained with goat, anti-mouse phycoerythrin (PE) (31862, Invitrogen, MA, USA, 1:500) and Hoechst 33342 (H3570, Invitrogen, MA, USA, 1:1,000) secondary antibodies, in addition to a fluorescein isothiocyanate (FITC) -conjugated antibody against CD45 (clone J33, A07782, Beckman Coulter, CA, USA, 1:50), for 30 min at room temperature. After PBS washing, the microfluidic chip was scanned for trapped cells with an optical microscope (ECLIPSE Ti2, NIKON, Tokyo, Japan).
View of microfluidic chip and cross-section. The microfluidic chip has 3 tubes: wash/stain, blood sample, and suction.
Isolation of single and clustered CTCs for Case 9 to Case 15. A total of 5 ml of whole blood was collected in an EDTA collection tube (Venoject II vacuum blood collection tube). Sample processing and evaluation were done as described by Masuda et al. Briefly, the blood sample was diluted with the same volume of PBS (EDTA-2Na solution) containing a PE-conjugated antibody against EpCAM (IgG2b; HEA-125, 130-113-264, MiltenyiBio, 1:200) and a FITC-conjugated antibody against CD45 (1:50), plus an Alexa Fluor 647-conjugated antibody against HER2 (clone 24D2, 324412, BioLegend, CA, USA, 1:50) to avoid clogging. The blood sample was separated as described above. The trapped cells were washed once with 600 μl PBS and trapped cells were directly stained on the microfluidic chip with Hoechst 33342 (H3570, Invitrogen, 1:100) at room temperature for 15 min. After PBS washing, rare cell sorting microscopy was used to detect trapped cells on the microfluidic chip.
Evaluation of CTC identification. CTCs trapped on the open-channel microfluidic chip were visually detected with a fluorescence microscope; images of the entire filter area were obtained using a 10x objective lens (excitation wavelengths of 355, 488, 565, and 650 nm and emission wavelengths of 460, 525, 578, and 660 nm) before automatic stitching into a single, generated image file. A CTC was defined as a single, intact round oval cell with a visible nucleus (Hoechst 33342 positive) that stained positively with anti-EpCAM but negatively with anti-CD45. Clusters were defined as aggregates of two or more CTCs. Quantification of single and clustered CTCs are presented as the number of cells and clusters in 5 ml of peripheral blood. For evaluation of the HER2 levels, surface HER2 in the CTC subset was assessed using HER2-Alexa Fluor 647.
To confirm whether the RCS could detect cancer cells after EpCAM-positive staining, the identification rate of cancer cells on the chip by fluorescent labeling was compared and evaluated with an automatic fluorescence microscope attached to this device and an inverted microscope (ECLIPSE Ti2) for Cases 1-8. For Cases 9-15, cells on the chip were automatically detected with an automatic fluorescence microscope and computer system.
Results
Evaluation of CTC identification. First, we used the antibody-based method to distinguish cancer cells and leukocytes via RCS. We first equally mixed human colorectal cancer HT-29 cells with whole blood from healthy volunteers (spiked samples) and, as shown in Figure 2, we successfully detected EpCAM-positive cells. To confirm that RCS can detect EpCAM-positive cancer cells, we compared RCS microscopy (Figure 2A) to optical microscopy (Figure 2B), resulting in a concordance rate of about 95%.
Detection of EpCAM-positive cells. EpCAM-positive HT-29 cells as detected by rare cell sorter (RCS) (A) and optical microscopy (B).
Patient and tumor characteristics. Clinicopathologic characteristics of all patients are summarized in Table I. The median age was 58 years (range=38-83 years) and the patient cohort comprised of 15 women. The included histologic types were invasive ductal carcinoma (14 cases) and ductal carcinoma in situ (1 case). The histological HER2 statuses were as follows: HER2 positive (6 cases), HER2 negative (8 cases), and unknown (1 case). At the time of blood collection, 6 patients had metastatic disease, 11 patients were undergoing chemotherapy, and 6 patients were undergoing radiotherapy (Table I).
Patient characteristics.
CTC and clustered CTC enumeration. We successfully detected EpCAM-PE/Cytokeratin-PE-positive cells (Figure 3A) and captured images of EpCAM-PE/Cytokeratin-PE-positive cell clusters (Figure 3B). The number of CTCs and CTC clusters are shown in Table II.
Detection of circulating tumor cells (CTCs). EpCAM-positive/Cytokeratin-positive single cell (A) and cluster cells (B) were detected. EpCAM-positive cluster (C) and EpCAM-PE-positive/HER2-Alexa Fluor647-positive cells (D) were detected.
Number of circulating tumor cells (CTCs)/CTC clusters and HER2 statuses.
HER2 status evaluation in CTCs. We detected EpCAM-positive CTCs (Figure 3C) and HER2-positive/EpCAM-positive CTCs (Figure 3D). In HER2-positive tissue (Cases 11 and 12), HER2-positive CTCs were detected and, as expected, HER2-negative tissue (Cases 14 and 15) had no observable HER2-positive CTCs. In Case 11, we detected 5 CTCs, but only 1 CTC was HER2-positive, while, in Case 12, we detected 48 CTCs with 4 CTCs labeled as HER2-positive. We took data from 4 patients (Cases 11, 12, 14 and 15) with detectable CTCs and evaluated their HER2 statuses, comparing the results with HER2 evaluations of primary tumors. The concordance rate of HER2 statuses between tissue and CTCs was 100% (Table II).
Discussion
In this study, we constructed a rare cell sorting system, featuring microfluidic chips, to identify and isolate both single and clustered CTCs simultaneously from peripheral blood samples of breast cancer patients. We additionally tested 2 staining methods to elucidate any variability between methods in clinical practice.
Some studies have reported that molecular alteration profiles in cancer cells change during tumor progression and/or treatment with molecular targeted drugs (13-16). Currently, HER2 status is determined at initial diagnosis by analyzing primary tumor tissue. However, as several studies have demonstrated that HER2 status may change during disease progression (7, 8), non-invasive reassessment of CTCs for HER2 in relapsed patients is a strategy with potent clinical applications. Therefore, we performed a study to non-invasively evaluate HER2 status in CTCs of breast cancer patients and compare data with primary tumor tissue samples, speculating that these circulating cells may represent evolution of metastatic tumors into distinct phenotypes. Such data would be useful during a clinical course to plan alternative or adjunct therapies as well as predicting response to chemo- or radiotherapy.
Prediction of therapy response is mostly measured using tumor biopsies; however, the sampled tissue may not be representative of the entire tumor (or metastases) because of tumor heterogeneity and/or clonal evolution during disease progression. As an example, the expression of HER2 has been previously reported as an important biomarker for breast cancer response and, thus, frequent assessment of HER2 on tumor cells would be beneficial for predicting response (17, 18). To reduce the invasiveness of such sampling, we developed an imaging-based analytical method (RCS) that enables CTC-identification simultaneously with evaluation of HER2 heterogeneity. In this study, all tissues after surgery were checked for HER2 status since the correlation between CTC and tissue HER2 statuses is clinically meaningful. Only part of the captured CTCs was HER2-positive among all detected CTCs. One possible reason is subtype heterogeneity within primary tissue or that cancer cells might acquire phenotypical changes during progressive disease. This indicates that some subpopulations of CTCs may be accurate for use as biomarkers due to reflecting the real-time morphology of such metastatic cancer cells.
In conclusion, we developed a non-invasive, antibody-based imaging modality that effectively isolates and analyzes CTCs from clinical cancer patients to characterize these peripheral cells for therapy response prediction. Future validation studies will serve to both fine tune and accumulate a body of phenotypic data to increase the success of therapy prognoses based on molecular attributes of metastatic cells.
Acknowledgements
The project described was supported in part by award number 19he2202003h0201 from the Japan Agency for Medical Research and Development. The authors would like to thank Dr. Bryan J. Mathis of the University Hospital of Tsukuba International Medical Center for language revision.
Footnotes
Authors’ Contributions
T.B., T. Masuda and F.A. supplied RCS. A.U. and H.B. supplied clinical samples. T. Mori and A. F. performed the experiments. S.M. supervised the entire project. All Authors read and agreed to publication of the article.
Conflicts of Interest
The Authors do not have any potential conflicts of interest to declare.
- Received July 19, 2022.
- Revision received August 10, 2022.
- Accepted August 24, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.