A total of 22 patients, diagnosed with colorectal cancer, were enrolled in this study between October 2018 und August 2020. Before treatment, all patients passed a complete clinical evaluation including clinical history, physical examination, rectoscopy/colonoscopy, relevant blood examination and chest/abdominal computed tomography. Local stage was determined according to the TNM classification of the UICC (21). The recruitment criteria were as follows: Histologically confirmed, invasive colorectal carcinoma (ICD10: C18/C20); primary diagnosis. The characteristics of all patients enrolled in this study are shown in Table I. All patients were treated according to current treatment guidelines for colon (ICD10: C18) or rectal (ICD10: C20) cancer (22). Long term R/CT for rectal cancer was performed as follows: Radiation dose: 50.4 Gy (single dose 1.8 Gy); target volume: Rectal cancer and region of pelvic lymphatic drainage; chemotherapy: 5-Fluorouracil (10 patients), Capecitabine (1 patient), 5-Fluouracil/Oxaliplatin (3 patients). Individual therapy decisions were within the discretion of the attending physician and independent of any data collected in the course of this study. For patients receiving neoadjuvant therapy, 7.5 ml peripheral blood samples were obtained 1-7 days before initiation of R/CT, 17±3 days after the first cycle of R/CT, and after the completion of R/CT (1-7 days before surgery). For patients with only or additional adjuvant therapy, blood samples were obtained 1-7 days before surgery, 6-8 weeks after surgery (before initiation of adjuvant therapy), 17±3 days after the first cycle of adjuvant chemotherapy, and on the last day of therapy, respectively.

The study was based on the Ethics Declaration of Helsinki and was approved by the Ethics Committee of the University of Bayreuth. Participants provided their written informed consent to participate in this study.

For all rectal cancer patients, treatment responses were assessed according to the pathological results after surgery, and graded by histological evaluation of the surgical specimens according to the criteria described by Dworak et al (23). The grade of tumor regression was defined as follows: Grade 0: No regression; Grade 1: Dominant tumor mass with obvious fibrosis and/or vasculopathy; Grade 2: Dominantly fibrotic changes with few tumor cells or groups (easy to find); Grade 3: Very few tumor cells (difficult to find microscopically) in fibrotic tissue with or without mucous substance; Grade 4: No tumor cells, only fibrotic mass (total regression/response).

For a proper assessment of therapy response, we additionally compared the tumor size and lymph node status as assessed by computed tomography and/or endosonography of each patient before and after neoadjuvant R/CT. According to Dworak regression grade, as well as TNM re-staging, each patient was individually assigned either to the group of good or poor responders to neoadjuvant R/CT (Table II).

Peripheral blood (7.5 ml) from 22 patients with colorectal cancer at different stages of disease was drawn into blood count tubes containing ethylenediaminetetraacetic acid (EDTA) as an anticoagulant and processed 24 h after collection.

The Maintrac^® approach was used for identification, quantification and further characterization of CETCs (22). To this end, 1 ml of EDTA-blood was subjected to red blood cells lysis at 4˚C for 15 min using 14 ml erythrocyte lysis buffer (Qiagen GmbH). Remaining cells were spun down at 700 x g for 7 min at rt and resuspended in 500 µl of PBS/EDTA buffer. Immunostaining was performed by adding 4 µl fluorescein-isothiocyanate (FITC)-conjugated anti-human epithelial cell adhesion molecule (EpCAM) antibody (clone HEA-125; Miltenyi Biotec GmbH) to 25 µl of the cell suspension (about 10⁷ cells/100 µl) and incubation for 20 min at 4˚C in the dark. The corresponding isotype control for EpCAM (mouse IgG1 K FITC; Miltenyi Biotec) was used at the same final concentration. In case of co-staining of Ki-67, additional 2.5 µl of phycoerythrin (PE)-conjugated anti-Ki-67 antibody (clone B56; BD Biosciences) was added prior to incubation. Subsequently, all samples were diluted in PBS/EDTA buffer to a total volume of 250 µl. A defined volume of the cell suspension and propidium iodide (PI; Sigma-Aldrich; Merck KGaA) was transferred to the wells of ELISA-plates (Greiner Bio-one). Co-staining of cells with Ki-67 was performed without PI. Red and green fluorescence of the cells was examined using a Fluorescence Scanning Microscope ScanR (Olympus), enabling detection and relocation of cells for visual examination of EpCAM-, PI- or Ki-67-positive cells. For quantification of CETCs, only vital CETCs with intact cell morphology and without PI staining were counted. For daily verification of optical components and detectors of the microscope, fluorospheres (Flow-Check 770; Beckman Coulter) were used.

Only a small subpopulation of CETCs possessing additional stem cell properties is able to grow into metastases. By enumeration of CETCs able to clonally grow into CETC microspheres under specific conditions, we specified and quantified this subpopulation. Therefore, CETCs and leukocytes were isolated from peripheral blood as described earlier, plated at a density of 2x10⁵ cells/ml in RPMI-1640 supplemented with l-glutamine, HEPES, penicillin/streptomycin and growth factors such as EGF, insulin and hydrocortisone, and incubated under standard cell culture conditions (37˚C, 5% CO₂) in a sterile incubator. Every five days, the cultures were inspected under an inverted light microscope (PrimoVert) and fresh culture medium was added. Between days 21 and 28 of incubation, spheres were collected from the culture flasks, pelleted (250 x g, 7 min), and resuspended in 500 µl PBS. Immunostaining of spheres was performed using FITC-conjugated mouse anti-human EpCAM-antibody (clone HEA-125; Miltenyi Biotec GmbH), PE-conjugated mouse anti-human CD44-antibody (BD Biosciences) or mouse anti-human CD133-antibody (clone 7; BioLegend) for 20 min at 4˚C in the dark. The samples were then diluted in PBS/EDTA and transferred into the wells of a 96-well microtiter plate (Greiner Bio-one). Analysis of fluorescence was performed using a fluorescence scanning microscope (ScanR; Olympus). To verify vitality, PI staining of spheres was performed before the analysis. Finally, only vital CTC spheres with intact morphology and without PI staining were counted.

In case of surgery of the primary tumor, a small piece of tissue from the middle of the tumor (ø depending on the size of the tumor) was obtained in a sterile falcon in 10 ml transportation medium (RPMI-1640, 5% FBS, 5 µg/ml insulin, 2.75 µg/ml transferrin, 20 mM sodium selenite, 55 µg/ml sodium pyruvate, 1 µM hydrocortisone, 1,000 U/ml penicillin, 1,000 µg/ml streptomycin, 250 mg/ml amphotericin B, 15 mM HEPES, 100 µg/ml gentamycin, 5 µg/ml metronidazole) directly from the operating theater and kept at 4˚C for transportation. All samples were processed within 24 h after withdrawal. For further processing, the tumor tissue was washed 3-5 times in PBS by extensive shaking and put into a sterile petri dish. Before the tissue was chopped into small pieces of about 1 mm in diameter by anti-parallel movement of two scalpels, it was covered with a small amount of sphere culture medium (RPMI-1640 supplemented with l-glutamine, HEPES, penicillin/streptomycin and growth factors such as EGF, insulin and hydrocortisone). After one more washing step with PBS, the tissue was enzymatically homogenized with Accumax™ solution (Sigma-Aldrich; Merck KGaA) for 45 min under continuous mixing at rt. Then the cell suspension was filtered using a cell strainer (mesh size 0.44 µm; Greiner Bio-one) to eliminate bigger cell clumps and centrifuged at 240 x g for 10 min at rt. The resulting cell pellet was resuspended in 1 ml of culture medium and the number of vital cells was determined by bromophenol blue staining. Finally, the cells were plated in a concentration of approximately 0.6x10⁶ vital cells/ml in culture medium in 6-well plates and incubated at standard cell culture conditions (37˚C, 5% CO₂) for several weeks. All cultures were checked for bacterial infections daily and in the case of a minor infection isolated and treated with additional antibiotics, or in case of a major infection, discarded. If primary tumor spheres were detectable after a few weeks, a small amount of the culture was harvested and immunostained for further characterization and documentation.

Statistical analysis was performed using SigmaPlot (version 14.0; Systat Software Inc.) for Windows. Comparisons between variables were performed using ANOVA (analysis of variance) followed by a post hoc test for parametric data, or Kruskal-Wallis test followed by Dunn's test for nonparametric data. The significance level was set at P<0.05.

A total of 22 patients with histologically confirmed colorectal cancer (16 patients with rectal cancer, 6 patients with colon cancer) were enrolled in this study. Patients' characteristics are given in Table I.

Considering all patients (ICD10: C18 and C20), 1 patient was at stage I (4%), 7 patients were at stage II (32%), 12 patients were at stage III (55%), and 2 patients were at stage IV (9%). Six patients (27%) suffered from colon cancer and 16 (73%) from rectal cancer. The age of the patients ranged from 51 to 80 years (median 65.5 years). The median number of CETCs of all 22 patients with colorectal cancer was 55 CETCs per 100 µl cell suspension (ranging from 0 to 640) from which colon cancer patients (ICD10: C18) had a median CETC number per 100 µl of 45 (ranging from 0 to 145), and rectal cancer patients (ICD10: C20) of 65 (range from 0 to 640). No statistically significant differences in CETC numbers were observed in correlation to tumor size, lymph node status or distant metastasis (data not shown).

Using the Maintrac^® method we detected CETCs in 100% of colorectal cancer patients included in this study. CETC numbers of all patients during the course of therapy are specified in Table SI. In addition to epithelial characteristics as assigned by immunostaining, we could demonstrate proliferative and stemness properties of CETCs under specific conditions, which were identical to those of cells derived from the primary tumor itself (Fig. 1).

CETCs from patient #1 were investigated for their proliferative activity by growing non-adhesive suspension cultures. Formation of EpCAM-positive spheres was observed after the first cycle of neoadjuvant R/CT (5 spheres/100 µl blood). Interestingly, CETCs from all other samples did not show any sphere formation. During surgery of patient 1, a small piece of tumor tissue was set aside and stored on ice until further processing. After separation and washing, primary tumor cells were cultured under the same conditions as CETCs, also resulting in the formation of spherical structures. Both, spheres from the primary tumor, as well as spheres from the peripheral blood of patient #1 were further characterized by immunostaining (Fig. 1). The viability of the spheres was ensured by counterstaining with PI (propidium iodide), which cannot permeate live cells. Expression patterns in primary tumor spheres of specific stem cell markers, which are regularly over-expressed in colorectal tumors (CD44 and CD133), correlated with those in spheres from peripheral blood. Moreover, primary tumor spheres expressed high levels of PD-L1.

14 patients with rectal cancer received neoadjuvant R/CT, 7 (50%) of whom showed a good response (Fig. 2) and 7 (50%) did not or only partially respond to the therapy (Fig. 3). In the group of good responders, the mean CETC number before R/CT was 105 CETCs/100 µl of cell suspension. After the first cycle of the R/CT it decreased to 47 per 100 µl. With a P-value of 0.543, the differences between the three time points did not reach statistical significance likely due to the small sample size. Nevertheless, the results show a trend. In detail, the CETC numbers declined in 5 of 7 patients (71%) and increased in only 2 patients (29%) with good response to neoadjuvant R/CT (Fig. 2). In the group of poor responders (7 patients), the mean CETC number was initially 40 CETCs/100 µl cell suspension, and increased continuously from 73 after the first cycle of the R/CT to 210 CETCs/100 µl cell suspension before surgery (Fig. 3). Again, the small number of participants (n=7) might be the major cause for the lack of statistical significance (P=0.428).

9 patients (41%; 6 patients with colon cancer and 3 patients with rectal cancer) received adjuvant chemotherapy and CETCs were quantified before surgery, before the beginning of chemotherapy and after the first cycle of adjuvant therapy (Fig. 4). Before surgery the median CETC number was 55/100 µl cell suspension, 6-8 weeks after surgery (before the beginning of the adjuvant CT) the median value was 65, and after the first cycle of CT the median was 20 CETCs/100 µl cell suspension. The difference between the mean values of the three time points was not statistically significant (P=0.114).

Interestingly, all patients showed decreasing CETC numbers under adjuvant chemotherapy.

Ki-67-positive CETCs were detected in 20 patients (91%) and the percentage ranged from 0-100 (median: 25 Ki-67-positive CETCs/100 µl cell suspension). The median of Ki-67-positive CETCs/100 µl cell suspension in colon cancer patients was 18 (ranging from 0 to 170), and in rectal cancer patients 25 (ranging from 0 to 169). Although the differences in the Ki-67-positive CETCs at the three time points were not statistically significant neither for patients with neoadjuvant R/CT (P=0.202), nor in the group of patients with adjuvant CT (P=0.151), there was a trend in the number of Ki-67-positive CETCs to decrease under adjuvant CT, and to increase in patients receiving neoadjuvant R/CT (Fig. 5).

Fig. 6 shows an example of a serial analysis of the CETC numbers during the therapy of a 63-year-old patient with stage III (T3, N1, M0; G2) rectal cancer. The patient was treated with neoadjuvant R/CT (Dworak 1, poor response), followed by surgery (R0-resection) and additional adjuvant chemotherapy. During neoadjuvant therapy the CETC numbers increased significantly and reached their maximum (225 CETCs/100 µl cell suspension) before surgical removal of the primary tumor. Eight weeks after surgery the CETC number had fallen to a level similar to that at the beginning of the R/CT. It continued to decrease until there were no residual CETCs detectable at the last day of the adjuvant CT. Until 9 months after completion of the adjuvant therapy, this patient has remained free of relapse.