INTRODUCTION
Colorectal cancer is one of the most common cancers worldwide, and is well known to be the main cause of the cancer death [
1,
2]. Early diagnosis and development of its treatment options have improved the treatment outcome continuously; however, it still ranks as the 4th and the 3rd highest mortality rates among men and women, respectively [
2]. Such high mortality rates are due to recurrence and metastasis of the cancer, which are shown in 50% of the patients [
3]. Thus, providing a proper plan for patients with colorectal cancer is necessary in order to improve the mortality rate by early detection of the metastasis of cancer.
The concept of circulating tumor cells was derived from such a background. Circulating tumor cells are tumor cells that circulate in the peripheral blood of the patients, and they are originally derived from a primary cancer. The presence of circulating tumor cells is one of the steps that are necessary to provoke distant metastasis. This process includes detachment of cancer cells from the primary lesion and their implantation at target organs through peripheral circulation, thus establishing the distant metastasis [
2]. Recently, many methods had been tried in order to extract circulating tumor cells effectively, and carcinoembryonic antigen (CEA) has been used as the typical marker for circulating tumor cells [
2,
4,
5]. However, practical usage of circulating tumor cells in clinical situations is still limited due to the facts that they are extracted in small amounts and that the method of extraction is yet to be standardized. Consequently, further development of a new marker that not only reflects the expression of circulating tumor cells but also is involved in the metastatic process of colon cancers is necessary in the future.
Cellular adhesion molecules are major molecules that are involved in the metastatic process of circulating tumor cells. In order for the circulating tumor cells to cause metastasis, they need to attach to the vascular wall and penetrate into surrounding tissues while they are circulating in the peripheral blood. During this process, an interaction between the tumor cells and the surrounding extracellular matrix near the target organ is required. At this very moment, many types of cellular adhesion molecules are expressed abnormally, especially, integrin. Integrin is a major cellular adhesion molecule that plays an important role in the migration and the invasion of the tumor cells [
6,
7]. Among these molecules, β
1-integrin is known to be overexpressed in many cancers, especially in metastatic colorectal cancer [
6].
Many articles have reported on circulating tumor cell or β1-integrin in colorectal cancer. Unfortunately, such previous investigations were mainly reporting the functions of each molecule. So far, very few articles have attempted to reveal the interactions between the molecules. Cancer metastasis is a process of several mechanisms, in which we assume that a relation may exist between the number of circulating tumor cells and the expression of the β1-integrin. In this study, we evaluated the expression level of β1-integrin in relation to different clinical characteristics of patients with colorectal cancer. The expression level of β1-integrin was compared to that of CEA, which is a marker for circulating tumor cells, in order to analyze their relationship. In this way, we aimed to investigate the function of β1-integrin, which might be utilized as a new marker to predict the presence of micrometastasis of the cancer and the prognosis for the patient.
METHODS
This study consisted of 2 groups of 30 patients each, for a total of 60 patients. The experimental group included patients who were diagnosed with colorectal cancer at the Surgery Department of Ewha Woman's University Mokdong Hospital and were to go surgery. The patients of the experimental group were diagnosed with an adenocarcinoma of the colorectum through tissue biopsy, and they had no history of cancer of other organs, metastasis that is accompanied by cancer of other organs, surgery, chemotherapy, radiotherapy, or any other procedures that were part of the treatment for cancer. The control group included patients suffering from benign diseases who had no history of cancer. Written informed consent was obtained for all the enrolled patients. The protocol of this study was confirmed by clinical audit committee of our institution (confirmation number: ECT12-10B-25).
Peripheral venous blood samples were taken from the experimental group during preoperative state under anesthesia and on the 7th postoperative day. In the control group, the blood samples were only taken before the surgery while the patients were under sedation. When the peripheral blood was sampled, an angio-needle was inserted, and first 5 mL of the sample was discarded so that contamination of the epithelial cells might be prevented. An additional 10 mL of venous blood was sampled and stored in an ethylenediaminetetraacetic acid bottle.
At room temperature, the blood was centrifuged for 10 minutes at 2,500 rpm, and the serum was removed. The buffy coat layer was separated, and RNA was extracted according to a conventional method [
8]. One microgram of the separated RNA was stored at 42℃ for 15 minutes, at 95℃ for 5 minutes, and at 4℃ for 5 minutes by using a reverse transcription system (Promega Co., Madison, WI, USA) consisting of a reverse transcription 1 × buffer, 5-mM MgCl
2, 10-mM deoxynucleotide (dNTP), 40 unit/µL recombinant RNasin ribonuclease inhibitor, 25 unit/µL avian myeloblastosis virus reverse transcriptase, and 0.5 µg of oligo dT 15 primer in order to reverse transcribe into cDNA. Primers of CEA, which is a marker for circulating tumor cells, β
1-integrin, which is a cell adhesion molecule, and β-actin, which is an internal control, were produced, and then polymerase chain reaction (PCR) was performed (
Table 1). GoTaq DNA polymerase, 5 × green GoTaq reaction buffer, dNTP mix (Promega Co.), each primer mixture, and distilled water were mixed together. Four microliter of CEA cDNA, 2.7 µL of β
1-integrin cDNA, and 1 µL of β-actin cDNA were mixed together to produce the PCR mixture. The amplification of the primer was processed through denaturation of CEA at 95℃ for 5 minutes, and then at different temperatures of 95℃, 55℃, and 72℃ for 30 seconds each for 35 cycles. Lastly, the mixture was stored for 5 minutes at 72℃. In the case of β
1-integrin, each mixture that contained primer pairs was denatured at 95℃ for 5 minutes and at 95℃, 57℃, and 72℃ for 1 minute each for 35 cycles. The last process included storage at 72℃ for 5 minutes, after which the amplifying process was completed.
From the electrophoretic reverse transcription-polymerase chain reaction (RT-PCR) results that were obtained from each sample, we could assess the expression of CEA and β1-integrin. In order to semiquantify their expression levels, we measured the density of the observed bands of electrophoresis with the help of UN-SCAN-IT gel ver. 6.1 (Silk Scientific Inc., Orem, UT, USA). The results of the procedure were comparatively analyzed using the histological characteristics of the patients, and the expression levels of CEA and β1-integrin, along with their relationship, were analyzed at the same time.
The analysis of the clinical characteristics of the patients was done by using descriptive statistics, and we compared the positivities of CEA and β1-integrin by using the Mann-Whitney the analysis of variance, and the chi-square tests. In the experimental group, the comparison of each marker before and after surgery was completed with the paired t-test and the chi-square test. The comparison of the experimental and the control groups was done by using the Student t-test and the chi-square test. Also, the analysis of the correlation between the positivity of CEA and that of β1-integrin was done by using Pearson correlation coefficient and the chi-square test in order to identify the statistical meaning of the results. The program used for this analysis was SPSS ver. 16.0 (SPSS Inc., Chicago, IL, USA), and we considered the results to be significant when the P-value was less than 0.05.
DISCUSSION
Various diagnostic methods and treatment options are being developed in order to improve the survival rate of patients with colorectal cancer. Moreover, many markers are used to evaluate the progression of the cancer in the patients and to predict the prognosis. So far, many investigations had been done, but only the extraction of CEA within the blood has been found to be useful clinically [
9]. Accordingly, new markers that will allow us to predict the spread and the recurrence of the cancer and to evaluate the reaction of the patient to the treatment need to be developed in order to improve the survival rate of the patients.
Since the theory of circulating tumor cells was introduced by Ashworth [
10] in 1869, its clinical importance has been emphasized and has been the subject of many investigations. Circulating tumor cells are produced at the beginning of the metastatic process, and they are rarely observed in a healthy individual [
4]. They are known to be able to not only metastasize to distant organs, but also able to return to the primary organ and self-seed the cancer [
2]. The circulating tumor cells play an important role in diagnosing the cancer and predicting recurrence and metastasis, so they may be used to evaluate the reaction of the patient to the treatment. The extraction of circulating tumor cells may be a new way to predict the prognosis of the patient [
2]. According to many articles written about circulating tumor cells, they were expressed at higher levels in patients with colorectal cancer than in normal control patients [
8,
11], and the expression level tended to be higher for higher histological stage of the cancer [
8,
12]. Moreover, metastasis and recurrence were shown to occur at an earlier time in a patient who showed expression of circulating tumor cells [
3,
8], and their survival rate was reported to be lower as well [
3,
13]. According to the study of Wong et al. [
14], among the patients suffering from colorectal cancer with the presence of circulating tumor cells, 88% showed decreased levels of such cells after surgery. Similar results were observed in this study. The marker that was used to detect the circulating tumor cells was expressed especially at higher levels in patients with colorectal cancer, and the expression level was shown to be significantly decreased after surgery.
The molecular extraction of circulating tumor cells was tried for the first time by Smirnov et al. [
15] in 2005 in patients with colorectal cancer, and similar investigations have been reported since then. The markers that are most commonly used to detect the circulating tumor cells in colorectal cancer are CEA, cytokeratin 19, cytokeratin 20, epithelial cell adhesion molecule (EpCAM), human telomerase reverse transcriptase, etc., and among these, CEA is the most standardized marker [
2,
4]. Many methods for extracting circulating tumor cells have been tried; unfortunately, many of them have not yet been standardized. Thus, the usefulness of the various methods is a subject of debate [
5]. Among the enlisted methods, Cell Search System [
5] utilizes a semi-automatic system that dyes EpCAM immunologically, and its use is increasing these days. Its pitfall is low sensitivity, and the system cannot be used in cells that show EpCAM negativity. What's most important is that it costs approximately 800 US dollars per each sample to be tested and is difficult to use on a clinical basis [
4,
16]. Therefore, we selected RT-PCR [
2,
5,
8], which is widely used to extract circulating tumor cells, as our extraction method due to its easy operation and high sensitivity compared to the other methods.
Every day, approximately 10
6 cells are produced and excreted from 1 g of primary cancer into the circulatory system. Through the process called anoikis, a process of cellular degeneration that is caused by the loss of cell-matrix interaction, circulating tumor cells survive within the blood [
2,
17]. Unfortunately, the number of such circulating cells that are measured in reality is quite low (1 in 10
5-10
7), and this phenomenon is due to the dissolution of many tumor cells by the shear force [
2]. Eventually less than 0.1% of the cells will survive 24 hours after generation, and the percentage of circulating tumor cells that actually contribute to metastasis is reported to be less than 0.01% [
17]. Many efforts have been made to extract circulating tumor cells; however as described above, limitations on extracting the cells effectively still exist due to the very small number of cells and the unsettled extraction method. Accordingly, developing a new marker for patients that may show higher extraction level of circulating tumor cells compared to CEA, an ability to reflect the presence of circulating tumor cells and their involvement in the metastatic process of colorectal cancer, would be clinically meaningful.
The circulating tumor cells were reported to be related to factors such as tissue inhibitor of metalloproteinase 1, carbonic anhydrase IX, and epidermal growth factor receptor [
18,
19]; however, we assumed that research on a substance that is involved in the cell-matrix interaction, which may influence the metastasis more directly, seemed important. One such important substance is integrin. Integrin is a heterodimeric cell surface receptor that mediates the interaction between the cell and the extracellular matrix, and it is included in the glycoprotein family. It consists of 18 α-subunits and 8 β-subunits, and it forms 24 heterodimers through combinations of different bonds [
6]. Very little integrin is observed in most normal tissues, and its expression level is increased abnormally in tumor cells [
6]. Integrin enables tumor cells to move to target tissue by acting as a driving force with an adhesion function, and it accelerates the degradation of the basement membrane, easing the invasion by tumor cells. It accelerates the process of mobility and invasion of the cell through modification of many signaling pathways, such as the focal adhesion kinase and the src family kinase pathways [
6,
7,
20]. Integrin is also known for its ability to maintain the shape of the cell and induce cellular survival and proliferation, and it is involved in gene transcription, as well [
20]. Recently, thanks to such various functions of integrin, many researchers have been turning their attention to the development of new treatments that involve integrin as a target. Their argument includes the power of integrin to prevent the growth and metastasis of the cancer by injecting the antagonists of integrin, thus inhibiting the movement of and the invasion by tumor cells. Actually, many authors have suggested that growth of the cancer can be inhibited by blocking integrin [
7,
21]. In an investigation that utilized human breast cancer cell lines, monoclonal antibody against α5β
1-integrin was observed to inhibit the growth of tumor cells [
22]. In a similar mouse model in which liver metastasis of colorectal cancer was induced, the sizes of the primary cancer and the liver metastatic lesion were greatly decreased with the injection of both monoclonal antibody to α5β
1-integrin and fluorouracil, and this reduction seemed to be greater than it was in other groups that received fluorouracil only [
23].
The interaction between the cell and the matrix as mediated by integrin is organ-specific compared to that of other cell adhesion molecules [
2,
6], and β
1-integrin seems to be expressed at higher level in colorectal cancer [
24,
25]. Especially, it is reported to be related to metastatic colorectal cancers [
26]. β
1-integrin promotes the growth, survival, and invasion of many different types of tumor cells involving colorectal cancer by its attachment to different α-subunits. According to Yang et al. [
27], α
5β
1-integrin was expressed at higher levels in patients suffering from colorectal cancer than in normal patients, and the expression level was higher in cases of lower differentiation level and higher histological stage of the cancer. Moreover, the 5-year survival rate was significantly reduced in patients with high expression levels of α
5β
1-integrin. In the investigation of Bartolome et al. [
28] who used a mouse model, the results proved that liver metastasis of colorectal cancer might be caused by activation of α
5β
1-integrin. According to many reports of previous years, β
1-integrin is closely related to the progression of colorectal cancer, and it is regarded as a factor that worsens the prognosis of the patients [
6,
7,
25].
A small number of articles have reported that researchers tried to extract tumor cells from locations other than the bloodstream, such as lymph nodes, bone marrow, and peritoneal fluid. Their results showed a similar correlation to the prognosis for the patients [
2,
29,
30]. Wong et al. [
14,
29] reported that tumor cells were extracted in molecular tests even in patients with colorectal cancer and no metastasis to lymph nodes, and their survival rate was reduced due to a more frequent development of metastasis. The level of β
1-integrin was observed at higher levels in patients with colorectal cancer than in those with benign diseases (26.7% vs. 93.3%), and the expression level was higher for higher stages of colorectal cancer. Even in cases with the absence of lymphatic metastasis, vascular invasion, and lymphatic invasion, the expression level of β
1-integrin was quite high, but with no statistical significance.
When the expression levels that were measured before and after surgery in patients with colorectal cancer were compared, the expression level was significant reduced after surgery. This result proves the theory that the amount of β
1-integrin tends to decrease significantly as the tumor burden within the body becomes less. The expression level of β
1-integrin in patients with colorectal cancer is 93.3%, which is higher than the expression level of CEA, 73.3%. An analysis of the relationship between β
1-integrin and CEA showed a significant correlation between both patients with colorectal cancer and patients with benign diseases. Despite the lesser amount of β
1-integrin compared to that of CEA, which is specific of tumor cells, β
1-integrin can be expressed in endothelial cells, epithelial cells, white blood cells, platelets, and normal cells [
6]. Because the RNA of β
1-integrin was obtained from only the buffy coat layer, we cannot clearly confirm that the relationship we have discovered from our result proves whether the actual presence of β
1-integrin reflects the presence of circulating tumor cells or not. Consequently, additional investigations that use tumor tissue or lymph nodes other than blood samples e thought to be necessary in order to obtain more meaningful results.
This is a case-control study, which shows better utilization in regards to clinical application of the results that were observed in the investigation, and the significant results that were acquired from comparisons of the two markers seem to be very meaningful considering that no research articles have ever reported the relationship between β1-integrin and circulating tumor cells. Additionally, the measurement of β1-integrin by using RT-PCR is very simple, noninvasive, and cheap, and it can predict the presence of micrometastasis at an earlier stage compared to computed tomography and magnetic resonance imaging in patients with colorectal cancer. Thus, it seems to be a method that may be used repeatedly and usefully in such patients. In our opinion, the results of this study may be used during the follow-up of patients. In this way, patients with higher expression of β1-integrin can be selected, recurrence and metastasis in those patients can be discovered at an earlier stage, and more intensified treatment can be considered while planning treatment options. We also suggest that the results of the study be used for monitoring both the reaction of the patient to treatment and the progression of the disease after surgery. Furthermore, we think that new treatment methods can be investigated, with β1-integrin as a new target.
The expression level of β1-integrin was higher in patients with colorectal cancer, and that level was significantly reduced after surgery. Moreover, it showed a close relationship with the expression of CEA, which serves as a marker for circulating tumor cells. In conclusion, β1-integrin was shown to be a new potential marker that can be used in patients with colorectal cancer to predict the progression of the cancer and the prognosis after surgery and that, along with circulating tumor cells, can be used to detect micrometastasis earlier.