Abstract
Background: The purpose of the present study was to determine the extent of expression of proteins that are the elements of the plasminogen activation system, i.e. urokinase plasminogen activator (uPA), receptor of uPA (uPAR), plasminogen activating inhibitor-1 (PAI-1) and tissue plasminogen activator (tPA) in the primary site of rectal adenocarcinoma, as well as in lymph node metastases, and their correlation with clinical and pathological data. Materials and Methods: The study material consisted of archival paraffin blocks from 108 patients with confirmed oncological diagnosis, treated in 2002 and 2003 in the Lower Silesian Oncology Centre in Wroclaw. This material was used for immunohistochemical reactions with antibodies against uPA, uPAR, PAI-1 and tPA. The extent of expression was evaluated semi-quantitatively based on the immunoreactive score according to Remmele and Stegner. The obtained results were correlated with clinical and pathological data: stage of the disease (modified Dukes' classification of Astler and Coller), grade of histological malignancy, event-free survival and overall survival. Results: It was found that cytoplasmic expression of uPA, uPAR, PAI-1 and tPA was higher in cells of tumor and metastases of perienteric lymph nodes in comparison to normal rectal tissues. Positive correlation was shown between expression of PAI-1 and tPA, uPA and uPAR, as well uPA and PAI-1 (r=0.70, r=0.77 and r=0.34, respectively; p<0.05 for all). Additionally, in patients with low and moderate expression of uPA and uPAR, the overall survival rate was higher in comparison to patients with high expression of the studied markers. Conclusion: The intensity of expression of uPA and uPAR might be a prognostic factor of survival time for patients with primary rectal adenocarcinomas.
The plasminogen activation system (PAS) consists of: plasminogen, proenzyme of plasmin (glycoprotein); serine protease, plasmin; two plasminogen activators: urokinase-type (uPA) and tissue-type (tPA); receptor for uPA, uPAR; and two plasminogen-activating inhibitors belonging to the family of serine protease inhibitors (SERPIN): PAI-1 and PAI-2 (1-4). Under physiological conditions, PAS participates in fibrinolysis and in processes in which basal membrane destruction-related tissue restructuring occurs: e.g. rupturing of Graaf's follicle during ovulation and blastocyst implantation, mammary gland involution after lactation, and wound healing (1).
PAS might play a role in cancer cell invasion and formation of metastases by means of catalysis of proteolytic processes, degradation of basal membrane proteins and extracellular matrix, as well as the initiation of intracellular signal transduction cascade (1, 2, 4, 5). Moreover, significantly higher expression of uPA, uPAR and PAI-1 was found in neoplastically -changed cells than in corresponding normal tissues (6-8). In patients with high expression of uPA, uPAR and PAI-1 in cancer cells, shorter overall (OS) and event-free (EFS) survival were found in comparison to patients having low expression of those markers (8). On the other hand, tPA expression was correlated with a good prognosis (2, 10). High expression of PAS elements was observed in cancer cells of the breast (8, 9), lung (11), prostate (12) and ovary (13, 14), in squamous cell head and neck cancer (15, 16) and squamous cell skin cancer (17). Few studies have been published so far wherein the prognostic value of PAS in colorectal cancer was evaluated. They show inter alia that high uPA, uPAR and PAI-1 expression is associated with shortened survival. However, as far as we are aware of, there are no studies on the prognostic effect of tPA, nor on how the elements of PAS affect relapse of the disease or on their expression in cancerous lymph nodes.
The aim of the present study was to determine the extent of expression of PAS elements (uPA, uPAR, PAI-1, tPA) in cells of primary rectal adenocarcinoma and in metastases of perienteric lymph nodes. Additionally, the purpose was to correlate the intensity of expression of these PAS markers with clinical and pathological data of rectal adenocarcinoma cases namely: stage of the disease (modified Dukes' classification of Astler and Coller), grade of histological malignancy, EFS and OS (18, 19).
Materials and Methods
Patients. The study material consisted of archival paraffin blocks from 108 patients of age ranging from 34 to 98 years (39 women and 69 men), treated in 2002 and 2003 in Lower Silesian Oncology Centre in Wroclaw. In all cases, the diagnosis of rectal adenocarcinoma was confirmed by histopathological examination. The grade of histological malignancy was verified according to the WHO classification (18) (Table I). Clinical and pathological features were determined based on patients' medical records, including analysis of survival time, time to relapse and the stage of carcinogenesis according to the modified Dukes' classification of Astler and Coller, which was used during the treatment and assessed the depth of infiltration of the intestinal wall and metastasis to lymph nodes and distal organs (19) (Table I). In recent years, this classification has been replaced by TNM classification system (20). Clinical stages according to TNM classification system and their corresponding stages according to the modified Dukes' classification of Astler and Coller are shown in Table II. The control group consisted of 15 normal rectum specimens (surgical margins) from the same patients, taken during tumor removal surgery. All studies were conducted with the consent of the Ethics Committee of Wroclaw Medical University (consent no.: KB-67/2009).
Immunohistochemistry (IHC). Reactions were carried out on paraffin sections with the use of mouse monoclonal antibodies anti-uPA (dilution 1:50; American Diagnostica, Stamford, CT, USA), anti-uPAR (dilution 1:50; Dako, Glostrup, Denmark), anti-PAI-1 (dilution 1:50; Leica-Microsystems, Wetzlar, Germany) and anti-tPA (dilution 1:100; Thermo Scientific, Waltham, MA, USA). Proteins were visualized through subsequent incubations with the solution of biotinylated antibodies, streptavidin-horseradish peroxidase complex and diaminobenzidine (DAB) chromogen. All reagents were obtained from Dako. The semiquantitative immunoreactive score (IRS) according to Remmele and Stegner, which indicates the percentage of positively stained cells and the intensity of the color reaction, was used to evaluate the extent of expression of PAS antigens in cancer cells (Table III) (21). With the use of an Olympus BX-41 double-headed microscope, two pathologists (MZ, PD) independently conducted the evaluation of encoded preparations.
Statistical analysis. The results were subjected to statistical analysis with the use of STATISTICA 7.1 (StatSoft, Krakow, Poland). After analysis using the Shapiro-Wilk test, for subsequent analysis the following tests were used: Kruskal–Wallis test, Mann–Whitney U-test and Spearman's rank correlation. For the evaluation of patient survival and the probability of relapse, the log-rank test (Mantel-Cox test) was used. Test results are presented with the use of Kaplan–Meier curves. Statistically significant p-values for individual tests were accepted at p<0.05.
Distribution of 108 cases of rectal adenocarcinoma taking into account grade of histological malignancy and stage according to modified Dukes' classification of Astler and Coller (19).
Comparison of TNM and Dukes' (modified by Astler and Coller) classification (19).
Results
IHC reactions were conducted on all 108 cases of rectal adenocarcinoma and 15 samples of normal tissue and showed cytoplasmic expression of variable intensity (Figure 1). Preparations evaluated for the expression of uPA, uPAR, PAI-1 and tPA were divided into two independent groups according to the IRS: one of low and moderate expression (score=0-4) and the other of high expression (score=6-12). The IHC expression of individual PAS elements in neoplastically changed tissues is shown in Table IV. The intensity of all PAS element expression in tumor cells was higher than that in control tissue. The differences were statistically significant for PAI-1, uPA and uPAR (p<0.01, p<0.001 and p<0.0001, respectively). Only for tPA there was no statistical significance found, although its expression in neoplastically-changed tissue was also higher (Figure 2).
Evaluation of the extent of the plasminogen activation system element expression with the use of immunoreactive score according to Remmele and Stegner (21). Σ=A×B (score ranges from 0 to 12).
Distribution of the plasminogen activation system elements expression in 108 cases of rectal adenocarcinoma (immunoreactive score).
Expression of the plasminogen activation system elements in rectal adenocarcinoma cells. Cytoplasmic immunohistochemical reaction of high intensity (×100): A: Urokinase plasminogen activator, B: urokinase plasminogen activator receptor, C: plasminogen activating inhibitor-1, D: tissue plasminogen activator.
The expression of PAS elements was investigated in primary tumors as well as in metastatic perienteric lymph nodes in a sub-group of 43 cases (39.81%) of lymph node metastasis (Dukes' stage C). However, no statistically significant differences were found between the expression of proteins in the group of primary tumor and metastases (Figure 2).
Extent of expression of urokinase plasminogen activator (uPA), receptor of uPA (uPAR), plasminogen activating inhibitor-1 (PAI-1) and tissue plasminogen activator (tPA) in primary rectal adenocarcinomas and their metastases of lymph nodes in comparison to normal rectal tissue (control vs. tumor: *p<0.001; **p<0.0001; ***p<0.01). IRS: immunoreactive score.
In order to investigate the relationship between the expression of individual PAS elements in cancer, analysis of Spearman's rank correlation coefficients was performed. A strong positive correlation was found between the intensity of expression of PAI-1 and tPA, as well uPA and uPAR, whereas a moderate correlation was found between expression of uPA and PAI-1 (r=0.70, r=0.77 and r=0.34, respectively; p<0.05 for all).
Furthermore, the association was studied between IHC expression of PAS elements and clinicopathological factors. In cases of rectal adenocarcinoma, expression of uPA and uPAR increased with increasing histological malignancy grade, whereas PAI and tPA expression showed an opposite tendency, reaching the highest values in G1 grade. However, the differences in expression between different malignancy grades was not statistically significant (data not shown).
Moreover, no statistically significant differences were found in the expression of PAS elements according to the stage of disease. Nevertheless, a tendency was visible towards increasing PAS element expression in higher Astler-Coller stages. Particularly in stage C1, a clearly higher uPA, PAI-1 and tPA expression was observed (data not shown).
In order to determine the effect of PAS elements on prognosis, OS and EFS was evaluated depending on the extent of expression of the studied markers. Statistically significant difference was found in patients' survival depending on the expression of uPA and uPAR. Patients with low and moderate expression of uPA (Figure 3A) and uPAR (Figure 3B) had a longer survival in comparison to patients with high expression. However, no statistically important difference was found in patients' survival depending on the intensity of expression of PAI-1 and tPA. Moreover, no specific differences were found in the time to relapse in patients with different expression of PAS elements (data not shown).
Discussion
Colorectal cancer is one of the most commonly diagnosed malignant tumors in humans. Its most frequent localization (30-50% of cases) is the rectum, with the predominant histological pattern being adenocarcinoma. The incidence of colorectal cancer is higher mainly in developing and highly developed countries. It is an important social problem because it frequently occurs at the maximum of a patient's professional activity (the risk of the disease increases from 50 years onward, with the maximum after 70 years) (22). The treatment of colorectal cancer depends on the stage of the disease, which is the most important prognostic factor. In patients with cancer cell infiltration beyond the wall of the intestine and with metastases to nearby lymph nodes, it is necessary to include radio- and chemotherapy in addition to surgical treatment. Patients with disease limited to the intestinal wall and without lymph node metastases usually do not require additional oncological treatment after the surgical removal of tumor. However, based on current clinical observations, it seems necessary to determine the sub-group of patients in need of such treatment (22). Therefore, it is important to search for new prognostic factors that would allow such a therapeutic approach to these lesions.
In recent years, it has been observed that expression of PAS elements might be a prognostic factor in various types of cancer. Their expression is significantly higher in various types of cancer cells than in corresponding normal tissues (6-8, 23). The extent of expression of PAS elements is usually associated with shorter EFS and OS (9). Only expression of tPA seems to be associated with a good prognosis (2, 10).
Tumors of the large intestine show higher expression of uPA, uPAR and PAI-1 than normal intestinal tissue (23-30). Expression of uPAR was shown to increase during neoplastic transformation of adenocarcinoma into invasive colorectal cancer. Due to carcinogenesis, in the normal intestinal epithelium expression of PAI-1 is also increased (24, 31). Those results clearly show the role of PAS elements in neoplastic transformation of cells of the mucous membrane of the large intestine. Our studies confirmed higher expression of uPA, uPAR and PAI-1 in tumor in comparison to normal rectal tissue. However, there exist discrepancies regarding tPA expression. In the studied material, tPA expression was higher (although without statistical significance) in cancer cells, but this was not confirmed in other reports. Herszenyi et al. (26) and Raigoso et al. (32) showed that the level of tPA was significantly higher in normal mucous membrane than in neoplastic cells in patients with colorectal cancer (26, 32). A similar intensity of expression of PAS elements was observed in the primary tumor and lymph nodes analyzed here. There are no other reports available on this subject as far as we are aware. However, these results might suggest that the evaluation of expression in tumor cells only seems to be sufficient to determine the role and prognostic values of PAS elements.
Dependency of patient survival on the extent of urokinase plasminogen activator (uPA) (A) and receptor of uPA (uPAR) (B) expression in rectal adenocarcinoma (p<0.01 and p<0.05, respectively).
The next stage of our study was the determination of the relationship between PAS elements and selected clinical and pathological factors. According to literature reports, expression of uPA, uPAR, PAI-1 and PAI-2 correlates with cancer stage, the size of the tumor, the grade of malignancy, depth of infiltration and cancer cell invasion into blood vessels (23, 25, 27-29, 33-36). Herszenyi et al. showed significantly higher expression of uPA and PAI-1 in tumors of the highest grade of malignancy (G3) in comparison with cancer of low (G1) and moderate (G2) malignancy (26). On the other hand, Raigoso et al. did not find significant correlation between the level of tPA expression and clinicopathological features, such as Dukes' stage or the grade of histopathological malignancy (32).
In our material, no significant changes were found in the expression of PAS elements depending on the grade of histological malignancy and Dukes' stage of the disease. However, an increasing trend in uPA and uPAR level of expression, as well as a decreasing trend in PAI-1 and tPA expression with increasing grade was observed. The trend for increasing PAS expression was also shown in higher Dukes' stages of the disease.
Another important issue is the evaluation of the effect of PAS element expression on patients' OS and EFS. The results published so far show that increased expression of uPA, uPAR and PAI-1 is an independent negative prognostic factor of shorter survival (26). However, the effect of tPA on the prognosis and on the expression of PAS element have not been evaluated in previous studies. Herszenyi et al. showed longer 5-year survival in patients with a low level of uPA and PAI-1 (26). Higher expression of uPA, uPAR and PAI-1 is also associated with more frequent occurrence of liver metastases (33, 34, 37). Abe et al. (25), Terada et al. (33) and Halamkova et al. (35) showed that high tumor expression of uPAR was associated with shorter survival (25, 33, 35). In contrast to Angenete et al. who showed that PAI-1 expression is an independent negative prognostic factor of shorter OS in colorectal cancer (28), no such relationship was found for the rest of the PAS elements. Märkl et al. showed that high PAI-1 expression is an independent prognostic factor of distant metastasis in colorectal cancer (23). On the other hand, Raigoso et al. found that high expression of tPA in patients with surgical colorectal cancer was correlated with shorter OS and EFS (32), whereas Herszenyi et al. did not find any correlation between expression of tPA and survival (26). Halamkova et al. showed, however, that higher expression of uPA, uPAR, PAI-1 and PAI-2 was associated with poorer response to the treatment (36).
In our studies, a statistically significant difference was found in patients' survival depending on the expression of uPA and uPAR. Patients with low and moderate expression of uPA and uPAR had longer survival in comparison to patients with high expression. Therefore, high expression of uPA and uPAR seems to be a negative prognostic factor of shorter survival. We did not find a similar relationship for the expression of PAI-1 and tPA. In the case of PAI-1, there was a trend towards higher percentage of survival in patients with low and moderate expression, but this was statistically insignificant. There are no reports in the literature regarding the effect of expression of PAS elements on relapse-free survival in rectal cancer. In the analyzed material, we found no differences in time to relapse for patients with different expression of individual elements of PAS, which may have been due to the small size of the study group.
The results of our observations, wherein showing the effect of PAS elements on the prognosis, seem to confirm the hypothesis of the involvement of this system in the development and progression of tumors, also including malignant colorectal carcinomas. Tumors having high expression of PAS elements have increased invasiveness and metastatic potential inter alia due to their stronger ability to proteolytically degrade the surrounding extracellular matrix (2). Our studies and the obtained results seem to confirm this mechanism indirectly. In the future, the presented data might result in the use of PAS elements in prognostic and predictive evaluation supporting colorectal cancer therapy.
Acknowledgements
The work was supported by the funds of the research project no. 1710AM entitled: ”Expression of elements of the plasminogen activation system (uPA, uPAR, PAI-1, tPA) in different stages of colorectal cancer and its prognostic value”.
- Received July 8, 2015.
- Revision received August 22, 2015.
- Accepted August 31, 2015.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.