Date Published: April 16, 2018
Author(s): George Emmanouil, George Ayiomamitis, Adamantia Zizi-Sermpetzoglou, Maria Tzardi, Andrew Moursellas, Argyro Voumvouraki, Elias Kouroumalis.
To study the expression of angiodrastic chemokines in colorectal tumors and correlate findings with clinicopathological parameters and survival.
The proangiogenic factor VEGF, the angiogenic chemokines CXCL8 and CXCL6, and the angiostatic chemokine CXCL4 were measured by ELISA in tumor and normal tissue of 35 stage II and III patients and correlated with the histopathology markers Ki67, p53, p21, bcl2, EGFR, and MLH1 and 5-year survival. The Wilcoxon and chi-square tests were used for statistical comparisons.
There was a significant increase of CXCL6 (p = 0.005) and VEGF (p = 0.003) in cancerous tissue compared to normal. Patients with lower levels of CXCL8 and CXCL4 lived significantly longer. Patients with loss of EGFR expression had higher levels of CXCL8 while p21 loss was associated with higher levels of CXCL6. Chemokine levels were not correlated with TNM or Dukes classification. Strong expression of p53 was accompanied by decreased survival.
(1) The angiogenic factors CXCL6 and VEGF are increased in colorectal cancer tissue with no association with the clinical stage of the disease or survival. (2) However, increased levels of tissue CXCL8 and CXCL4 are associated with poor survival. (3) Strong expression of p53 is found in patients with poor survival.
The incidence of cancer is increasing every year. Colorectal cancer (CRC) is the second most common cause of cancer mortality in the Western world . Many factors both environmental and genetic are implicated in the propagation and mortality caused by CRC. Among various trophic factors, chemokines have a predominant role.
Patients with biopsy-confirmed colorectal cancers were recruited for participation in the current study. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the University Hospital of Heraklion, Heraklion. In all cases, written consent from the participants was obtained. Selection of patients was based on the following criteria:
Only patients with stages II and III according to TNM classification (stages B and C according to Dukes classification as modified by Astler-Coller) were included. Metastatic disease (type IV) patients were excluded [39–41].Only patients with a curative (R0) surgical resection who did not receive adjuvant chemotherapy (either refused or were not considered as candidates by the attending doctors) were included.Only patients with at least a 5-year follow-up (or death before that) were included in the report. Patients lost to follow-up were not included.
For every patient, immunohistochemical detection of MLH1, Ki-67, bcl2, p53, and p21 and EGFR protein expression were studied in tumor tissue by immunoperoxidase staining in 3 steps using a Dako kit as previously described .
Concentration of 4 chemokines (CXCL8, CXCL4, CXCL6, and VEGF) was calculated through ELISA protocols. All tissue samples, both cancer and control, were homogenized with a glass homogenizer in 1 ml 0.25% BSA phosphate-buffered saline (PBS) on ice. Immediately after homogenization, the samples were aliquoted and frozen at −80°C till further analysis. For each one of these chemokines, commercially available monoclonal antibodies and biotinylated antibodies were obtained (R&D Systems).
Statistical analysis was performed with the IBM SPSS statistics software version 19. Results are expressed as means ± standard deviation of the mean and were depicted as box plots. The nonparametrical test Wilcoxon signed-rank test for paired samples was used when the Kolmogorov-Smirnov method showed that the distribution of values was not normal. The chi-square test for the analysis of nonparametric data in 2 × 2 tables was used for associations between histopathology markers and 5-year survival. Statistical significance was set at the 5% level (p = 0.05).
The Kolmogorov-Smirnov test showed that chemokine concentrations in both tumor and normal tissues were not normally distributed. Figure 1 shows that there are no significant differences between tumor and normal tissue for CXCL8 (p = 0.177) and CXCL4 (p = 0.795).
Due to lack of normality in distribution, the Wilcoxon test was used for comparisons. As shown in Table 2, there is a significant difference in 5-year survival for CXCL8 and CXCL4. Patients who survived had significantly lower levels of those chemokines as compared to nonsurvivors. The same tendency existed for CXCL6 and VEGF, but this was not statistically significant due to the high scattering of results.
No difference of chemokine levels was found according to Dukes or TNM staging (Tables 4 and 5).
It is usually stated that bowel adenocarcinomas arise from epithelial cells. However, it is accepted today that the interaction between tumor cells and the tumor microenvironment is equally important for tumor evolution. Among others, expression of angiodrastic agents is particularly effective as they can induce or inhibit neovascularization, a process vital for tumor progression [47, 48]. Colorectal cancer is the second leading etiology of cancer death in Western countries. Almost half of the patients die of metastatic disease after curative surgery despite adjunct chemotherapy [49, 50]. TNM classification is considered to be the best prognostic factor in early stages of colon cancer .