Evaluation of angiogenesis with serum and tissue vascular endothelial growth factor, angiopoietin-1 and angiopoietin-2 levels in relation to clinicopathological features in lung cancer patients

Objective: The aim of this study was to investigate serum and tissue levels of vascular endothelial growth factor (VEGF), angiopoietin-1 (Ang-1) and angiopoietin-2 (Ang-2) and to evaluate correlations between serum and tissue parameters with respect to clinicopathological features in patients with lung cancer. Methods: The study was conducted on 34 patients with stage I-IV primary lung cancer and 32 healthy controls. Preoperative serum, tumor and matched normal tissue VEGF, Ang-1 and Ang-2 levels were determined with ELISA. Results: Serum Ang-2 levels were significantly increased in lung cancer patients (p < 0.001). VEGF levels were significantly higher in tumor tissue than in matched normal tissue (p < 0.001). Ang-1 and Ang-2 levels were significantly higher in normal tissue than in tumor tissue (p < 0.001). A significant negative correlation was found between normal lung tissue Ang-2 and serum VEGF levels (r = − 0.400, p = 0.019). A significant correlation was observed between serum and lung tissue Ang-2 levels (r = 0.397, p = 0.020). Conclusion: This study clearly demonstrated that VEGF, Ang-1 and Ang-2 are all involved in lung cancer process. This was the first study to show a correlation between serum and tissue levels of Ang-2 in lung cancer. This finding might be the basis for therapeutic strategies against lung cancer.

Angiogenesis is an essential process in tumorigenesis and pathogenesis of human malignancy [3]. When solid tumors become larger than 2-3 mm 3 , they cannot fulfill their needs of oxygen and nutrients through diffusion and need angiogenesis [4]. Angiogenesis is controlled by a balance of stimulating and inhibiting regulators involved in multiple pathways which result in endothelial cell proliferation, differentiation and organization into a functional network of vascular channels [4]. When the balance between stimulating and inhibiting factors in angiogenesis is disrupted in favor of stimuli, neovascularization occurs in the tumor tissue [5,6].
VEGF is one of the most important angiogenic factors and effective in all stages of angiogenesis [10]. It is overexpressed by a majority of solid tumors, which increases vascular permeability and promotes endothelial cell proliferation, survival and migration in both physiological and pathological angiogenesis [11].
Other important factors in angiogenesis are angiopoietin-1 (Ang-1) and angiopoietin-2 (Ang-2). The former stimulates interaction between endothelial cells and extracellular matrix around it and stabilizes vessels. The latter antagonizes the stabilizing effect of Ang-1 and causes destabilization in vessels [12,13]. The destabilization created by Ang-2 as an anti-angiogenic factor goes through angiogenic change in the presence of VEGF [9].
Most of the studies have investigated VEGF, Ang-1 and Ang-2 in serum or tissue, whereas we used both serum and tumor tissue and matched normal tissue in lung cancer patients. Thus, the aim of this study was to determine VEGF, Ang-1 and Ang-2 levels in patients with lung cancer and to evaluate relations between these angiogenic factors in terms of clinicopathological parameters in these patients.

Subjects
The study was approved by Clinical Research Ethics Committee of Dokuz Eylül University School of Medicine. It included 34 patients with stage I-IV primary lung cancer and 32 healthy controls. Thirty two of the patients were diagnosed with non-small cell lung cancer (NSCLC) including 18 squamous cell carcinoma, nine adenocarcinoma and five mixed adenocarcinoma and two were small cell lung cancer (SCLC). The patients were selected at the Department of Thoracic Surgery at Dokuz Eylül University Hospital and informed consent was obtained from all the cases. Exclusion criteria were malignancies other than those of the lungs, lung carcinoid tumors, and positive history of chemotherapy and/or radiotherapy. Healthy controls who had already undergone routine and clinical analyses were selected from Smoking Cessation Outpatient Clinic and assigned into the control group. Demographic and anthropometric data and medical history of the subjects and the healthy controls were recorded. Routine biochemical and clinical analyses were performed in the Central Laboratory of Dokuz Eylül University Hospital directly after sample collection. Preoperative blood specimens were collected and stored at −40°C until their analyses were made.
In addition, tumor tissues (300-400 mg) and matched normal lung tissues 5 cm away from tumor were obtained from the fields of the surgical specimens approved by the pathologist. These specimens were rinsed with 0.9% NaCl and kept at −80°C until they were analyzed. Clinicopathological features including age, gender, tumor size, histological types of tumors, histological differentiation degree, tumor node metastasis (TNM) on pathological examinations and invasions into veins, arteries, lymph vessels and nerve tissue were evaluated. Tumor stage was determined according to 2004 WHO classification of lung tumors [14].

Measurement of VEGF, Ang-1 and Ang-2 levels
Blood samples were centrifuged at 4000 rpm for 10 min and aliquots of supernatants were stored at −80°C until testing. Frozen serum samples were mixed thoroughly after thawing and centrifuged before analysis. Samples with more than a trace of hemolysis were discarded.
VEGF, Ang-1 and Ang-2 protein levels were measured by ELISA according to the manufacturer's instructions (R&D Systems, Abingdon, Oxfordshire, UK). The absorbance was measured at 450 nm using a microplate reader (BioTek KC4, USA). Quantifications were achieved by construction of standard curves using known concentrations. The results were expressed as picogram per milliliter (pg/mL) for serum and picogram per milligram of protein (pg/mg protein) for tissue samples. Assay reproducibility was assessed by precision of intra-day measurements (n = 10). The intra-day CVs for VEGF, Ang-1 and Ang-2 were 8.4%, 6.9% and 7.3%, respectively.

Statistical analysis
Statistical analyses were performed using SPSS 15.0. All measurements were done in triplicate. Each variable was checked for normality of distribution by Shapiro-Wilk normality test. Student's t-test was used for normally distributed variables in the analysis of continuous variables. Variables with non-Gaussian distribution were compared by using Mann-Whitney U test and Kruskal-Wallis test. The data were expressed as mean ± standard deviation for normally distributed variables and median and range for non-normally distributed variables. Pearson and Spearman correlation coefficients were used for the analysis of correlations for continuous normally and non-normally distributed variables, respectively. Values of p < 0.05 were considered to indicate a statistically significant difference.

Results
A total of 34 patients with primary lung cancer presenting to respiratory surgery department and 32 healthy controls referring to smoking cessation outpatient clinic of Dokuz Eylül University Hospital were included into the study. Out of 34 patients, 18 (53%) had squamous cell carcinoma, nine (26%) had adenocarcinoma, five (15%) had mixed adenocarcinoma and two (6%) had small cell carcinoma. Diagnoses in pathological examinations revealed that tumors were well-differentiated in 12 patients, moderately differentiated in 10 patients and poorly-differentiated in 12 patients. Thirty-two individuals (26 males and six females) referring to smoking cessation outpatient clinic were found to be healthy based on routine investigations. There was no significant difference in age between the patients and the healthy controls (p > 0.97).
Serum VEGF, Ang-1 and Ang-2 levels were compared between the patients with lung cancer and the healthy controls. While the patients and the controls did not have a significant difference in terms of serum VEGF and Ang-1, the patients had significantly higher levels of Ang-2 levels (p < 0.001) ( Table 1). VEGF, Ang-1 and Ang-2 levels in tumor tissues and matched normal tissues in the patients were determined. VEGF levels were significantly higher in tumor than in normal matched normal tissues (p < 0.001) ( Table 2). Ang-1 and Ang-2 levels were significantly lower in tumor than in matched normal tissues (p < 0.001) ( Table 2).
VEGF, Ang-1 and Ang-2 levels in serum and tumor tissues were evaluated with respect to clinicopathological parameters of lung cancer patients. Serum VEGF levels were significantly higher in the patients with tumor stages T3 and T4 than in those with tumor stages T1 and T2 (p = 0.042) ( Table 3). Ang-1 levels in tumor tissues were also significantly higher in the patients aged over 65 years than in those younger than 65 years old and in the patients with moderately and poorly differentiated tumors than in those with well-differentiated tumors (p = 0.038 and p = 0.044, respectively) ( Table 3). No significant differences were found in sex, tumor size, histological type, lymph node, perineural invasion, artery invasion, lymphovascular invasion and vein invasion between VEGF, Ang-1 and Ang-2 levels in serum and tumor tissues (data not shown).
In the patient group, there was a significant positive correlation between serum VEGF and Ang-1 levels (r = 0.446, p = 0.000), a significant negative correlation between serum VEGF and normal tissue Ang-2 levels (r = − 0.400, p = 0.019) and a significant positive correlation between serum Ang-2 levels and normal tissue Ang-2 levels (r = 0.397, p = 0.020).

Discussion
Lung cancer is the leading cause of cancer related deaths with a survival rate of 10% [15]. The reason for the low survival rate is the lack of early diagnosis methods. Therefore, early diagnosis and effective treatment of this malignant disease become more important [16]. Just as all malignant solid tumors, tumors in lung cancer need new vascularization in their primary foci and metastases so that they can fulfill their needs of oxygen and nutrients for growth [4]. There is a widespread agreement that angiogenic activation of a tumor results from disruption of a balance between positive and negative factors of angiogenesis [5,6]. It is important to use stimulating and inhibiting factors in angiogenesis as early markers of monitoring tumor progression, to decide strategies to be used in treatment and to develop new treatment methods. Several studies have revealed that VEGF and angiopoietins play a critical role in tumor angiogenesis [7][8][9].
In the present study, VEGF, Ang-1 and Ang-2 levels in serum, tumor tissues and matched normal tissue supernatants from lung cancer patients and healthy controls were determined and obtained results were evaluated together with clinicopathological parameters. Furthermore, the patients were divided into groups according to clinicopathological parameters including gender, age, tumor differentiation, histological type, lymph node (N), tumor invasion (perineural, venous and lymphatic), and VEGF, Ang-1 and Ang-2 levels were compared between these groups.
Regarding VEGF, despite lack of a significant difference between serum levels in the patients and healthy controls, the higher values found in tumor tissues than in matched normal lung tissues seem to support the critical Results were expressed as median (interquartile range); a p < 0.05 was statistically significant. Results were expressed as median (interquartile range); a p < 0.05 was statistically significant.
role of VEGF in angiogenesis. It has also been reported before that although VEGF levels in tumor cells can be increased, serum VEGF levels may not be sufficiently high to measure [22]. On the other hand, as the patients were subdivided into groups, serum VEGF were significantly higher in the patients with tumor stages T3 and T4 than in those with tumor stages T1 and T2. This suggests that VEGF can be associated with tumor size and confirms the idea that it is one of the most important angiogenic factors in angiogenesis mediated tumor growth. Most of the studies have reported no statistically significant difference for tumor stage and histological type in terms of VEGF [22,28,29]. Similar to our findings, increased serum VEGF levels were found in advanced tumor stages compared to those in early tumor stages [30][31][32].
Having an angiogenic effect in normal and developing tissues, Ang-1 levels were significantly lower in tumor tissues than in normal lung tissues, which is consistent with the results of a study by Park et al. [28]. Conflicting findings about Ang-1 from other studies suggest Ang-1 expression is heterogeneous and does not always reflect the whole tissue [27][28][29][30]. Due to its angiogenic effect, Ang-1 has a crucial role in development and growth of tumors. In the present study, higher levels of Ang-1 found in moderately and poorly differentiated tumors compared to highly differentiated ones show more aggressive growth.
Serum Ang-2 levels were found to be significantly higher in the lung cancer patients than in the healthy controls. It has been proposed that increased Ang-2 levels are a sign of survival and prevent tumor growth. However, VEGF upregulation accompanied by increased Ang-2 levels stimulates angiogenesis at tumor margins, which causes tumor growth. Consistent with the present study, other studies found significantly higher Ang-2 levels in lung cancer patients than in healthy controls [22,28,33,34]. In addition, Canadas et al. measured Ang-2 levels in lung cancer patients with a method different from the one used in the present study (Luminex Technology) and reported significantly higher Ang-2 levels [35]. When angiopoietins were evaluated according to the tumor stage and histologic type, supporting our findings most of the studies have found no significant relation [3,12,22,29]. On the other hand, Park et al. have shown that serum Ang-2 levels were increased in stage II than those in stage I [28].
There have been few studies on the relation between clinicopathological features and VEGF, Ang-1 and Ang-2 in lung cancer patients, with conflicting results. These studies have been directed towards evaluation of either one or two of the parameters serum VEGF, Ang-1 or Ang-2 and their results have been presented semi-quantitatively  [12, 17-19, 21, 22, 24, 27, 28, 35]. However, in the current study, both serum and tumor and matched normal tissue levels of these parameters were investigated. In the present study, the significant negative correlation found between serum VEGF and lung tissue Ang-2 reveals clearly the opposing roles of these factors. On the other hand, a moderately significant positive correlation between serum VEGF and Ang-1 was found. This positive correlation was supported by the fact that both VEGF and Ang-1 are angiogenic factors. Park et al. reported that serum VEGF was correlated with Ang-1 and Ang-2, but there was no correlation between Ang-1 and Ang-2 [28]. However, in the present study, there was not a significant correlation between serum VEGF and serum Ang-2 levels. One of the most striking findings of the present study was presence of a significant positive correlation between normal tissue Ang-2 and serum Ang-2, reflecting a parallel rise in both tissue and serum. This finding suggests that serum levels of these parameters can be a potential surrogate marker.
The limitation of this study is that the number of patients in the subgroups based on histological tumor types was relatively small (nine patients with adenocarcinoma and 18 patients with squamous cell carcinoma). Therefore, further studies with larger samples are needed.
The results of this study are important in that it increases our understanding of the role of different angiogenic parameters in lung cancer. To conclude, this study formed a basis for further studies on angiogenic and antiangiogenic factors in lung cancer, especially Ang-2. The finding that Ang-2 levels were increased in both normal lung tissue and serum will provide guidance for follow-up and evaluation of patients and creation of new treatment strategies.