Study design
This was an analytical cross-sectional study among patients undergoing surgery at TMH during the period of October 2015 to September 2017. This study was conducted in the oncology and pathology department at Tongji medical hospital in Wuhan, Hubei, China, a department with the capacity of 4,000-6,000 beds.
Sampling:
Convenient sampling was method deployed in this study. Tissue samples of patient who attended at Tongji Medical Hospital in oncology and pathology department were collected after filling precoded questionnaire. A precoded questionnaire was used to obtain information of the participants .The questionnaire was divided into demographic characteristics, clinical characteristics and staining analysis. All tissue samples of patients above 19 years of age of both sexes who underwent curative or palliative surgery were included into the study. Patients whose biopsy sample for histopathological analysis could not be obtained and did not consent for the study were excluded from the study. Their samples were taken for further analysis of VEGF-C by immunohistochemistry. To ensure internal validity of the study precoded questionnaire was tested. XY and BM were trained how to fill the questionnaire.
Specimen Collection
Tissue samples analyzed in this study were collected from Chinese patients (n=161) who had undergone either a palliative surgical procedure or a therapeutic surgical procedure between October 2015 and September 2017 at Tongji medical hospital (Hubei, Wuhan, China). Patient data retrieved included: age, sex, date of birth, history of smoking and alcohol consumption, TNM stage, anatomical location of tumor, histopathological pattern of the cancer, tumor grade, Bormann’s classification, vascular and lymphatic involvement by tumor, tumor grading (well differentiated, moderate differentiated, and poorly differentiated), and positive VEGF status. TMN staging criteria were used according to the American Joint Committee on Cancer (AJCC) [17]. Tissues were then taken for further analysis of expression of VEGF-C.
Immunohistochemical analysis of VEGF -C
Immunohistochemical staining was done using a 4-μm thick paraffin-embedded section and then treated with 0.3% hydrogen peroxide at room temperature for ten minutes. These sections were heated in a solution (pH 6.0) of 1% mmol/L of trisodium citrate in a microwave for the extraction of antigen. Incubation of sections was done at a humidity condition of 4°C of primary antibodies (mouse monoclonal VEGF-C antibody) [1:100, DAKO]. Chosen slides were then washed three times in 0.1 mmol/L PBS for about 2 minutes and then incubated at standard room temperature with horseradish peroxidase (Envision, DAKO) conjugated mouse secondary antibody for 30 minutes. Negative control for VEGF-C detection was done using normal rabbit antibody IgG after development was done with 3,3'-Diaminobenzidine.
VEGF -C scoring according to immunohistochemistry
Two pathologists who were unaware of the clinical outcome of the patients did a pathological analysis of immunohistochemistry. The analysis of the staining was exclusively restricted to tumor cell reactions. VEGF-C stained results were classified, according to the intensity of color and percentage of epithelial cells that showed specific immunoreactivity, into 0, 1, 2, and 3 designations. If the summation of intensity and percentage was in the 0-2 ranges, it was considered a weak expression of VEGF-C, while a range of 3- 6 was regarded as a strong expression of VEGF-C [18, 19, 20].
Statistical data analysis
Data were entered into Excel, imported for analysis into Stat software version
13.v, and analyzed according to the objectives. Continuous variables were analyzed using mean and standard deviation. For categorical variables frequency and proportion was used. Chi-square test was used for comparison between the categorical variables groups. The odds ratio was used to measure strength of the association between predictor and outcome variables. A p-value of less than 0.05 was reported as significant. Predictor variables, which were found to be significant on univariate analysis, were subjected to multivariate logistic regression analysis to test the significance of the association between these variables and the outcome. Adjusted odds ratio presented at 95% Confidence Interval and P- value less than 0.05 were considered statistically significant.