| Objective: 1.To determine the expression of E-cadherin and N-cadherin,markers of epithelial-mesenchymal transition in gastric cancer tissues after operation,and to analyze their relationship with clinicopathological parameters.2.To analyze metabolic profiles of metabolites in serum samples and explore the relationship between epithelial-mesenchymal transition and serum metabolites in gastric cancer.3.To explore the independent risk factors of epithelial-mesenchymal transition in gastric cancer,a risk scoring model based on serum metabolites and clinicopathological data was established.Methods: 1.The expression of E-cadherin and N-cadherin was detected by immunohistochemical staining in 51 patients.The patients with epithelial-mesenchymal transition,i.e.those with low expression of E-cadherin and high expression of N-cadherin,were included as a group.The remaining patients did not undergo epithelial-mesenchymal transition.The patients were classified as another group.SPSS software was used to analyze the relationship between the two groups and their clinicopathological parameters.2.The metabolic profiles of metabolites in serum samples were analyzed by high performance liquid chromatography-mass spectrometry(LC-MS).The original data were obtained.Then the two groups of patients were analyzed by data analysis software.The differentiated metabolites were screened by searching HMDB(Human Metabonomics Database).3.The independent risk factors associated with epithelial-mesenchymal transition were selected by single factor and multifactor logistic regression.Results: 1.In 51 cases of gastric cancer,the expression of N-cadherin was high and E-cadherin was low in 24 cases,and the rest 27 cases were detected by immunohistochemistry.PCA(Principal Component Analysis)with Markerview software showed that the two groups of patients could be easily separated;Peakview software was used to observe the difference of metabolites between the two groups in serum total ion flow chart;179 serum metabolites were found to be different between the two groups by t-test of cations in metabolites(P < 0.05).By using BRB-Array Tools software for discriminant analysis and inquiring HMDB(Human Metabolomics Database),seven specific serum metabolites were screened out.By using RStudio software and clustering analysis,the corresponding thermal map can be obtained,which can be observed to distinguish two groups of patients.2.PCA(Principal Component Analysis)by Markerview software showed that the two groups could be separated;Peakview software was used to get the total ion flow chart of the serum,which could observe the existence of different metabolites between the two groups;t test(make P < 0.05)showed that there were 179 different serum metabolites between the two groups.After discriminant analysis with BRB-Array Tools software,13 products with 80%-100% contribution were selected.By inquiring HMDB(Human Metabolomics Database),the names of metabolites can be selected.By clustering analysis with RStudio software,the corresponding thermal map can be obtained,and it can be observed that two groups of patients can be distinguished.3.In combination with logistic regression,a risk scoring model for predicting epithelial-mesenchymal transition in gastric cancer was established.The ROC curve of this model as a diagnostic test showed better predictive of epithelial-mesenchymal transition compared to a single independent risk factor.Conclusion: 1.Patients with epithelial-mesenchymal transformation are prone to lymph node metastasis,TNM staging is relatively late,and the degree of differentiation is low.It was not related to age,sex and tumor size.2.There are differences in serum metabolites between patients with gastric cancer who have epithelial-mesenchymal transformation and those who have not.3.Trihexosylceramide and lysophosphatide(22:4)are independent risk factors for epithelial-mesenchymal transition in gastric cancer. |
PDF Full Text Request