Abstract:

Objective To analyze gene expression data of esophageal adenocarcinoma （EAC） in the Gene Expression Omnibus （GEO） and The Cancer Genome Atlas （TCGA） databases and clarify the relationship between the potential core genes and tumor lymphocyte infiltration in the EAC， and to provide the molecular targets for the diagnosis and treatment of EAC. Methods The high-throughput chip datasets GSE13898， GSE26886， GSE74553， and GSE92396， including EAC and normal esophageal tissues， were downloaded from the GEO database by searching for “esophageal adenocarcinoma”. The limma package of R software was used to screen the differentially expressed genes （DEGs） in EAC tissue and esophageal normal tissue， and the common DEGs were obtained through Venn diagram. After the DEGs were analyzed by STRING database， the results were imported into Cytoscape software to screen the core genes and construct the protein-protein interaction （PPI） network. The Gene Expression Profiling Interactive Analysis （GEPIA） database was used to verify the expression levels of core genes. The UALCAN and Kaplan-Meier Plotter databases were used to analyze the correlations between the core genes and prognosis and clinical data of the EAC patients. The Tumor Immune Estimation Resource （TIMER） database was used to analyze the relationship between core genes and tumor immune infiltration. Gene Ontology （GO） functional and Kyoto Encyclopedia of Genes and Genomes （KEGG） signaling pathway enrichment analysis were performed to analyze the positively correlated genes of core genes obtained from the LinkedOmics database. Results A total of 340 DEGs were obtained from the intersection of DEGs from the four GEO datasets， including 127 upregulated genes and 213 downregulated genes. After screening with the STRING database and Cytoscape software， the key core genes with the highest scores were secreted phosphoprotein 1 （SPP1） and transforming growth factor beta 1 （TGFB1）. The GEPIA database analysis results showed that compared with esophageal normal tissue， the expression levels of SPP1 and TGFB1 mRNA in cancer tissue were significantly increased （P<0.01）. The 1-year， 3-year， and 5-year overall survival of the EAC patients in SPP1 low expression group was higher than those in SPP1 high expression group （HR=10.1， P<0.05； HR=3.09， P<0.05； HR=2.32， P<0.05）， and the 5-year overall survival of the EAC patients in TGFB1 low expression group was higher than that in TGFB1 high expression group （HR=2.36， P<0.05）. The UALCAN database analysis results showed that compared with esophageal normal tissue， the expression levels of SPP1 and TGFB1 mRNA in cancer tissue of the EAC patients with stage Ⅱ-Ⅲ and N1-N2 lymph node metastasis were significantly increased （P<0.01）.The TIMER analysis results showed that the expression levels of SPP1 and TGFB1 mRNA in cancer tissue of the EAC patients were positively correlated with the infiltration of macrophages （r=0.353，P<0.01； r=0.187，P<0.05） and dendritic cells （r=0.236，P<0.01； r=0.221，P<0.01）. The GO and KEGG pathway enrichment analysis results showed that SPP1， TGFB1， and their top 50 positively correlated genes mainly participated in the biological processes such as cell migration， cell activity， and angiogenesis， and signaling pathways such as tumor proteoglycans， extracellular matrix （ECM）-receptor interaction， and phosphatidylinositol 3-kinase （PI3K）/protein kinase B （AKT）. Conclusion SPP1 and TGFB1 are closely associated with clinical staging， lymph node metastasis， and overall survival of the EAC patients. High expressions of SPP1 and TGFB1 may lead to the infiltration of the macrophages and dendritic cells， and change the tumor microenvironment. SPP1 and TGFB1 may become new targets for the diagnosis and treatment of EAC.

Key words: Esophageal adenocarcinoma, Bioinformatics, Survival analysis, Prognosis, Tumor immune infiltration