Identification Of Core Gene In Esophageal Squamous Cell Carcinoma By Weighted Gene Co-expression Network Analysis And Its Clinical Significance

Objectives The incidence and mortality of esophageal cancer in China is the highest in the world.The most important histological type is esophageal squamous cell carcinoma(ESCC),and its overall 5-year survival rate is less than25%.Lack of early diagnosis of it and does not have targeted therapy drugs are the reasons for the low survival rate of patients with ESCC.With the advent of high-throughput data,our research on malignant tumors at the molecular level has reached a new height.Weighted gene co-expression network analysis(WGCNA)is a method that can identify important factors in scale-free networks.Compared with other methods such as differential expression method,WGCNA can not only make full use of the information of all genes,but also decrease the dimension of thousands of genes,eliminating the problem of multiple hypothesis testing and correction.WGCNA is particularly suitable for complex multi-sample transcriptome data processing.In this study,WGCNA was used to analyze high-throughput data in order to find core gene for ESCC.We hope the core gene has the ability to be a biomarker for ESCC.Methods(1)WGCNA analysis was performed on the microarray data of GSE20347 and GSE29001 by R language.Intersection gene sets were obtained by Venn method.To construct protein-protein interaction networks,STRING on-line database was used.Core gene was found by the MCODE plug-in in Cytoscape software,and gene set enrichment Analysis(GSEA)was performed to obtain potential pathways.(2)The expression levels of core gene in ESCC tissue and normal esophageal tissue were verified at the transcriptional and protein levels by qRT-PCR and Western Blot,and the differences between the two groups were compared by student’s T test.(3)The pathological slides and clinicopathological feature of patients with ESCC were collected,and the relationship between the expression levels of core gene and the clinicopathological factors was analyzed by immunohistochemical techniques.Kaplan-Meier curve and log-rank test were used for survival analysis.Cox proportional hazards model was performed to analyze the hazard ratio(HR)for overall survival(OS)and disease-free survival(DFS).To construct a prediction model for survival probability of patients with ESCC by using nomogram.(4)Combining the microarray data of GEO,the cancer genome atlas(TCGA)and our study,we conducted a meta-analysis of the expression of core genes,and used the STATA 15.1 software to construct the forest plot of standardized mean difference(SMD).The funnel plot,Egger’s and Begg’s plot were used to detect the publication bias of results,and the sensitivity analysis was performed to test the stability of the effect size estimates.We drew the receiver operating characteristic(ROC)curve of each dataset and calculated the integrated area under the curve(AUC),sensitivity(SEN),specificity(SPE),etc.To explore the diagnostic performance of core genes,Fagan’s nomogram was constructed to get the predicted probability of core genes for ESCC.Finally,in order to obtain the comprehensive prognostic value of core genes for ESCC,we performed a meta-analysis by searching Chinese and English literature databases such as Pub Med,Google Scholar,CNKI,Wan fang,and VIP.In this study,if P<0.05,the difference between groups was considered to be statistically significant.Results(1)A total of 661 intersecting genes relevant to ESCC were identified after WGCNA,of which the core gene was COL11A1.The single-gene GSEA results suggested that ECM receptor interaction was a potential pathway of COL11A1 in ESCC.(2)COL11A1 expression level in ESCC tissues was higher than that in normal tissues at both the transcription and protein levels.(3)IHC results showed that the expression of COL11A1 ESCC tissues was significant with low Grade(P = 0.002),high N stage(P < 0.001),and high TNM stage(P < 0.001).COL11A1 was an independent prognostic risk factor for patients,with HR = 1.863 [1.124-3.086],P = 0.016 for OS,HR = 1.691[1.041-2.749],P = 0.034 for DFS.A prediction model for survival probability of patients with ESCC by using nomogram,with C-index=0.721 [0.668-0.774].(4)We collected 670 ESCC samples and 482 non-cancerous samples.The comprehensive results showed that COL11A1 in ESCC was higher than that in non-cancerous tissues(SMD = 1.62 [1.26-1.99]),and the expression of COL11A1 could distinguish two of them(overall AUC = 0.95 [0.92-0.96],SEN= 0.93 [0.85-0.97],SPE = 0.92 [0.89-0.94]).When the pre-test probability was20%,Fagan’s nomogram suggested that COL11A1 expression level can increase the positive probability of ESCC to 74%,and decrease the false negative probability to 3%.A total of 600 patients with ESCC were included in 4 studies for prognostic meta-analysis,and it was confirmed that COL11A1 was a prognostic risk factor for OS,with comprehensive HR = 1.82 [1.33-2.31],P <0.001.Conclusions COL11A1 was the core gene of ESCC by using WGCNA,and its expression level in ESCC is up-regulated compared with non-cancerous esophageal tissue.High COL11A1 expression is a prognostic risk factor for patients with ESCC.Moreover,COL11A1 expression level could well separate ESCC from non-cancerous tissues.COL11A1 has the potential to become a biomarker and therapeutic target for ESCC.