| Research background and purposeRenal cell carcinoma is second only to prostate cancer and bladder cancer in terms of incidence of genitourinary cancers.Clear cell renal cell carcinoma(cc RCC)is the most prevalent pathological type of renal cell carcinoma and accounts for more than 3/4 of renal malignancies,and its onset is insidious,with no specific manifestations in early stages,and most patients are already in the middle or late stages when detected.Although the detection rate of early-stage renal cancer has gradually increased,early surgical intervention and other means have effectively changed and improved the prognosis of these early-stage renal cancer patients.However,the mortality rate of kidney cancer also remains stable and high due to postoperative metastasis and recurrence.Epithelial-mesenchymal transition(EMT)is the initial step that mediates the acquisition of metastatic capacity by cells,and EMT is critical in starting and promoting tumor cell invasion and metastasis,according to a growing number of studies.Therefore,In this study,we propose to combine EMT-related genes with data from public databases to build a prognostic model for patients with cc RCC and try to elaborate the relationship between model genes and the prognosis of patients with cc RCC,which can provide theoretical support for subsequent experimental studies.MethodsThe transcriptomic expression dataset and matching clinical information files of cc RCC were downloaded from the public database TCGA,and the data were collated using R software and Perl program,then EMT-associated genes(EAGs)were extracted,and the differentially expressed EAGs associated with cc RCC were screened by differential analysis and weighted correlation network analysis(WGCNA).The samples were randomly divided 1:1 into Tain group and Test group while GO and KEGG functional enrichment analysis were performed on differentially expressed EAGs.The EMT-related risk prognostic model was constructed by univariate COX,LASSO regression,and multifactor COX regression and the efficacy of the model was validated by the area under the Kaplan-Meier(K-M)survival curve,receiver operating characteristic curve(ROC)curve,and independent prognosis analysis in the Train group,Test group,and total data set.Then we constructed a nomogram using independent prognostic factors to predict the overall survival of patients at 1,3,5,7,and 10 years.Finally,the correlation analysis between the tumor microenvironment score and prognostic model was built.To investigate immune cell infiltration in high and low-risk groups,we performed a single sample set enrichment analysis(ss GSEA),and the Tumor Immune Dysfunction and Exclusion(TIDE)score was used to predict the efficacy of immunotherapy in high and low-risk groups.ResultsA total of 195 differentially expressed cc RCC-associated EAGs were obtained after differential analysis and WGCNA analysis.Four genes(MICALL2,IRF6,RPL22L1,MTHFD2)were finally screened for participation in the model construction by univariate COX regression,LASSO regression,and multivariate COX regression analysis.The samples were divided into high-risk group and low-risk group based on mid-risk Score,and the K-M survival curves in Train,Test,and total data sets showed that the high-risk group curves were all lower than the low-risk group curves(P <0.001),and patients in the high-risk group had lower survival rate than those in the low-risk group.There were survival differences between high-risk and low-risk patients in different age groups,gender,grade,stage,T stage,and M0 stage,and all differences were statistically significant.In addition,on multiple ROC curves,the area under the curve exceeded 0.68.Independent prognostic analysis showed that risk Score,age,and stage could be used as independent prognostic factors for patients with cc RCC.The independent prognostic factors were included in the nomogram construction,and the C index of the nomogram was calculated to be 0.777.Meanwhile,the calibration curve basically matched with the diagonal line,so this nomogram possessed a quite accurate predictive ability.Correlation analysis of tumor microenvironment scores showed a positive correlation between the constructed model and immune scores.Analysis of ss GSEA immune infiltration showed a significant difference in the extent of multiple immune cell infiltration between the high and low-risk groups,and the TIDE results predicted a low efficacy of immunotherapy in the high-risk group.ConclusionThis study successfully constructed a prognostic prediction model for cc RCC consisting of four EMT-related genes based on the TCGA database with good predictive efficacy.The model risk Score was an independent prognostic factor,which could independently predict the prognosis of patients with cc RCC.The constructed nomogram can predict the survival rate of patients with cc RCC quite accurately and provide a prognostic reference for patients.In addition,we found significant differences in infiltration of multiple immune cells between high and low-risk groups,low efficacy of immunotherapy and high likelihood of immune escape in high-risk group. |
