Construction Of Breast Cancer Prognostic Risk Model And Validation Of Nrf2 Gene Expression In The Model

Objective: To study the expression level,mutation frequency and possible signal pathway of breast cancer prognostic gene in breast cancer tissue through the establishment of breast cancer risk prognosis model,Identify and develop new biological markers for breast cancer to provide a theoretical basis for accurate targeted therapy of breast cancer patients.Methods: 1.TCGA database was used to screen the differentially expressed genes,combined with single factor and multivariate Cox regression to further screen the prognosis related m RNAs of breast cancer to construct the risk prognosis model of breast cancer.The area AUC under ROC curve was used to evaluate the stability of the model.According to the median value of model score,patients were divided into high-risk and low-risk groups.Kaplan-Meier survival analysis(log rank test)was used to evaluate the prognosis difference of patients with different model scores and verify the clinical application efficiency of the model.Finally,1/3 of the original data was randomly selected as the verification set for internal verification of the model.Three independent risk genes SLC1A1,Nrf2 and PGR were screened by GSE3744,a breast cancer data set in GEO database.2.The real-time fluorescence quantitative PCR and immunohistochemistry were used to further analyze the expression of Nrf2 gene in benign breast epithelial cells and breast cancer cells.The corresponding Nrf2 gene mutation data were obtained from TCGA database,and the differences between wild group and mutant group were analyzed to study the mutations of Nrf2 gene.According to the expression of Nrf2 gene,the mutations of Nrf2 gene were analyzed in the subgroups of high and low expression population.The mutation frequency distribution regulated by CUL3-KEAP1-Nrf2 gene was analyzed,and the single gene KEGG enrichment analysis was conducted to clarify the possible mechanism of Nrf2 gene.Results: 1.A total of 514 transcriptome samples were downloaded from TCGA database,including 41 paracancerous tissues and 473 cancerous tissues.The results of differential analysis showed that there were 1266 m RNAs,395 of which were up-regulated and 871 down regulated.2.Single factor and multivariate Cox regression analysis showed that SLC1A1,Nrf2 and PGR were independent risk genes for prognosis.Based on the regression coefficient and expression of the above variables,the risk score of breast cancer patients' prognosis risk prediction model was constructed,risk score = 0.112 * SLC1A1 + 0.582 * Nrf2 + 0.166 * PGR.3.Kaplan-Meier analysis showed that there was a significant difference in OS curve between the two groups(P = 0.00057),and the survival rate of high-risk patients was significantly lower than that of low-risk patients.This result was also verified in the verification set(P = 0.0066).ROC curve analysis of SLC1A1,Nrf2 and pgr3 genes showed that AUC of training set model was 0.734 at 120 months OS,AUC = 0.689.4.The results of Kaplan-Meier analysis in GSE32744 data set showed that there was a significant difference(P = 0.0155)in the survival curve between the high-risk group and the low-risk group of Nrf2 gene,while there was no significant difference(P = 0.2581 and 0.0823)in the survival curve between the high-risk group of SLC1A1 and the high-risk group of PGR gene;Wilcox's rank sum test showed that the expression of Nrf2 gene in breast cancer was significantly higher than that in normal tissues(P < 0.01).5.The expression level of Nrf2 gene in MCF-7 breast cancer cells was significantly higher than that in MCF-10 a breast benign epithelial cells.Immunohistochemistry further confirmed that the expression level of Nrf2 in breast cancer tissues was significantly higher than that in adjacent tissues.6.The Waterfalls package analyzed the mutation data of the Nrf2 gene.The results showed that the expression of the wild group and the mutant group in the body was significantly different(P <0.05).Statistics found that in CUL3-KEAP1-Nrf2 gene axis,missense has the highest mutation frequency of Nrf2 gene?Mutation,the highest mutation frequency of CUL3 gene on the same gene axis is the same as Nrf2 gene,while the most mutation event of KEAP1 gene on the same gene axis is 3'UTR.The most frequently mutated genes were PIK3CA(38%),TP53(28%)and TTN(18%),TP53(41%),PIK3CA(31%)and TTN(20%).7.The Cluster Profiler package predicts the KEGG enrichment pathway of the Nrf2 gene,the results suggest that Nrf2 is enriched to KEGG?WNT?SIGNALING?PATHWAY,KEGG?PATHWAYS?IN?CANCER and KEGG?MAPK?SIGNALING?PATHWA Y and KEGG?MTOR?SIGNALING?PATHWAY and other signal pathways.Conclusion: 1.The breast cancer prognostic risk assessment model was successfully constructed,which provides certain guidance value for breast cancer patients' clinical risk stratification and early clinical intervention;2.Nrf2 gene may become a potential biological target for breast cancer treatment.