Identification Of Immune-related Genes In Breast Cancer And Prognostic Risk Signature Establishment

Object: In this study,bioinformatic methods were used for data mining.We aim to identify the immune-related genes(IRGs)in breast cancer(BRCA)and develop an IRGs signature to predict the prognosis of BRCA patients.The expression of the core gene MMP9 was verified with clinical samples.Method: The mRNA expression profiles and clinical data of BRCA patients were downloaded from TCGA databases,and then the differentially expressed IRGs were screened.We also used the least absolute shrinkage and selection operator(LASSO)Cox regression model to construct a multigene signature with IRGs of BRCA patients,and then the formula for risk score was constructed.Based on the median of risk score,we divided patients into low-and high-risk groups.We also evaluated the efficacy of the IRGs prognostic signature with the GEO validation cohort(GSE42568 and GSE16446).Moreover,we evaluated the predictive capability of the IRGs prognostic signature using the Kaplan-Meier survival curve and ROC curve analysis.Then,we explored the differences in tumor immune cell infiltration and tumor mutation burden(TMB)between the different risk groups.The STRING online tool was used for detecting the protein-protein interaction(PPI)network between the IRGs in this signature,and we used Cytoscape software to screen the top 10 hub genes.The intersection between the IRGs in prognostic signature and the top 10 hub genes in the PPI network was selected as the core genes.The expression of core genes in BRCA was performed using the GEPIA and GSCALite databases.The survival curve of BRCA for core genes was performed by Kaplan-Meier plotter online platform.The correlation of core genes and six immune-infiltrating cells was evaluated based on the TIMER database.Gene Ontology(GO)term enrichment and Kyoto Encyclopedia of Genes and Genomes(KEGG)pathway analyses were performed using DAVID tools.Clinical BRCA tissues were collected and analyzed by RT-PCR and immunohistochemical(IHC)assays.MDSCs and the expression of core genes on MDSCs were quantified by flow cytometry.Result: In this study,426 differentially expressed IRGs were screened based on the expression data of BRCA in the TCGA database.By conducting LASSO Cox regression analysis,a 15-IRGs prognostic signature for predicting survival in patients with BRCA was developed.ROC analyses and the nomogram suggested that the signature showed an excellent predictive capability for BRCA.The prognostic signature was further validated in the GSE42568 and GSE16446 cohorts.According to the signature,patients with BRCA were divided into two risk groups.Functional annotation and the pathway enrichment analysis showed that the high-risk group enriched many tumor features and immune-related pathways.Moreover,the immune cell types of the tumor immune microenvironment and the TMB value of patients in the high-risk group were significantly different from those in the low-risk group.The multivariate Cox regression revealed that the IRGs prognostic signature is an independent prognostic factor of BRCA.Finally,verified by clinical samples,the expression of core gene MMP9 in the tumor and adjacent normal tissues was consistent with the bioinformatic results.The expression level of MMP9 in BRCA tissues and serum was significantly up-regulated compared with the healthy control group.Immunohistochemical analysis revealed that the expression level of MMP9 was related to the histologic grade,clinical stage,PAM50 subtype,and tumor size in BRCA.Both M-MDSC and PMN-MDSC expressed high levels of MMP9 in BRCA,confirmed by flow cytometry.Conclusion: We identified a novel IRGs prognostic signature that could predict BRCA patients’ prognoses.Our work elucidated that the IRGs may serve as an indispensable player in the complexity and diversity of tumor progression and facilitate personalized treatment in BRCA.