| ObjectiveBreast Cancer(BC)is the leading cause of cancer-related death in females.Early diagnosis,effective prognostic prediction,and dynamic monitoring of therapeutic effects play a vital role in improving the survival outcome of BC patients.DNA methylation regulates the pathological and physiological processes of cells.Aberrant DNA methylation is involved in various processes of carcinogenesis and progression,including regulation of energy metabolism,angiogenesis,maintenance of proliferative signals,epithelial-mesenchymal transformation,invasion,and metastasis.But its roles and potential mechanisms in the development,prognosis,and treatment of BC remain unclear.This study explored a reliable DNA methylation biomarker related to the diagnosis,prognosis,and treatment of BC and evaluated its value in clinical application.MethodsWhole-genome methylation datasets integrated from the Cancer Genome Atlas and Gene Expression Omnibus database were profiled.A three-stage selection procedure(discovery,training,and validation)was utilized to screen out the prominent biomarkers and establish a robust prognostic signature from more than 300,000 Cytosine-phosphate-Guanine(CpG)sites.Subsequently,we integrated the 10-CpG-based signature and clinicopathological risk factors to build a composite nomogram and performed a follow-up evaluation.Moreover,we explored the differences in the immune microenvironment among different BC risk subgroups by gene set enrichment analyses and immunogenomics analyses.Multi-omics data analyses were performed to decipher specific genomic alterations in low-and high-risk patients.Additionally,we also analyzed the role of the risk score in predicting response to several treatment options.ResultsIn this study,from the perspective of epigenetic modification of CpG sites,a 10-CpGbased prognostic signature was established based on bioinformatics analyses.It was sufficiently validated that this signature could independently predict the prognosis of BC patients.Furthermore,we integrated the identified independent risk factors,consisting of risk score,age,ER status,and positive lymph node status to construct a novel methylation-clinicopathological nomogram to assist in the individualized prognostic prediction in BC patients.Gene set enrichment analyses showed that immunoactivity-related pathways were significantly enriched in the low-risk group.Further immunogenomics analyses demonstrated that the derived risk score was negatively correlated with the characteristics of the inflamed tumor microenvironment,suggesting that the shaping of the intratumor non-inflamed microenvironment was an important factor for the poor prognoses of patients with high risk scores.In multi-omics data analyses,the identified high-risk patients were characterized by upregulation of TP53 mutation,tumor mutation burden,and copy number variation burden.In terms of therapeutic response,high-risk patients exhibited lower responses to immunotherapy and chemotherapy(such as gemcitabine,etoposide,and vinorelbine).Potential therapeutic targets of patients in the high-risk group were obtained through CMap(Connectivity Map)analysis.ConclusionOur work highlighted the complementary role of the 10-CpG-based signature in estimating the overall survival of BC patients and contributed to clarifying the potential role of CpG site modification in the occurrence and development of BC.In addition,this study also revealed the close association between the 10-CpG-based signature and the immune microenvironment of BC,shedding new light on improving the immunotherapy response of patients.Potential therapeutic targets were also discovered for BC. |
PDF Full Text Request