Weighted gene correlation network analysis identifies RSAD2, HERC5, and CCL8 as prognostic candidates for breast cancer.
Jianing Tang,Qian Yang,Qiuxia Cui,Dan Zhang,Deguang Kong,Xing Liao,Jiangbo Ren,Yan Gong,Gaosong Wu
DOI: https://doi.org/10.1002/jcp.28980
2020-01-01
Journal of Cellular Physiology
Abstract:As the most commonly diagnosed malignant tumor in female population, the prognosis of breast cancer is affected by complex gene interaction networks. In this research weighted gene co-expression network analysis (WGCNA) would be utilized to build a gene co-expression network to identify potential biomarkers for prediction the prognosis of patients with breast cancer. We downloaded GSE25065 from Gene Expression Omnibus database as the test set. GSE25055 and GSE42568 were utilized to validate findings in the research. Seven modules were established in the GSE25065 by utilizing average link hierarchical clustering. Three hub genes, RSAD2, HERC5, and CCL8 were screened out from the significant module (R-2 = 0.44), which were considerably interrelated to worse prognosis. Within test dataset GSE25065, RSAD2, and CCL8 were correlated with tumor stage, grade, and lymph node metastases, whereas HERC5 was correlated with lymph node metastases and tumor grade. In the validation dataset GSE25055 and RSAD2 expression was correlated with tumor grade, stage, and size, whereas HERC5 was related to tumor stage and tumor grade, and CCL8 was associated with tumor size and tumor grade. Multivariable survival analysis demonstrated that RSAD2, HERC5, and CCL8 were independent risk factors. In conclusion, the WGCNA analysis conducted in this study screened out novel prognostic biomarkers of breast cancer. Meanwhile, further in vivo and in vitro studies are required to make the clear molecular mechanisms.