Table_3_Identification and Validation of a Five-Gene Signature Associated With Overall Survival in Breast Cancer Patients.docx

BackgroundRecent years, the global prevalence of breast cancer (BC) was still high and the underlying molecular mechanisms remained largely unknown. The investigation of prognosis-related biomarkers had become an urgent demand. ResultsIn this study, gene expression profiles and clinical information of breast cancer patients were downloaded from the TCGA database. The differentially expressed genes (DEGs) were estimated by Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis. A risk score formula involving five novel prognostic associated biomarkers (EDN2, CLEC3B, SV2C, WT1, and MUC2) were then constructed by LASSO. The prognostic value of the risk model was further confirmed in the TCGA entire cohort and an independent external validation cohort. To explore the biological functions of the selected genes, in vitro assays were performed, indicating that these novel biomarkers could markedly influence breast cancer progression. ConclusionsWe established a predictive five-gene signature, which could be helpful for a personalized management in breast cancer patients.研究背景 近年来，全球乳腺癌（breast cancer, BC）的患病率仍居高不下，其潜在分子机制仍未完全阐明，针对预后相关生物标志物的研究已成为迫切需求。 研究结果 本研究从癌症基因组图谱（The Cancer Genome Atlas, TCGA）数据库中下载了乳腺癌患者的基因表达谱与临床信息。通过基因本体（Gene Ontology, GO）分析与京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes, KEGG）分析，鉴定得到差异表达基因（differentially expressed genes, DEGs）。随后通过最小绝对收缩和选择算子（Least Absolute Shrinkage and Selection Operator, LASSO）构建了包含5个新型预后相关生物标志物（EDN2、CLEC3B、SV2C、WT1及MUC2）的风险评分模型。随后在TCGA全队列与独立外部验证队列中，进一步验证了该风险模型的预后价值。为探究筛选得到的基因的生物学功能，本研究开展了体外实验，结果显示这些新型生物标志物可显著影响乳腺癌的进展进程。 研究结论 本研究构建了一个五基因预测特征模型，可为乳腺癌患者的个性化诊疗管理提供有益参考。