Table_3_Identification and Validation of m6A-Related lncRNA Signature as Potential Predictive Biomarkers in Breast Cancer.doc

The metastasis and poor prognosis are still regarded as the main challenge in the clinical treatment of breast cancer (BC). Both N6-methyladenosine (m6A) modification and lncRNAs play vital roles in the carcinogenesis and evolvement of BC. Considering the unknown association of m6A and lncRNAs in BC, this study therefore aims to discern m6A-related lncRNAs and explore their prognostic value in BC patients. Firstly, a total of 6 m6A-related lncRNAs were screened from TCGA database and accordingly constructed a prognostic-predicting model. The BC patients were then divided into high-risk and low-risk groups dependent on the median cutoff of risk score based on this model. Then, the predictive value of this model was validated by the analyses of cox regression, Kaplan-Meier curve, ROC curve, and the biological differences in the two groups were validated by PCA, KEGG, GSEA, immune status as well as in vitro assay. Finally, we accordingly constructed a risk prognostic model based on the 6 identified m6A-related lncRNAs, including Z68871.1, AL122010.1, OTUD6B-AS1, AC090948.3, AL138724.1, EGOT. Interestingly, the BC patients were divided into the low-risk and high-risk groups with different prognoses according to the risk score. Notably, the risk score of the model was an excellent independent prognostic factor. In the clinical sample validation, m6A regulatory proteins were differentially expressed in patients with different risks, and the markers of tumor-associated macrophages and m6A regulators were co-localized in high-risk BC tissues. This well-validated risk assessment tool based on the repertoire of these m6A-related genes and m6A-related lncRNAs, is of highly prognosis-predicting ability, and might provide a supplemental screening method for precisely judging BC prognosis.

转移与不良预后仍是乳腺癌（BC）临床治疗面临的主要挑战。N6-甲基腺嘌呤（m6A）修饰与长链非编码RNA（lncRNAs）均在乳腺癌的发生发展过程中发挥关键作用。鉴于目前乳腺癌中m6A修饰与lncRNAs的关联尚不明确，本研究旨在筛选与m6A相关的lncRNAs，并探讨其在乳腺癌患者中的预后价值。首先，本研究从癌症基因组图谱（The Cancer Genome Atlas, TCGA）数据库中共筛选出6个m6A相关lncRNAs，并据此构建了预后预测模型。基于该模型的风险评分中位数分界值，将乳腺癌患者分为高风险组与低风险组。随后，通过Cox回归分析、Kaplan-Meier曲线、受试者工作特征（Receiver Operating Characteristic, ROC）曲线验证了该模型的预测价值，并通过主成分分析（Principal Component Analysis, PCA）、京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes, KEGG）富集分析、基因集富集分析（Gene Set Enrichment Analysis, GSEA）、免疫状态分析以及体外实验验证了两组间的生物学差异。最终，本研究成功构建了基于6个已鉴定m6A相关lncRNAs的风险预后模型，所涉及的lncRNAs包括Z68871.1、AL122010.1、OTUD6B-AS1、AC090948.3、AL138724.1及EGOT。有趣的是，根据风险评分可将乳腺癌患者分为预后存在显著差异的低风险组与高风险组。值得注意的是，该模型的风险评分是一项优秀的独立预后因素。在临床样本验证中，不同风险组患者的m6A调控蛋白表达存在差异，且肿瘤相关巨噬细胞标志物与m6A调控因子在高风险乳腺癌组织中存在共定位现象。本研究验证的这款基于这套m6A相关基因与m6A相关lncRNAs的风险评估工具，具备优异的预后预测能力，可为乳腺癌预后的精准判断提供补充筛查手段。