Background We investigated the miRNA-m6A related gene network and identified a miRNA-based prognostic signature in patients with esophageal cancer using integrated genomic analysis. Methods We obtained expression data for m6A-related genes and miRNAs from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets. Survival analysis was conducted to identify potential prognostic biomarkers. LASSO Cox regression was performed to construct the overall survival (OS) associated prediction signature. We used the Kaplan–Meier (K–M) curve and receiver operating characteristic (ROC) curves to explore the signature’s efficiency and accuracy. Interactions between the m6A-related genes and miRNAs were identified in starBase3.0 and used to construct the miRNA-m6A related gene network. Results We found that HNRNPC, YTHDF, ZC3H13, YTHDC2, and METTL14 were dysregulated in esophageal cancer tissues. Multivariate Cox regression analysis revealed that HNRNPC may be an independent risk factor for OS. Five hundred twenty-two potential upstream miRNAs were obtained from starBase3.0. Four miRNAs (miR-186, miR-320c, miR-320d, and miR-320b) were used to construct a prognostic signature, which could serve as a prognostic predictor independent from routine clinicopathological features. Finally, we constructed a key miRNA-m6A related gene network and used one m6A-related gene and four miRNAs associated with the prognosis. The results of our bioinformatics analysis were successfully validated in the human esophageal carcinoma cell lines KYSE30 and TE-1. Conclusion Our study identified a 4-miRNA prognostic signature and established a key miRNA-m6A related gene network. These tools may reliably assist with esophageal cancer patient prognosis.

中文翻译：

---

食管癌患者m6A相关基因miRNA的网络分析

背景 我们研究了 miRNA-m6A 相关基因网络，并使用整合基因组分析确定了食管癌患者基于 miRNA 的预后特征。方法 我们从癌症基因组图谱 (TCGA) 和基因表达综合 (GEO) 数据集中获得 m6A 相关基因和 miRNA 的表达数据。进行生存分析以确定潜在的预后生物标志物。进行 LASSO Cox 回归以构建与总生存 (OS) 相关的预测特征。我们使用 Kaplan-Meier (K-M) 曲线和接收者操作特征 (ROC) 曲线来探索签名的效率和准确性。在starBase3.0中鉴定了m6A相关基因和miRNA之间的相互作用，并用于构建miRNA-m6A相关基因网络。结果 我们发现 HNRNPC、YTHDF、ZC3H13、YTHDC2、和 METTL14 在食管癌组织中失调。多变量 Cox 回归分析显示，HNRNPC 可能是 OS 的独立危险因素。从starBase3.0中获得了522个潜在的上游miRNA。四种 miRNA（miR-186、miR-320c、miR-320d 和 miR-320b）用于构建预后特征，可作为独立于常规临床病理学特征的预后预测因子。最后，我们构建了一个关键的 miRNA-m6A 相关基因网络，并使用了一个 m6A 相关基因和四个与预后相关的 miRNA。我们的生物信息学分析结果在人食管癌细胞系 KYSE30 和 TE-1 中成功验证。结论 我们的研究确定了 4-miRNA 预后特征并建立了关键的 miRNA-m6A 相关基因网络。