Table7_Modification Patterns of DNA Methylation-Related lncRNAs Regulating Genomic Instability for Improving the Clinical Outcomes and Tumour Microenvironment Characterisation of Lower-Grade Gliomas.XLSX

Background: DNA methylation is an important epigenetic modification that affects genomic instability and regulates gene expression. Long non-coding RNAs (lncRNAs) modulate gene expression by interacting with chromosomal modifications or remodelling factors. It is urgently needed to evaluate the effects of DNA methylation-related lncRNAs (DMlncRNAs) on genome instability and further investigate the mechanism of action of DMlncRNAs in mediating the progression of lower-grade gliomas (LGGs) and their impact on the immune microenvironment.
Methods: LGG transcriptome data, somatic mutation profiles and clinical features analysed in the present study were obtained from the CGGA, GEO and TCGA databases. Univariate, multivariate Cox and Lasso regression analyses were performed to establish a DMlncRNA signature. The KEGG and GO analyses were performed to screen for pathways and biological functions associated with key genes. The ESTIMATE and CIBERSORT algorithms were used to determine the level of immune cells in LGGs and the immune microenvironment fraction. In addition, DMlncRNAs were assessed using survival analysis, ROC curves, correlation analysis, external validation, independent prognostic analysis, clinical stratification analysis and qRT-PCR.
Results: We identified five DMlncRNAs with prognostic value for LGGs and established a prognostic signature using them. The Kaplan–Meier analysis revealed 10-years survival rate of 10.10% [95% confidence interval (CI): 3.27–31.40%] in high-risk patients and 57.28% (95% CI: 43.17–76.00%) in low-risk patients. The hazard ratio (HR) and 95% CI of risk scores were 1.013 and 1.009–1.017 (p < 0.001), respectively, based on the univariate Cox regression analysis and 1.009 and 1.004–1.013 (p < 0.001), respectively, based on the multivariate Cox regression analysis. Therefore, the five-lncRNAs were identified as independent prognostic markers for patients with LGGs. Furthermore, GO and KEGG analyses revealed that these lncRNAs are involved in the prognosis and tumorigenesis of LGGs by regulating cancer pathways and DNA methylation.
Conclusion: The findings of the study provide key information regarding the functions of lncRNAs in DNA methylation and reveal that DNA methylation can regulate tumour progression through modulation of the immune microenvironment and genomic instability. The identified prognostic lncRNAs have high potential for clinical grouping of patients with LGGs to ensure effective treatment and management.

背景：DNA甲基化（DNA methylation）是一类重要的表观遗传修饰，可影响基因组不稳定性并调控基因表达。长链非编码RNA（long non-coding RNAs，lncRNAs）通过与染色体修饰或重塑因子相互作用，实现对基因表达的调控。当前亟需评估DNA甲基化相关长链非编码RNA（DNA methylation-related lncRNAs，DMlncRNAs）对基因组不稳定性的影响，并进一步探究DMlncRNAs介导低级别胶质瘤（lower-grade gliomas，LGGs）进展的作用机制，以及其对免疫微环境的影响。
方法：本研究分析所用的LGG转录组数据、体细胞突变谱及临床特征，均取自CGGA、GEO与TCGA数据库。研究通过单因素、多因素Cox回归及Lasso回归分析，构建DMlncRNA预后特征模型。采用GO（Gene Ontology）与KEGG（Kyoto Encyclopedia of Genes and Genomes）分析，筛选与关键基因相关的通路及生物学功能。利用ESTIMATE和CIBERSORT算法，测定LGG样本中的免疫细胞水平与免疫微环境组分。此外，本研究还通过生存分析、ROC曲线、相关性分析、外部验证、独立预后分析、临床分层分析及qRT-PCR，对DMlncRNAs进行系统性评估。
结果：本研究共鉴定出5个具有LGG预后价值的DMlncRNAs，并以此构建了预后评分模型。Kaplan-Meier分析结果显示，高风险组患者的10年生存率为10.10% [95%置信区间（confidence interval，CI）：3.27%~31.40%]，低风险组患者的10年生存率为57.28%（95% CI：43.17%~76.00%）。基于单因素Cox回归分析，风险评分的风险比（hazard ratio，HR）及95% CI分别为1.013与1.009~1.017（p < 0.001）；经多因素Cox回归分析后，该结果分别为1.009与1.004~1.013（p < 0.001）。据此，这5个lncRNAs被确定为LGG患者的独立预后标志物。进一步的GO与KEGG分析表明，这些lncRNAs通过调控癌症通路及DNA甲基化，参与LGG的预后进程与肿瘤发生发展。
结论：本研究结果为lncRNAs在DNA甲基化中的功能提供了关键参考信息，并揭示DNA甲基化可通过调控免疫微环境与基因组不稳定性，进而影响肿瘤进展。本次鉴定出的预后相关lncRNAs，在LGG患者的临床分组以实现精准诊疗与规范化管理方面，具备较高的应用潜力。