DataSheet1_Comprehensive Analysis of N6-Methyladenosine-Related Long Noncoding RNA Prognosis of Acute Myeloid Leukemia and Immune Cell Infiltration.ZIP

N6-Methyladenosine-related long noncoding RNAs play an essential role in many cancers’ development. However, the relationship between m6A-related lncRNAs and acute myelogenous leukemia (AML) prognosis remains unclear. We systematically analyzed the association of m6A-related lncRNAs with the prognosis and tumor immune microenvironment (TME) features using the therapeutically applicable research to generate effective treatment (TARGET) database. We screened 315 lncRNAs associated with AML prognosis and identified nine key lncRNAs associated with m6A by the LASSO Cox analysis. A model was established based on these nine lncRNAs and the predictive power was explored in The Cancer Genome Atlas (TCGA) database. The areas under the ROC curve of TARGET and TCGA databases for ROC at 1, 3, and 5 years are 0.701, 0.704, and 0.696, and 0.587, 0.639, and 0.685, respectively. The nomogram and decision curve analysis (DCA) showed that the risk score was more accurate than other clinical indicators in evaluating patients’ prognoses. The clusters with a better prognosis enrich the AML pathways and immune-related pathways. We also found a close correlation between prognostic m6A-related lncRNAs and tumor immune cell infiltration. LAG3 expression at the immune checkpoint was lower in the worse prognostic cluster. In conclusion, m6A-related lncRNAs partly affected AML prognosis by remodeling the TME and affecting the anticarcinogenic ability of immune checkpoints, especially LAG3 inhibitors. The prognostic model constructed with nine key m6A-related lncRNAs can provide a method to assess the prognosis of AML patients in both adults and children.

N6-甲基腺苷（N6-Methyladenosine，简称m6A）相关长链非编码RNA（long noncoding RNAs，简称lncRNAs）在多种癌症的发生发展中发挥关键作用。然而，m6A相关lncRNAs与急性髓系白血病（acute myelogenous leukemia，简称AML）预后之间的关联仍不明确。本研究依托治疗性应用研究以生成有效治疗（TARGET）数据库，系统分析了m6A相关lncRNAs与患者预后及肿瘤免疫微环境（tumor immune microenvironment，简称TME）特征的关联。研究团队筛选出315个与AML预后相关的lncRNAs，并通过LASSO Cox分析（LASSO Cox analysis）鉴定出9个与m6A相关的关键lncRNAs。基于上述9个lncRNAs构建预后模型，并在癌症基因组图谱（The Cancer Genome Atlas，简称TCGA）数据库中验证了模型的预测效能。TARGET与TCGA数据库的1年、3年、5年ROC曲线（Receiver Operating Characteristic curve）下面积分别为0.701、0.704、0.696与0.587、0.639、0.685。列线图（nomogram）与决策曲线分析（decision curve analysis，简称DCA）结果显示，风险评分在评估患者预后方面的准确性显著优于其他临床指标。预后较好的样本聚类组富集了AML相关通路与免疫相关通路。本研究同时发现，预后相关m6A相关lncRNAs与肿瘤免疫细胞浸润程度密切相关。免疫检查点（immune checkpoint）LAG3（淋巴细胞活化基因3，LAG3）的表达水平在预后较差的样本聚类组中显著更低。综上，m6A相关lncRNAs可通过重塑肿瘤免疫微环境、调控免疫检查点的抗肿瘤活性（尤其是LAG3抑制剂），部分影响AML的预后。本研究基于9个关键m6A相关lncRNAs构建的预后模型，可为成人与儿童AML患者的预后评估提供可行的参考方案。