Yuan Shushu, Cong Zhirong, Ji Jiali, Zhu Li, Jiang Qi, Zhou Ying, Shen Qian, Damiani Daniela, Xu Xiaohong, Li Bingzong
Department of Hematology, The Second Affiliated Hospital of Soochow University, Suzhou, China.
Department of Oncology, Affiliated Tumor Hospital of Nantong University, Nantong, China.
Ann Transl Med. 2022 Aug;10(16):902. doi: 10.21037/atm-22-3858.
Most previous studies have focused on the intrinsic carcinogenic pathways of tumors; however, little is known about the potential role of N6-methyladenosine (m6A) methylation in the tumor immune microenvironment (TIME). To better diagnose and treat acute myeloid leukemia (AML), we sought to examine the correlation between m6A regulatory factors and immune infiltration in cases of AML. At the same time, a prognostic model was constructed to predict the survival of AML.
We extracted data from The Cancer Genome Atlas (TCGA) database, including ribonucleic acid sequencing (RNA-seq) transcriptome data and data on the corresponding clinical characteristics of AML patients. We identified two m6A modification patterns with distinct clinical outcomes and found a significant relationship between them. Simultaneous discovery of distinct m6A clusters associated with the tumor immune microenvironment [immune cell types and Estimation of STromal and Immune cells in MAlignant Tumor tissues using Expression data (ESTIMATE) algorithm] are closely related. Next, we implemented Lasso (Least Absolute Shrinkage and Selection Operator) Cox regression to build a predictive model in the 2-m6A regulator TCGA dataset to further explore m6A prognostic features in AML, and perform correlation validation.
We identified 2 molecular subtypes (Clusters 1 and 2) by the consistent clustering of significant m6A regulators in AML. Cluster 2 was associated with a higher immune score and obvious immune cell infiltration, and thus patients in Cluster 2 had a poorer prognosis than those in Cluster 1 (P<0.05). Additionally, the 2 m6A-related signatures representing the independent prognostic factors in AML were screened to construct a prognostic risk-score model. We found that patients with low-risk scores had higher immune scores than those with high-risk scores (P<0.05).
Our research confirmed that m6A methylation plays an important role in AML. Further provide new directions for the prognosis and treatment of AML.
以往大多数研究都集中在肿瘤的内在致癌途径；然而，关于N6-甲基腺苷（m6A）甲基化在肿瘤免疫微环境（TIME）中的潜在作用知之甚少。为了更好地诊断和治疗急性髓系白血病（AML），我们试图研究m6A调控因子与AML患者免疫浸润之间的相关性。同时，构建了一个预后模型来预测AML患者的生存情况。
我们从癌症基因组图谱（TCGA）数据库中提取数据，包括核糖核酸测序（RNA-seq）转录组数据以及AML患者相应的临床特征数据。我们识别出两种具有不同临床结局的m6A修饰模式，并发现它们之间存在显著关联。同时发现与肿瘤免疫微环境[免疫细胞类型以及使用表达数据估计恶性肿瘤组织中的基质和免疫细胞（ESTIMATE）算法]相关的不同m6A簇密切相关。接下来，我们实施套索（最小绝对收缩和选择算子）Cox回归，在2-m6A调控因子TCGA数据集中构建预测模型，以进一步探索AML中的m6A预后特征，并进行相关性验证。
我们通过对AML中显著的m6A调控因子进行一致性聚类，识别出2种分子亚型（簇1和簇2）。簇2与更高的免疫评分和明显的免疫细胞浸润相关，因此簇2中的患者预后比簇1中的患者更差（P<0.05）。此外，筛选出代表AML中独立预后因素的2个与m6A相关的特征，构建了预后风险评分模型。我们发现低风险评分的患者比高风险评分的患者具有更高的免疫评分（P<0.05）。
我们的研究证实m6A甲基化在AML中起重要作用。进一步为AML的预后和治疗提供了新方向。
