Despite that immune responses play important roles in acute myeloid leukemia (AML), immunotherapy is still not widely used in AML due to lack of an ideal target. Therefore, we identified key immune genes and cellular components in AML by an integrated bioinformatics analysis, trying to find potential targets for AML. Eighty-six differentially expressed immune genes (DEIGs) were identified from 751 differentially expressed genes (DEGs) between AML patients with fair prognosis and poor prognosis from the TCGA database. Among them, nine prognostic immune genes, including NCR2, NPDC1, KIR2DL4, KLC3, TWIST1, SNORD3B-1, NFATC4, XCR1, and LEFTY1, were identified by univariate Cox regression analysis. A multivariable prediction model was established based on prognostic immune genes. Kaplan–Meier survival curve analysis indicated that patients in the high-risk group had a shorter survival rate and higher mortality than those in the low-risk group (P < 0.001), indicating good effectiveness of the model. Furthermore, nuclear factors of activated T cells-4 (NFATC4) was recognized as the key immune gene identified by co-expression of differentially expressed transcription factors (DETFs) and prognostic immune genes. ATP-binding cassette transporters (ABC transporters) were the downstream KEGG pathway of NFATC4, identified by gene set variation analysis (GSVA) and gene set enrichment analysis (GSEA). To explore the immune responses NFATC4 was involved in, an immune gene set of T cell co-stimulation was identified by single-cell GSEA (ssGSEA) and Pearson correlation analysis, positively associated with NFATC4 in AML (R = 0.323, P < 0.001, positive). In order to find out the immune cell types affected by NFATC4, the CIBERSORT algorithm and Pearson correlation analysis were applied, and it was revealed that regulatory T cells (Tregs) have the highest correlation with NFATC4 (R = 0.526, P < 0.001, positive) in AML from 22 subsets of tumor-infiltrating immune cells. The results of this study were supported by multi-omics database validation. In all, our study indicated that NFATC4 was the key immune gene in AML poor prognosis through recruiting Tregs, suggesting that NFATC4 might serve as a new therapy target for AML.

尽管免疫反应在急性髓细胞性白血病（AML）中起着重要作用，但由于缺乏理想的靶标，免疫治疗仍未广泛用于AML中。因此，我们通过综合的生物信息学分析鉴定了AML中的关键免疫基因和细胞成分，试图寻找AML的潜在靶标。从TCGA数据库中预后良好的AML患者中的751个差异表达基因（DEG）中鉴定出86个差异表达免疫基因（DEIG）。其中，通过单变量Cox回归分析确定了9个预后免疫基因，包括NCR2，NPDC1，KIR2DL4，KLC3，TWIST1，SNORD3B-1，NFATC4，XCR1和LEFTY1。基于预后免疫基因建立了多变量预测模型。P<0.001），表明该模型的有效性。此外，活化的T细胞4（NFATC4）的核因子被认为是通过差异表达转录因子（DETFs）和预后免疫基因的共表达确定的关键免疫基因。ATP结合盒转运蛋白（ABC转运蛋白）是NFATC4的下游KEGG途径，通过基因组变异分析（GSVA）和基因组富集分析（GSEA）进行鉴定。为了探索NFATC4参与的免疫反应，通过单细胞GSEA（ssGSEA）和Pearson相关分析鉴定了与T细胞共刺激的免疫基因组，其与AML中的NFATC4正相关（[R = 0.323， P<0.001，阳性）。为了找出受NFATC4影响的免疫细胞类型，应用CIBERSORT算法和Pearson相关分析，发现调节性T细胞（Tregs）与NFATC4相关性最高（[R = 0.526， P肿瘤浸润免疫细胞的22个子集中的AML中<0.001（阳性）。这项研究的结果得到了多组学数据库验证的支持。总体而言，我们的研究表明NFATC4是招募Tregs导致AML预后不良的关键免疫基因，这表明NFATC4可能成为AML的新治疗靶点。