Zhu Zongsi, Li Bing, Li Ping
Department of Hematology, Tongji Hospital, Tongji University School of Medicine, Shanghai, China.
J Cancer. 2025 Jul 11;16(10):3202-3215. doi: 10.7150/jca.113136. eCollection 2025.
Acute myeloid leukemia (AML) remains an incurable hematological malignancy characterized by significant treatment resistance. Necroptosis, a newly recognized form of programmed cell death, has been implicated in tumor development and progression; however, its specific role in AML is not yet fully understood. We integrated transcriptomic and clinical data from TCGA and GEO database (GSE37642) to identify differentially expressed necroptosis-related genes (NRGs) between AML and normal samples from GTEx. Consensus clustering was performed to classify AML samples based on NRG expression profiles. Kaplan-Meier survival analysis, GSVA, and ssGSEA were employed to assess survival differences, biological functions, and immune cell infiltration between clusters. Differentially expressed genes (DEGs) identified between NRG clusters underwent LASSO and Cox proportional hazards regression analyses to develop a prognostic risk model. A nomogram integrating age and risk score was constructed and validated in independent cohorts (GSE12417). A nomogram integrating age and risk score was developed. CNV, TMB, immune profiles, and drug sensitivity were also analyzed. Importantly, qRT-PCR was performed using and normal PBMCs to experimentally validate the expression levels of three key NRGs identified by the model (STAT5B, MAP3K7, and BCL2L11). Two distinct NRG clusters were identified. Cluster B exhibited poorer prognosis, higher immune cell infiltration, and enriched signaling pathways, including TGF-β, JAK-STAT, ERBB, MAPK, and VEGF. The developed prognostic nomogram demonstrated robust predictive capability (integrated AUC = 0.645). The high-risk group displayed positive correlations with naive B cells, eosinophils, activated/resting memory CD4 T cells, and CD8 T cells, while negatively associated with memory B cells, resting mast cells, and follicular helper T cells. Drug sensitivity analysis indicated increased sensitivity to Bcl-2 inhibitors, checkpoint kinase inhibitors, and MAPK-MEK pathway inhibitors in the high-risk group. qRT-PCR results confirmed that STAT5B was significantly upregulated, while MAP3K7 and BCL2L11 were significantly downregulated in AML cells compared to normal PBMCs, consistent with bioinformatic predictions. Our study elucidates a significant association between suppressed necroptosis and adverse prognosis in AML. We highlight the role of NRGs in modulating the immune microenvironment of AML and identify potential therapeutic targets and drugs, providing valuable insights for improving clinical outcomes in AML patients.
急性髓系白血病（AML）仍然是一种无法治愈的血液系统恶性肿瘤，具有显著的治疗抵抗性。坏死性凋亡是一种新认识的程序性细胞死亡形式，已被证明与肿瘤的发生和进展有关；然而，其在AML中的具体作用尚未完全明确。我们整合了来自TCGA和GEO数据库（GSE37642）的转录组和临床数据，以鉴定AML与来自GTEx的正常样本之间差异表达的坏死性凋亡相关基因（NRG）。进行一致性聚类以根据NRG表达谱对AML样本进行分类。采用Kaplan-Meier生存分析、GSVA和ssGSEA来评估各聚类之间的生存差异、生物学功能和免疫细胞浸润情况。对NRG聚类之间鉴定出的差异表达基因（DEG）进行LASSO和Cox比例风险回归分析，以建立一个预后风险模型。构建了一个整合年龄和风险评分的列线图，并在独立队列（GSE12417）中进行了验证。开发了一个整合年龄和风险评分的列线图。还分析了拷贝数变异（CNV）、肿瘤突变负荷（TMB）、免疫谱和药物敏感性。重要的是，使用AML细胞和正常外周血单核细胞（PBMC）进行qRT-PCR，以实验验证该模型鉴定出的三个关键NRG（STAT5B、MAP3K7和BCL2L11）的表达水平。鉴定出两个不同的NRG聚类。聚类B表现出较差的预后、更高的免疫细胞浸润以及富集的信号通路，包括TGF-β、JAK-STAT、ERBB、MAPK和VEGF。所开发的预后列线图显示出强大的预测能力（综合AUC = 0.645）。高危组与幼稚B细胞、嗜酸性粒细胞、活化/静息记忆CD4 T细胞和CD8 T细胞呈正相关，而与记忆B细胞、静息肥大细胞和滤泡辅助性T细胞呈负相关。药物敏感性分析表明高危组对Bcl-2抑制剂、检查点激酶抑制剂和MAPK-MEK通路抑制剂的敏感性增加。qRT-PCR结果证实，与正常PBMC相比，AML细胞中STAT5B显著上调，而MAP3K7和BCL2L11显著下调，这与生物信息学预测一致。我们的研究阐明了AML中坏死性凋亡受抑制与不良预后之间的显著关联。我们强调了NRG在调节AML免疫微环境中的作用，并鉴定出潜在的治疗靶点和药物，为改善AML患者的临床结局提供了有价值的见解。
