目的 探索非凋亡调节性细胞死亡基因（NARCDs）在肺腺癌中的潜在生物学功能，并评估其在肺腺癌中的预后预测价值。方法 从癌症基因组图谱和基因表达图谱数据库中下载肺腺癌的转录组数据。我们使用R软件识别肺腺癌组织与正常组织之间表达差异的NARCDs。采用单变量Cox回归分析和最小绝对收缩和选择算子Cox回归构建NARCDs签名。通过Kaplan-Meier生存曲线、受试者工作特征曲线以及单变量和多变量Cox回归分析评估NARCDs签名的预测能力。采用基因集变异分析、Gene Ontology和Kyoto Encyclopedia of Genes and Genomes对NARCDs签名进行功能富集分析。此外，分析高和低NARCDs得分组之间的肿瘤突变负荷、肿瘤微环境、肿瘤免疫功能障碍和排斥得分以及化疗药物敏感性的差异。最后，使用STRING和Cytoscape软件构建NARCDs和免疫相关基因的蛋白质相互作用网络。结果我们鉴定了与预后相关的34个表达差异显著的NARCDs，其中16个基因（ATIC、AURKA、CA9、ITGB4、DDIT4、CDK5R1、CAV1、RRM2、GAPDH、SRXN1、NLRC4、GLS2、ADRB2、CX3CL1、GDF15和ADRA1A）被选择用于构建NARCDs签名。NARCDs签名被确定为独立的预后因子（P < 0.001）。功能分析显示高NARCDs得分组和低NARCDs得分组在错配修复，p53信号通路和细胞周期方面存在显著差异（均P < 0.05）。NARCDs低得分组具有较低的肿瘤突变负荷，较高的免疫得分，较高的肿瘤免疫功能障碍和排斥得分以及较低的药物敏感性（均P < 0.05）。此外，NARCDs和免疫相关基因的蛋白质相互作用网络中的10个中心基因（CXCL5、TLR4、JUN、IL6、CCL2、CXCL2、ILA、IFNG、IL33和GAPDH）均为免疫相关基因。结论基于16个基因的NARCDs预后模型是一个独立的预后因子，能够有效预测肺腺癌患者的临床预后，并为临床治疗提供帮助。

Objective To explore the potential biological functions of non-apoptotic regulatory cell death genes (NARCDs) in lung adenocarcinoma, and their prognostic prediction value in lung adenocarcinoma. Methods Transcriptome data of lung adenocarcinoma were downloaded from The Cancer Genome Atlas and Gene Expression Omnibus databases. We identified differentially expressed NARCDs between lung adenocarcinoma tissues and normal tissues with R software. NARCDs signature was constructed with univariate Cox regression analysis and the least absolute shrinkage and selection operator Cox regression. The prognostic predictive capacity of NARCDs signature was assessed by Kaplan-Meier survival curve, receiver operating characteristic curve, and univariate and multivariate Cox regression analyses. Functional enrichment of NARCDs signature was analyzed with gene set variation analysis, Gene Ontology, and Kyoto Encyclopedia of Genes and Genomes. In addition, differences of tumor mutational burden, tumor microenvironment, tumor immune dysfunction and exclusion score, and chemotherapeutic drug sensitivity were analyzed between the high and low NARCDs score groups. Finally, a protein-protein interaction network of NARCDs and immune-related genes was constructed by STRING and Cytoscape software. Results We identified 34 differentially expressed NARCDs associated with the prognosis, 16 (ATIC, AURKA, CA9, ITGB4, DDIT4, CDK5R1, CAV1, RRM2, GAPDH, SRXN1, NLRC4, GLS2, ADRB2, CX3CL1, GDF15, and ADRA1A) of which were selected to construct a NARCDs signature. NARCDs signature was identified as an independent prognostic factor (P < 0.001). Functional analysis showed that there were significant differences in mismatch repair, p53 signaling pathway, and cell cycle between the high NARCDs score group and low NARCDs score group (all P < 0.05). The NARCDs low score group had lower tumor mutational burden, higher immune score, higher Tumor Immune Dysfunction and Exclusion score, and lower drug sensitivity (all P < 0.05). In addition, the 10 hub genes (CXCL5, TLR4, JUN, IL6, CCL2, CXCL2, ILA, IFNG, IL33, and GAPDH) in protein-protein interaction network of NARCDs and immune-related genes were all immune-related genes. Conclusion The NARCDs prognostic model based on 16 genes is an independent prognostic factor, which can effectively predict the clinical prognosis of patients of lung adenocarcinoma and provide help for clinical treatment.