Table3_Comprehensive landscape of junctional genes and their association with overall survival of patients with lung adenocarcinoma.DOCX

ObjectivesJunctional proteins are involved in tumorigenesis. Therefore, this study aimed to investigate the association between junctional genes and the prognosis of patients with lung adenocarcinoma (LUAD).
MethodsTranscriptome, mutation, and clinical data were retrieved from The Cancer Genome Atlas (TCGA). “Limma” was used to screen differentially expressed genes. Moreover, Kaplan–Meier survival analysis was used to identify junctional genes associated with LUAD prognosis. The junctional gene-related risk score (JGRS) was generated based on multivariate Cox regression analysis. An overall survival (OS) prediction model combining the JGRS and clinicopathological properties was proposed using a nomogram and further validated in the Gene Expression Omnibus (GEO) LUAD cohort.
ResultsTo our knowledge, this study is the first to demonstrate the correlation between the mRNA levels of 14 junctional genes (CDH15, CDH17, CDH24, CLDN6, CLDN12, CLDN18, CTNND2, DSG2, ITGA2, ITGA8, ITGA11, ITGAL, ITGB4, and PKP3) and clinical outcomes of patients with LUAD. The JGRS was generated based on these 14 genes, and a higher JGRS was associated with older age, higher stage levels, and lower immune scores. Thus, a prognostic prediction nomogram was proposed based on the JGRS. Internal and external validation showed the good performance of the prediction model. Mechanistically, JGRS was associated with cell proliferation and immune regulatory pathways. Mutational analysis revealed that more somatic mutations occurred in the high-JGRS group than in the low-JGRS group.
ConclusionThe association between junctional genes and OS in patients with LUAD demonstrated by our “TCGA filtrating and GEO validating” model revealed a new function of junctional genes.

研究目的：连接蛋白（junctional proteins）参与肿瘤发生过程。本研究旨在探讨连接基因与肺腺癌（lung adenocarcinoma, LUAD）患者预后之间的关联。
研究方法：从癌症基因组图谱（The Cancer Genome Atlas, TCGA）中获取转录组、突变及临床数据。采用"Limma"包筛选差异表达基因（differentially expressed genes）。通过Kaplan–Meier生存分析筛选与LUAD预后相关的连接基因。基于多变量Cox回归分析构建连接基因相关风险评分（junctional gene-related risk score, JGRS）。结合JGRS与临床病理特征，通过列线图（nomogram）构建总体生存期（overall survival, OS）预测模型，并在基因表达综合数据库（Gene Expression Omnibus, GEO）的LUAD队列中进行外部验证。
研究结果：据我们所知，本研究首次证实了14个连接基因的mRNA水平（CDH15、CDH17、CDH24、CLDN6、CLDN12、CLDN18、CTNND2、DSG2、ITGA2、ITGA8、ITGA11、ITGAL、ITGB4及PKP3）与LUAD患者的临床结局存在相关性。基于这14个基因构建JGRS评分，高JGRS评分与患者高龄、更高临床分期及更低免疫评分显著相关。据此，本研究构建了基于JGRS的预后预测列线图。内部及外部验证结果显示，该预测模型具有良好的预测效能。机制分析表明，JGRS评分与细胞增殖及免疫调控通路密切相关。突变分析显示，高JGRS组的体细胞突变数量显著多于低JGRS组。
研究结论：本研究通过"TCGA筛选、GEO验证"模型证实了LUAD患者连接基因与OS的相关性，揭示了连接基因的全新功能。