世界上最常见的癌症之一是肺癌，而腺癌（LUAD）占据了其中相当大的比例。根据美国国家癌症研究所（NCI）的数据，每10万人中新发病例和死亡率如下：每年肺癌和支气管癌的新发病例率为50.0/10万人。男性和女性每年的死亡率为35.0/10万人。DNA甲基化是最早被发现和广泛研究的表观遗传调控机制之一，其异常与癌症的发生和发展密切相关。然而，需要进一步探索DNA甲基化和LUAD的预后价值，以改善LUAD患者的生存预测。LUAD的转录组数据和临床数据从TCGA和GEO数据库中下载，Illumina人类甲基化450芯片（450k芯片）数据从TCGA数据库中下载。首先，校正两个数据库中表达的基因的交集，进行差异分析，并使用MethylMix包评估甲基化数据以获得差异甲基化基因。使用单变量和多变量Cox回归分析筛选出独立的预后基因，并使用单变量Cox分析开发了一个甲基化预后模型，并使用GEO数据库中的GSE30219数据集加以验证。进行甲基化高风险组和低风险组的生存分析，并构建了基于甲基化的基因预后模型。最后，使用癌症药物敏感性基因组学（GDSC）对LUAD基因签名相关的潜在药物进行预测。本研究共纳入TCGA数据库的555个样本和GSE30219的307个样本，共确定了24个差异甲基化驱动基因。采用单变量和多变量Cox回归分析筛选出独立的预后基因，涉及2个基因：CFTR和PKIA。甲基化高风险组和低风险组之间的生存分析结果不同，CFTR高甲基化组和低甲基化组的预后差，而PKIA则相反。我们的研究揭示了CFTR和PKIA的甲基化状态可作为肺癌潜在的预后生物标志物和治疗靶点。© 2023年，法国国家卫生与医学研究所（INSERM）。

One of the most prevalent cancers in the world is lung cancer, with adenocarcinoma (LUAD) making up a significant portion of cases. According to the National Cancer Institute (NCI), there are new cases and fatality rates per 100,000 individuals as follows: New instances of lung and bronchial cancer occur annually at a rate of 50.0 per 100,000 persons. The yearly death rate for men and women is 35.0 per 100,000. DNA methylation is one of the earliest discovered and widely studied epigenetic regulatory mechanisms, and its abnormality is closely related to the occurrence and development of cancer. However, the prognostic value of DNA methylation and LUAD needs to be further explored to improve the survival prediction of LUAD patients.The transcriptome data and clinical data of LUAD were downloaded from TCGA and GEO databases, and the Illumina Human Methylation450 array (450k array) data were downloaded from the TCGA database. Firstly, the intersection of the expressed genes of the two databases is corrected, the differential analysis is performed, and the methylation data is evaluated by the MethylMix package to obtain differentially methylated genes. Independent prognostic genes were screened out using univariate and multivariate Cox regression analysis, and a methylation prognostic model was developed using univariate Cox analysis and validated with the GSE30219 dataset in the GEO database. Survival analysis between methylation high-risk and low-risk groups was performed and a methylation-based gene prognostic model was constructed. Finally, the prediction of potential drugs associated with the LUAD gene signature using Drug Sensitivity Genomics in Cancer (GDSC).In this study, a total of 555 samples from the TCGA database and 307 samples from GSE30219 were included, and a total of 24 differential methylation driver genes were identified. Univariate and multivariate Cox regression analyzes were used to screen out independent prognostic genes, involving 2 genes: CFTR, PKIA. Survival analysis was different between the methylation high-risk group and the low-risk group, the CFTR high methylation group and the low methylation group were poor, and the opposite was true for PKIA.Our study revealed that the methylation status of CFTR and PKIA can serve as potential prognostic biomarkers and therapeutic targets in lung cancer.© 2023. Institut National de la Santé et de la Recherche Médicale (INSERM).