CUI Yanqi 1 , YANG Jingrong 2 , Ni Lin 3 , LIAN Duohuang 2 , YE Shixin 2 , LIAO Yi 1 , ZHANG Jincan 2 , ZENG Zhiyong 1,2
  • 1. Fuzong Clinical College of Fujian Medical University, Fuzhou, 350025, P. R. China;
  • 2. Department of Cardiothoracic Surgery, The 900th Hospital of the Joint Support Force, Fuzhou, 350025, P. R. China;
  • 3. Department of General Surgery, The 900th Hospital of the Joint Support Force, Fuzhou, 350025, P. R. China;
ZENG Zhiyong, Email: 13295916182@163.com
Invasive lung adenocarcinoma; machine learning; biomarker; risk assessment model; nomogram.
10.7507/1007-4848.202303059
Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables References Cited by

Objective  To determine the prognostic biomarkers and new therapeutic targets of the lung adenocarcinoma (LUAD), based on which to establish a prediction model for the survival of LUAD. Methods  An integrative analysis was conducted on gene expression and clinicopathologic data of LUAD, which was obtained from the UCSC database. Subsequently, various methods, including screening of differentially expressed genes (DEGs), GO analysis, KEGG analysis, and GSEA, to analyze the data were employed. Our objective was to establish a five-gene panel risk assessment model using Cox regression and LASSO regression. Based on this model, we constructed a Nomogram to predict the probable survival of LUAD patients at different time points (1-year, 2-year, 3-year, 5-year, and 10-year). Finally, we evaluated the predictive ability of our model using Kaplan-Meier survival curves, ROC curves, and time-dependent ROC curves. The validation group further verified the prognostic value of the model. Results  The different-grade pathological subtypes' DEGs were mainly enriched in biological processes such as Metabolism of xenobiotics by cytochrome P450, Natural killer cell-mediated cytotoxicity, Antigen processing and presentation, and Regulation of enzyme activity, which were closely related to tumor development. Through Cox regression and LASSO regression, we constructed a reliable prediction model consisting of a five-gene panel (MELTF, MAGEA1, FGF19, DKK4, C14ORF105). The model demonstrated excellent specificity and sensitivity in ROC curves, with an area under the ROC curve (AUC) of 0.675, as well as in time-dependent ROC curves. The time-dependent ROC analysis revealed AUC values of 0.893, 0.713, and 0.632 for 1-year, 3-year, and 5-year survival, respectively. The advantage of the model was also verified in the validation group. Additionally, we developed a Nomogram that accurately predicted survival, as demonstrated by calibration curves and C-index. Conclusion  We have developed a prognostic prediction model for LUAD consisting of five genes. This novel approach offers clinical practitioners a personalized tool for making informed decisions regarding the prognosis of their patients.