Research Paper Volume 14, Issue 18 pp 7328—7347

Establishment and validation of individualized clinical prognostic markers for LUAD patients based on autophagy-related genes

Yuchang Fei1, , Junyi Xu2, , Liping Ge3, , Luting Chen4, , Huan Yu5, , Lei Pan6, , Peifeng Chen6, ,

  • 1 Department of Integrated Chinese and Western Medicine, The First People’s Hospital of Jiashan, Jiaxing, Zhejiang, China
  • 2 Information Center, The First People’s Hospital of Jiashan, Jiaxing, Zhejiang, China
  • 3 Department of Breast Surgery, Fudan University Shanghai Cancer Center, Xuhui, Shanghai, China
  • 4 Department of Integrated Chinese and Western Medicine, The First People’s Hospital of Wenling, Taizhou, Zhejiang, China
  • 5 Ningbo Yinzhou Second Hospital, Ningbo, Zhejiang, China
  • 6 Department of Oncology, The First Affiliated Hospital of Zhejiang Chinese Medical University, Hangzhou, Zhejiang, China

Received: November 13, 2021       Accepted: May 13, 2022       Published: September 29, 2022
How to Cite

Copyright: © 2022 Fei et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


There is considerable heterogeneity in the genomic drivers of lung adenocarcinoma, which has a dismal prognosis. Bioinformatics analysis was performed on lung adenocarcinoma (LUAD) datasets to establish a multi-autophagy gene model to predict patient prognosis. LUAD data were downloaded from The Cancer Genome Atlas (TCGA) database as a training set to construct a LUAD prognostic model. According to the risk score, a Kaplan-Meier cumulative curve was plotted to evaluate the prognostic value. Furthermore, a nomogram was established to predict the three-year and five-year survival of patients with LUAD based on their prognostic characteristics. Two genes (ITGB1 and EIF2AK3) were identified in the autophagy-related prognostic model, and the multivariate Cox proportional risk model showed that risk score was an independent predictor of prognosis in LUAD patients (HR=3.3, 95%CI= 2.3 to 4.6, P< 0.0001). The Kaplan-Meier cumulative curve showed that low-risk patients had significantly better overall (P<0.0001). The validation dataset GSE68465 further confirmed the nomogram’s robust ability to assess the prognosis of LUAD patients. A prognosis model of autophagy-related genes based on a LUAD dataset was constructed and exhibited diagnostic value in the prognosis of LUAD patients. Moreover, real-time qPCR confirmed the expression patterns of EIF2AK3 and ITGB1 in LUAD cell lines. Two key autophagy-related genes have been suggested as prognostic markers for lung adenocarcinoma.


LUAD: Lung adenocarcinoma; GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; LASSO: Least Absolute Shrinkage and Selection Operator; ARGs: Autophagy-related genes; PPI: Protein-Protein interaction; HPA: The Human Protein Atlas; GEO: Gene Expression Omnibus; TCGA: The Cancer Genome Atlas; BP: Biological processes; EIF2AK3: Eukaryotic translation initiation factor 2-alpha kinase 3; ITGB1: Integrin, beta 1; IL-17: Interleukin 17; ErbB: Epidermal Growth Factor Receptor erbB; PD-1: Programmed Cell Death Protein 1; PD-L1: Programmed Death Ligand-1; HR: Hazard ratio; DCA: Decision Curve Analysis; MA: Macroautophagy; MI: Microautophagy; CMA: Chain-mediated Autophagy; NSCLC: Non-small Cell Carcinoma; LC3: Autophagy-associated proteins; p62/ SQSTM1: Sequestosome 1; CNV: Copy number variations; SINP: Silica Nanoparticles; IGLUR: Ionic Glutamate Receptor; GRM4: glutamate metabotropic receptor 4; ROC: Receiver operating characteristic; T: Tumour; N: Lymph Node; M: Metastasis.