Research Paper Volume 12, Issue 21 pp 22233—22252

Development and validation of a novel prognostic signature in gastric adenocarcinoma

Rui Mao1, , Zheng Wang2, , Yuanchuan Zhang3, , YuanYuan Chen4, , Qian Liu2, , Tongtong Zhang5, , Yanjun Liu1,3, ,

  • 1 Affiliated Hospital of Southwest Jiaotong University, Chengdu, 610036, China
  • 2 Department of Colorectal Surgery, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100021, China
  • 3 The Center of Gastrointestinal and Minimally Invasive Surgery, The Third People’s Hospital of Chengdu, Chengdu, 610031, China
  • 4 Department of Pathology, The Third People’s Hospital of Chengdu, Chengdu, 610031, China
  • 5 Medical Research Center, The Third People’s Hospital of Chengdu, The Affiliated Hospital of Southwest Jiaotong University, The Second Chengdu Hospital Affiliated to Chongqing Medical University, Chengdu 610031, Sichuan, China

Received: April 27, 2020       Accepted: August 27, 2020       Published: November 8, 2020
How to Cite

Copyright © 2020 Mao 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.


Competing endogenous RNA networks have attracted increasing attention in gastric adenocarcinoma (GA). The current study aimed to explore ceRNA-based prognostic biomarkers for GA. RNA expression profiles were downloaded from TCGA and GEO databases. A ceRNA network was constructed based on the most relevant modules in the weighted gene coexpression network analysis. Kaplan-Meier (KM) survival analysis revealed prognosis-related RNAs, which were subjected to the multivariate Cox regression analysis. The predictive accuracy and discriminative ability of the signature were determined by KM analyses, receiver operating characteristic curves and area under the curve values. Ultimately, we constructed a ceRNA network consisting of 55 lncRNAs, 17 miRNAs and 73 mRNAs. Survival analyses revealed 3 lncRNAs (LINC01106, FOXD2-AS1, and AC103702.2) and 3 mRNAs (CCDC34, ORC6, and SOX4) as crucial prognostic factors; these factors were then used to construct a survival specific ceRNA network. Patients with high risk scores exhibited significantly worse overall survival than patients with low risk scores, and the AUC for 5-year survival was 0.801. A total of 112 GA specimens and the GSE84437 dataset were used to successfully validate the robustness of our signature by qRT-PCR. In summary, we developed a prognostic signature for GA, that shows better accuracy than the traditional TNM pathological staging system.


GA: gastric adenocarcinoma; TCGA: The Cancer Genome Atlas; GEO: Gene Expression Omnibus; lncRNAs: long noncoding RNAs; ROC: receiver operating characteristic; OS: overall survival; AUC: area under the curve; WGCNA: weighted gene coexpression network analysis; KEGG: Kyoto Encyclopedia of Genes and Genomes; qRT-PCR: real-time quantitative reverse transcription polymerase chain reaction; ceRNA: competing endogenous RNA.