Yi Liu1, Min Wu2, Jian Cao3, Yaning Zhu4, Yu Ma5, Yansong Pu6, Xueping Huo7, Jianhua Wang6. 1. Department of Oncology, Shaanxi Provincial People's Hospital, Xi'an, China. 2. Office of Scientific Research, Shaanxi Provincial People's Hospital, Xi'an, China. 3. Department of Pharmacy, Xijing 986 Hospital, Air Force Military Medical, Xi'an, China. 4. Department of Pharmacy, Shaanxi Provincial People's Hospital, Xi'an, China. 5. Department of Pathology, Shaanxi Provincial People's Hospital, Xi'an, China. 6. The Second Department of General Surgery, Shaanxi Provincial People's Hospital, Xi'an, China. 7. Department of Central laboratory, Shaanxi Provincial People's Hospital, Xi'an, China.
Abstract
Background: Glycolysis is a central metabolic pathway for tumor cells. However, the relationship between glycolysis and the prognosis of gastric cancer (GC) patients is not well established. In this study, we sought to construct a glycolysis-related gene signature for GC. Methods: The messenger ribonucleic acid (mRNA) expression profiles were analyzed using data from The Cancer Genome Atlas (TCGA) database. Glycolysis-related gene sets and pathways were obtained from the Molecular Signatures Database (MSigDB). Subsequently, a prognosis prediction model of the glycolysis-related genes was constructed using Cox and least absolute shrinkage and selection operator (LASSO) regression analyses. An external validation was conducted using data from the Gene Expression Omnibus (GEO) database. Risk scores were also calculated based on the signature. Finally, the correlations between the risk score and overall survival (OS), mutation, immune cell infiltration, immune score, and stromal score were examined in 22 types of infiltrating immune cells. Results: Fifty-five glycolysis-related genes were identified from TCGA database and MSigDB. Using the LASSO and Cox models, 4 novel genes (i.e., VCAN, EFNA3, ADH4, and CLDN9) were identified to construct a gene signature for GC prognosis prediction. The GC patients with low-risk scores had significantly better OS than those with high-risk scores in the training set. Similar results were also found in the independent GEO GSE84437 testing set. Additionally, the degree of cell infiltration in the low-risk group was significantly higher than that in the high-risk group in terms of naive B cells, plasma cells, and T follicular helper cells. In monocytes, M2 macrophages, resting dendritic cells, and resting Mast cells, the degree of infiltration in the high-risk group was significantly higher than that in the low-risk group. The immune score and stromal score of the high-risk group were also significantly higher than those of the low-risk group. Finally, the univariate and multivariate Cox regression analyses showed that 4 glycolysis-related genes were independent prognostic factors for GC. Conclusions: The established 4 glycolysis-related gene signature may serve as a reliable tool for the prognosis of GC patients and provide a potential glycolysis therapeutic target for GC. 2022 Annals of Translational Medicine. All rights reserved.
Background: Glycolysis is a central metabolic pathway for tumor cells. However, the relationship between glycolysis and the prognosis of gastric cancer (GC) patients is not well established. In this study, we sought to construct a glycolysis-related gene signature for GC. Methods: The messenger ribonucleic acid (mRNA) expression profiles were analyzed using data from The Cancer Genome Atlas (TCGA) database. Glycolysis-related gene sets and pathways were obtained from the Molecular Signatures Database (MSigDB). Subsequently, a prognosis prediction model of the glycolysis-related genes was constructed using Cox and least absolute shrinkage and selection operator (LASSO) regression analyses. An external validation was conducted using data from the Gene Expression Omnibus (GEO) database. Risk scores were also calculated based on the signature. Finally, the correlations between the risk score and overall survival (OS), mutation, immune cell infiltration, immune score, and stromal score were examined in 22 types of infiltrating immune cells. Results: Fifty-five glycolysis-related genes were identified from TCGA database and MSigDB. Using the LASSO and Cox models, 4 novel genes (i.e., VCAN, EFNA3, ADH4, and CLDN9) were identified to construct a gene signature for GC prognosis prediction. The GC patients with low-risk scores had significantly better OS than those with high-risk scores in the training set. Similar results were also found in the independent GEO GSE84437 testing set. Additionally, the degree of cell infiltration in the low-risk group was significantly higher than that in the high-risk group in terms of naive B cells, plasma cells, and T follicular helper cells. In monocytes, M2 macrophages, resting dendritic cells, and resting Mast cells, the degree of infiltration in the high-risk group was significantly higher than that in the low-risk group. The immune score and stromal score of the high-risk group were also significantly higher than those of the low-risk group. Finally, the univariate and multivariate Cox regression analyses showed that 4 glycolysis-related genes were independent prognostic factors for GC. Conclusions: The established 4 glycolysis-related gene signature may serve as a reliable tool for the prognosis of GC patients and provide a potential glycolysis therapeutic target for GC. 2022 Annals of Translational Medicine. All rights reserved.
Entities:
Keywords:
Gastric cancer (GC); LASSO and Cox regression; differential analysis; glycolysis; prognosis
Authors: Aaron M Newman; Chih Long Liu; Michael R Green; Andrew J Gentles; Weiguo Feng; Yue Xu; Chuong D Hoang; Maximilian Diehn; Ash A Alizadeh Journal: Nat Methods Date: 2015-03-30 Impact factor: 28.547