Xingxing Zhang1, Xu Chen1,2, Jiayun Liu1, Yaqi Li1,2, Jian Wu1, Menglin Chen1,2, Ruijuan Zhang1,2, Xintian Xu1, Tianyi Xu3,4, Qingmin Sun5. 1. Jiangsu Province Hospital of Chinese Medicine, Affiliated Hospital of Nanjing University of Chinese Medicine, Nanjing, 210029, Jiangsu, China. 2. No. 1 Clinical Medical College, Nanjing University of Chinese Medicine, Nanjing, 210023, Jiangsu, China. 3. National Genomics Data Center & CAS Key Laboratory of Genome Sciences and Information, Beijing Institute of Genomics, Chinese Academy of Sciences, Beijing, 100101, China. xuty@big.ac.cn. 4. China National Center for Bioinformation, Beijing, 100101, China. xuty@big.ac.cn. 5. Jiangsu Province Hospital of Chinese Medicine, Affiliated Hospital of Nanjing University of Chinese Medicine, Nanjing, 210029, Jiangsu, China. qingminsun@njucm.edu.cn.
Abstract
BACKGROUND: The importance of metabolism-related alterations in the development of gastric cancer (GC) is increasingly recognized. The present study aimed to identify metabolism-related genes to facilitate prognosis of GC patients. METHODS: Gene expression datasets and clinical information of GC patients were downloaded from TCGA and GEO databases. We scored the enrichment of human metabolism-related pathways (n = 86) in GC samples by GSV, constructed prognostic risk models using LASSO algorithm and multivariate Cox regression analysis, combined with clinical information to construct a nomogram, and finally cis score algorithm to analyze the abundance of immune-related cells in different subtypes. We used Weka software to screen for prognosis-related marker genes and finally validated the expression of the selected genes in clinical cancer patient tissues. RESULTS: We identified that two GC metabolism-related signatures were strongly associated with OS and the levels of immune cell infiltration. Moreover, a survival prediction model for GC was established based on six GC metabolism-related genes. Time-dependent ROC analysis showed good stability of the risk prediction scoring model. The model was successfully validated in an independent ACRG cohort, and the expression trends of key genes were also verified in the GC tissues of patients. DLX1, LTBP2, FGFR1 and MMP2 were highly expressed in the cluster with poorer prognosis while SLC13A2 and SLCO1B3 were highly expressed in the cluster with better prognosis. CONCLUSIONS: We identified a risk predictive score model based on six metabolism-related genes related to survival, which may serve as prognostic indicators and potential therapeutic targets for GC.
BACKGROUND: The importance of metabolism-related alterations in the development of gastric cancer (GC) is increasingly recognized. The present study aimed to identify metabolism-related genes to facilitate prognosis of GC patients. METHODS: Gene expression datasets and clinical information of GC patients were downloaded from TCGA and GEO databases. We scored the enrichment of human metabolism-related pathways (n = 86) in GC samples by GSV, constructed prognostic risk models using LASSO algorithm and multivariate Cox regression analysis, combined with clinical information to construct a nomogram, and finally cis score algorithm to analyze the abundance of immune-related cells in different subtypes. We used Weka software to screen for prognosis-related marker genes and finally validated the expression of the selected genes in clinical cancer patient tissues. RESULTS: We identified that two GC metabolism-related signatures were strongly associated with OS and the levels of immune cell infiltration. Moreover, a survival prediction model for GC was established based on six GC metabolism-related genes. Time-dependent ROC analysis showed good stability of the risk prediction scoring model. The model was successfully validated in an independent ACRG cohort, and the expression trends of key genes were also verified in the GC tissues of patients. DLX1, LTBP2, FGFR1 and MMP2 were highly expressed in the cluster with poorer prognosis while SLC13A2 and SLCO1B3 were highly expressed in the cluster with better prognosis. CONCLUSIONS: We identified a risk predictive score model based on six metabolism-related genes related to survival, which may serve as prognostic indicators and potential therapeutic targets for GC.