Construction of a lipid metabolism-related and immune-associated prognostic score for gastric cancer

Abstract

Abstract Background The interaction between tumor cells and immune or non-immune stromal cells creates a unique tumor microenvironment, which plays an important role in the growth, invasion and metastasis of gastric cancer (GC). Methods The candidate genes were selected to construct risk-score by univariate and multivariate Cox regression analysis. Nomograms were constructed by combining clinical pathological factors, and the model performance was evaluated by receiver operating characteristic curve, decision curve analysis, net reclassification improvement and integrated discrimination improvement. The functional enrichment between high-risk group (HRisk) and low-risk group (LRisk) was explored through GO, KEGG, GSVA and ssGSEA. CIBERSORT, quanTIseq and xCell were used to explore the immune cell infiltration between HRisk and LRisk. The relevant EMT scores, macrophage infiltration scores and various metabolic scores were calculated through the “IOBR” package and analyzed visually. Results Through univariate and multivariate Cox regression analysis, we obtained the risk-score of fittings six lipid metabolism related genes (LMAGs). Through survival analysis, we found that risk-score has significant prognostic significance and can accurately reflect the metabolic level of patients. The AUCs of the nomogram model incorporating risk-score 1, 3 and 5 years were 0.725, 0.729 and 0.749 respectively. In addition, it was found that the inclusion of risk-score could significantly improve the prediction performance of the model. It was found that the arachidonic acid metabolism and prostaglandin synthesis were up-regulated in HRisk, and more tumor metastasis related markers and immune related pathways were also enriched. Further study found that HRisk had higher immune score and M2 macrophage infiltration. More importantly, the immune checkpoints of tumor associated macrophages involved in tumor antigen recognition disorders increased significantly. We also found that ST6GALNAC3 can promote arachidonic acid metabolism and up-regulate prostaglandin synthesis, increase M2 macrophage infiltration, induce epithelial mesenchymal transformation, and affect the prognosis of patients. Conclusions Our research found a novel and powerful LMAGs signature. Six-LMAGs features can effectively evaluate the prognosis of GC patients and reflect the metabolic and immune status. ST6GALNAC3 may be a potential prognostic marker to improve the survival rate and prognostic accuracy of GC patients, and may even be a potential biomarker of GC patients, indicating the response to immunotherapy.