Identification of a Lipid Homeostasis Related Genes Signature in Predicting the Prognosis, Immunity, and Chemotherapeutic Effect for Patients With Gastric Cancer

Abstract

Abstract Purpose Gastric cancer (GC) is one of the most common cancers which causes millions of deaths annually. Lipids homeostasis is essential for the development of tumor because lipids metabolism is one of the most important metabolic reprogramming process within tumor. Elucidating the mechanism of lipids homeostasis in GC might significantly improve the treatment strategies and prognosis.Methods GSE62254 was applied to construct lipids homeostasis related genes signature (HGS) by multiple bioinformatic algorithms. HGS was successfully validated in TCGA-STAD and GSE84437. Besides, the correlations between HGS, immune/stromal cells infiltration and chemotherapeutic effect were analyzed as well.Result HGS was constructed based on eighteen lipids homeostasis related genes by WGCNA. Additionally, HGS was strongly associated with advanced TNM stages and it showed satisfactory value in predicting GC prognosis in three independent cohorts. Furthermore, we found HGS was associated with tumor mutation burden (TMB) and immune/stromal cells infiltration which were related with GC prognosis, indicating that lipids homeostasis had impact on the formation of tumor microenvironment (TME). In GSCAlite platform, PLOD2 and TGFB2 was proved to be positively related with chemotherapeutic resistance while SLC10A7 was a favorable factor for chemotherapy. Finally, CCK8 assay showed that disrupted lipids homeostasis could attenuate GC cell viability and specific genes expression levels were evaluated by RT-qPCR.Conclusion We identified a lipids homeostasis related genes signature which was correlated with the survival, clinical characteristics, TME, and chemotherapeutic of GC. This research provided a new perspective in improving prognosis and guiding individual chemotherapy for patients with GC.