Machine learning-random forest model was used to construct gene signature associated with cuproptosis to predict the prognosis of gastric cancer.

Gastric cancer (GC) is one of the most common tumors; one of the reasons for its poor prognosis is that GC cells can resist normal cell death process and therefore develop distant metastasis. Cuproptosis is a novel type of cell death and a limited number of studies have been conducted on the relationship between cuproptosis-related genes (CRGs) in GC. The purpose of the present study was to establish a prognostic model of CRGs and provide directions for the diagnosis and treatment of GC. Transcriptome and clinical data of patients with GC were collected from The Cancer Genome Atlas and Gene Expression Omnibus datasets. Single sample gene set enrichment analysis (GSEA) and the randomized forest method were used to establish the prognostic model. Kaplan-Meier survival curve, receiver operating characteristics diagram and a nomogram were used to evaluate the reliability of the model. GSEA and gene set variation analysis (GSVA) were used to examine enrichment pathways between high and low risk groups. Finally, immunohistochemical analysis was used to examine ephrin 4 (EFNA4) expression in GC samples and determine the prognosis of patients with GC based on the expression pattern of EFNA4. A group of 7 predictive models (RTKN2, INO80B, EFNA4, ELF2, MUSTN, KRTAP4, and ARHGEF40) was established which were correlated with CRGs. This model can be used as an independent prognostic factor to predict the prognosis of patients with GC. GSEA and GSVA results indicated that high risk patients with GC were mainly associated with the enrichment of ANGIOGENESIS and TGF_BETA_SIGNALING pathways. Finally, EFNA4 expression in GC was significantly higher than that in normal tissues, and patients with GC and high EFNA4 expression exhibited improved prognosis. In conclusion, the prognosis model based on CRGs could be used as the basis for predicting the potential prognosis of patients with GC and provide new insights for the treatment of GC.