Abstract
PURPOSE: Cuproptosis is a newly identified form of programmed cell death. We aimed to comprehensively discuss the correlation of cuproptosis with gastric cancer (GC) using bioinformatic methods. PATIENTS AND METHODS: This study selected GC bulk and single-cell RNA sequencing profiles from public databases. Based on the enrichment pattern of cuproptosis-related gene sets (CRGSs), GC patients were classified into different cuproptosis subtypes. A series of systematic analyses was performed to investigate the correlation of cuproptosis subtype with biological function and immune cell infiltration. In addition, we established a CRGS risk score signature to quantify GC patients' risk level, and analyzed the signature's relationship with clinical features, tumor microenvironment (TME) and treatment responses. Genes used for the construction of the risk score model were also detected in GC tumor and normal tissues by real-time quantitative polymerase chain reaction (RT-qPCR). RESULTS: First, analysis of scRNA-seq data revealed the alterations in CRGS enrichment scores for patients with GC and precancerous diseases. Then, based on large GC patient cohorts, two cuproptosis subtypes with significant differences in survival, biological function and TME were identified. Furthermore, we established a CRGS risk score signature. High-risk patients on the CRGS risk score signature with worse overall survival were characterized by higher immune and stromal contents in the TME, more advanced clinicopathological features, and better sensitivity to a wider range of anti-tumor drugs. Low-risk patients were correlated with higher tumor purity, and demonstrated more favorable clinical outcomes and higher sensitivity to immunotherapy. CONCLUSION: The current work elucidated that cuproptosis plays an important role in the regulation of TME landscapes in GC. Two cuproptosis subtypes with distinct TME characteristics were identified. In addition, the establishment of a CRGS risk score signature could provide novel insights into accurate prediction and personalized treatment for GC patients.