Development and validation of a cuproptosis-related gene signature for predicting prognosis and drug sensitivity in gastric cancer.

BACKGROUND

Cuproptosis, a recently identified copper-dependent form of cell death linked to the tricarboxylic acid cycle, remains poorly understood with respect to its association with prognosis and therapeutic response in gastric cancer (GC). This study aimed to construct a prognostic model based on cuproptosis-related genes (CRGs) and to evaluate its clinical utility for predicting patient survival and therapeutic response.

METHODS

Transcriptomic and clinical data of GC patients were obtained from the TCGA and GEO databases. Consensus clustering was applied to identify molecular subtypes based on CRG expression profiles. A prognostic model was developed using LASSO regression on differentially expressed genes, and a nomogram integrating clinical variables and the signature was constructed to enhance outcome prediction. The model's performance was assessed via Kaplan-Meier survival analysis, log-rank tests, Cox regression, and time-dependent ROC curves. Chemotherapy and immunotherapy response predictions were evaluated using the pRRophetic package and the TIDE algorithm, respectively. Single-cell RNA sequencing analysis was conducted using the Seurat package, and the expression of key genes was validated by qRT-PCR in cell lines and clinical specimens.

RESULTS

Comprehensive analysis of CRG expression and prognostic relevance identified two distinct cuproptosis-associated molecular subtypes in GC. A six-gene prognostic model was developed via LASSO regression. Patients were stratified into high- and low-risk groups. The high-risk group exhibited worse survival and more advanced TNM stages, whereas the low-risk group showed greater predicted sensitivity to chemotherapy and immunotherapy. ROC analysis confirmed the model's reasonable predictive performance.

CONCLUSIONS

This study reveals significant associations between CRG expression, predicted therapeutic response, and clinical prognosis in GC patients. The developed prognostic model may serve as a clinically useful tool for survival prediction and could facilitate personalized therapeutic strategies.