Development and Evaluation of a Novel Cuproptosis-Related lncRNA Signature for Gastric Cancer Prognosis.

Background: According to a growing body of research, long noncoding RNAs (lncRNAs) participate in the progress of gastric cancer (GC). Cuproptosis is a distinct kind of programmed cell death, separating it from several other forms of programmed cell death that may be caused by genetic programming. Consequently, it is crucial to examine cuproptosis-related lncRNAs (CRLs) prognostic importance for the prognosis and treatment response in GC. Method: The Cancer Genome Atlas (TCGA) database was used to retrieve RNA-seq data, pertinent clinical information, and somatic mutation data. A list of cuproptosis-related genes (CRGs) was obtained from prior work. We can distinguish prognostic CRLs using coexpression and univariate Cox analysis. Then, using CRLs, we developed a risk prediction model using multivariate Cox regression analysis and the least absolute shrinkage selection operator (LASSO) technique. To evaluate the diagnostic accuracy of this model, a Kaplan-Meier (K-M) survival analysis and a receiver operating characteristic (ROC) analysis were used. Moreover, the relationships between the risk model and immunological function, somatic mutation, and drug sensitivity were also investigated. Results: Using the multivariate Cox analysis technique, we developed a signature based on cuproptosis-related four lncRNAs. We then classified patients into high-risk and low-risk groups based on the likelihood of unfavorable outcomes. The model was subjected to further testing, including K-M survival analysis, ROC analysis, and multivariate Cox regression analysis, all of which proved the model's exceptional robustness and predictive capacity. In addition, a nomogram that has a strong capacity for prediction ability was built. This nomogram included age, gender, clinical grade, pathologic stage, T stage, and risk score. Furthermore, we discovered substantial disparities in immune function and the number of mutations carried by tumors between the high-risk and low-risk groups. Moreover, this research also found that the IC50 values for 27 chemotherapeutic drugs varied widely across patients within high- and low-risk groups. Conclusion: The proposed 4-CRLs signature is a promising biomarker to predict clinical outcomes in GC.