Identification of cuproptosis-related lncRNAs Prognostic Signature in Gastric cancer and Its Impact on the Tumor Microenvironment

Abstract Background: There is extensive genetic and transcriptional heterogeneity in Gastric cancer (GC). Increasing evidence has demonstrated that cuproptosis could exert an important function in tumor progression. Long noncoding RNAs (lncRNAs)，which exert a pivotal function in the development of GC, especially involving in tumor-associated immune progression. Thus, it is indispensable to establish a cuproptosis-related lncRNAs (crlncRNAs)signature. Methods: Based on 19 genes associated with cuproptosis, to investigate whether lncRNA was markedly associated with cuproptosis, Pearson correlation analysis came in handy. To identify characteristics of copper death-related lncRNAs and build predictive models, univariate Cox regression and least absolute shrinkage and selection operator (LASSO) analysis were used to identify 6 key cuproptosis-related lncRNAs (crlncRNAs). To validate the predictive power of the signature, Cox regression on the univariate and multivariate data, Kaplan-Meier survival analysis, receiver operating characteristic curves (ROC), area under the curve (AUC) are used in the training, testing, and total sets. We further examined the functional enrichment, the status of immune cells, immune cell infiltrations, landscape of mutation status, Tumor Immune Dysfunction and Exclusion (TIDE) score, correlation and drugs sensitivity among the different risk groups. Results: 16 prognostically related crlncRNAs(PRcrlncRNAs) were filtered to make the prognostic signature, 6 key crlncRNAs were included. The risk score was the independent parameter in predicting OS (Overall survival) according to Cox regression on the univariate and multivariate data and it is acceptable in predicting prognosis concerning the accuracy of the model, which is confirmed by ROC analysis in GC patients. Differences in median OS and PFS (progression-free survival) between high- and low-risk groups were statistically significant. The TIDE score was higher in high-risk patients, which could predict less chemosensitivity with some drugs in the light of the TIDE analysis. Conclusions: Our study is innovative to develop and validate a novel STAD-associated crlncRNAs model that could effectively instruct the prognosis, as well as participating in the immune microenvironment in STAD, which would provide a new insight in the development of molecularly targeted therapies associated to cuproptosis.