Construction and validation of a novel prognostic model using the cellular senescence-associated long non-coding RNA in gastric cancer: a biological analysis

Abstract

BackgroundThe onset and progression of many cancers, including gastric cancer (GC), are strongly influenced by cell senescence. Numerous studies have demonstrated that long non-coding RNA (lncRNA) impacts cell senescence, thus affecting cancer progression. However, it is not possible to develop a relevant predictive model for GC owing to the absence of a cell senescence-linked lncRNA. Since lncRNAs are linked to cellular senescence, the goal of this work was to create a prognostic signature for stomach adenocarcinoma (STAD) patients utilizing these lncRNAs.MethodsThrough the Pearson correlation, variance, and univariate Cox regression analyses, the cellular senescence lncRNAs that were related to the disease prognosis could be successfully identified. Using the least absolute shrinkage and selection operator (LASSO) regression algorithm, a predictive model that utilized the 11 cellular senescence-linked lncRNAs was constructed. Kaplan-Meier (KM) survival and the receiver operating characteristic (ROC) curve analyses, were employed for assessing the prognostic performance of the proposed model. In addition, ESTIMATE analysis of the low- and high-risk subtypes for the infiltration of various immune cells was carried out. Additionally, the CIBERSORT algorithm was utilized for investigating the infiltration status of numerous immune cells in both groups, while the expression of the immune checkpoint genes in the two groups, was also determined.ResultsIn this study, a new prognostic model was constructed using 11 cellular senescence-related lncRNAs. The findings revealed that the OS status of the patients in the low-risk group (category) was significantly higher compared to the high-risk category (P<0.001). The 1-year ROC-area under the curve (AUC) values for the risk score in the training group was 0.714, while the AUC value for the test and comprehensive groups were recorded to be 0.666 and 0.695, respectively, which were obviously due to stage, grade, age, etc. And based on univariate [hazard ratio (HR): 1.435; P<0.001; 95% confidence interval (CI): 1.295–1.589] and multivariate analyses (P<0.001; 95% CI: HR: 1.387; 1.247–1.543), it was noted that risk scores were effectively employed as a patient-independent prognostic factor.ConclusionsTaken together, these results suggest that cellular senescence-related lncRNAs are likely to be valuable prognostic markers for GC. They also reflect the situation of the STAD immune microenvironment and may provide direction for future GC treatment.