A Novel Aging-Related lncRNAs Signature-Based Model for Prognosis Prediction in Pancreatic Cancer

Abstract

Abstract Objective To identify aging-related long non-coding RNAs (ARLs) with prognostic significance and construct a prognostic model for pancreatic cancer (PC) patients. Methods Transcriptome information from PC samples and normal samples was obtained from the Cancer Genome Atlas Database (TCGA) and the Genotypic Tissue Expression Database (GTEx). Aging-related genes (ARGs) were obtained from the Human Aging Genome Resources (HAGR) and GeneCards. Correlation analysis was performed to screen out ARLs. Univariate regression, lasso regression, and multivariate regression were used to identify the target ARLs and construct the prognostic model for aging-related PC. Results A total of 1109 ARLs were identified, and 9 target ARLs were obtained to construct the risk score prognostic model. These target ARLs include AC245041.2, AC244153.1, AC091057.1, MIR3142HG, AL137779.2, AC145207.5, TDRKH-AS1, AC068620.2, and AC127024.6. The model showed an area under the curve (AUC) of 0.798 on the receiver operating curve (ROC) curve, indicating its effectiveness in predicting prognosis. Kaplan-Meier analysis demonstrated a significant difference in overall survival (OS) between the two groups based on the median risk (P＜0.001). To further assess prognosis, the risk score was combined with clinicopathological features to construct a nomogram for PC. Additionally, gene enrichment analysis (GSEA) and immunological correlation analysis revealed differences in gene enrichment level, immune infiltration, and the expression of immune checkpoint genes between the two groups. Conclusion The constructed prognostic model based on nine ARLs provides valuable insights for the prognosis management of PC patients and the development of promising biomarkers in the diagnosis and treatment of PC.