Prognostic Nomogram Based on Circular RNA-Associated Competing Endogenous RNA Network For Patients With Lung Adenocarcinoma

Abstract Evidence is increasingly indicating that circular RNAs (circRNAs) are closely involved in tumorigenesis and cancer progression. However, function and application of circRNAs in lung adenocarcinoma (LUAD) are still unknown. In this study, we selected differentially expressed circRNAs (DEcircRNAs), differentially expressed miRNAs (DEmiRNAs) and differentially expressed mRNAs (DEmRNAs) to establish a circRNA-associated competitive endogenous RNA (ceRNA) network and further constructed a prognostic signature for LUAD patients. Based on TargetScan prediction tool and Pearson correlation coefficient, we constructed a circRNA-associated ceRNA network including 11 DEcircRNAs, 8 DEmiRNAs and 49 DEmRNAs. Functional enrichment indicated that the ceRNA network might be involved in regulation of GTPase activity and endothelial cell differentiation. After removing the discrete points, a protein-protein interaction (PPI) network containing 12 DEmRNAs was constructed. Based on DEmRNAs within PPI network, we constructed a three-gene prognostic signature for LUAD patients using LASSO method. Patients in the training cohort could be categorized into high-risk or low-risk subgroup with significant survival difference (HR: 1.62, 95% CI: 1.12-2.35, log-rank p = 0.009). The prognostic performance was confirmed in an independent cohort (HR: 2.59, 95% CI: 1.32-5.10, log-rank p = 0.004). Combining three-gene signature with clinical risk characters, a nomogram was constructed. The calibration curves for probability of 3- and 5-year overall survival showed significant agreement between nomogram predictions and actual observations. Our findings provided a deeper understanding of the circRNA-associated ceRNA regulatory mechanism in LUAD pathogenesis and constructed a prognostic signature that could be a useful guide for personalized treatment of LUAD patients.