| Abstract: | ![]()
BackgroundBoth N6-methyladenosine (m6A) ribonucleic acid (RNA) methylation and ferroptosis regulators are demonstrated to have significant effects on the malignant clinicopathological characteristics of pancreatic adenocarcinoma (PAAD) patients. However, the currently available clinical indexes are not sufficient to predict precise prognostic outcomes pf PAAD patients accurately. This study aims to examine the clinicopathologic features of m6A RNA methylation and ferroptosis regulators in predicting the outcomes of different types of cancer.MethodsAs the foundation for this research, the differentially expressed genes (DEGs) between PAAD tissues and adjacent normal tissues were first identified. Next, dimensional reduction analysis (DCA) based on m6A RNA methylation regulators and ferroptosis regulators were performed and DEGs between good/poor prognosis PAAD patient clusters were identified. DEGs were then screened by Cox analysis, and finally a risk signature was established by least absolute shrinkage and selection operator (LASSO) analyses. The prediction model based on risk score was further evaluated by a validation set from Gene Expression Omnibus (GEO) database.ResultsIn total, 4 m6A RNA methylation regulator genes and 29 ferroptosis regulator genes were found to have close causal relationships with the prognosis of PAAD, and a risk score with 3 m6A methylation regulators (i.e., IGF2BP2, IGF2BP3, and METTL16) and 4 ferroptosis regulators (i.e., ENPP2, ATP6V1G2, ITGB4, and PROM2) was constructed and showed to be highly involved in PAAD progression and could serve as effective markers for prognosis with AUC value equaled 0.753 in training set and 0.803 in validation set.ConclusionsThe combined prediction model, composed of seven regulators of m6A methylation and ferroptosis, in this study more effectively reflects the progression and prognosis of PAAD than previous single genome or epigenetic analysis. Our study provides a broader perspective for the subsequent establishment of prognostic models and the patients may benefit from more precision management. |
