Tumor mutation burden-related long non-coding RNAs is predictor for prognosis and immune response in pancreatic cancer

Background Pancreatic cancer is one of the most common malignant tumors with extremely poor prognosis. It is urgent to identify promising prognostic biomarkers for pancreatic cancer. Methods A total of 266 patients with pancreatic adenocarcinoma (PAAD) in the Cancer Genome Atlas (TCGA)-PAAD cohort and the PACA-AU cohort were enrolled in this study. Firstly, prognostic tumor mutation burden (TMB)-related long non-coding RNAs (lncRNAs) were identified by DESeq2 and univariate analysis in the TCGA-PAAD cohort. And then, the TCGA-PAAD cohort was randomized into the training set and the testing set. Least absolute shrinkage and selection operator (LASSO) was used to construct the model in the training set. The testing set, the TCGA-PAAD cohort and the PACA-AU cohort was used as validation. The model was evaluated by multiple methods. Finally, functional analysis and immune status analysis were applied to explore the potential mechanism of our model. Results A prognostic model based on fourteen TMB-related lncRNAs was established in PAAD. Patients with High risk score was associated with worse prognosis compared to those with low risk score in all four datasets. Besides, the model had great performance in the prediction of 5-year overall survival in four datasets. Multivariate analysis also indicated that the risk score based on our model was independent prognostic factor in PAAD. Additionally, our model had the best predictive efficiency in PAAD compared to typical features and other three published models. And then, our findings also showed that high risk score was also associated with high TMB, microsatellite instability (MSI) and homologous recombination deficiency (HRD) score. Finally, we indicated that high risk score was related to low immune score and less infiltration of immune cells in PAAD. Conclusion we established a 14 TMB-related lncRNAs prognostic model in PAAD and the model had excellent performance in the prediction of prognosis in PAAD. Our findings provided new strategy for risk stratification and new clues for precision treatment in PAAD.