Abstract
Background: Prostate cancer (PCa) is one of the most common cancers in males around the globe, and about one-third of patients with localized PCa will experience biochemical recurrence (BCR) after radical prostatectomy or radiation therapy. Reportedly, a proportion of patients with BCR had a poor prognosis. Cumulative studies have shown that RNA modifications participate in the cancer-related transcriptome, but the role of pseudouridylation occurring in lncRNAs in PCa remains opaque. Methods: Spearman correlation analysis and univariate Cox regression were utilized to determine pseudouridylation-related lncRNAs with prognostic value in PCa. Prognostic pseudouridylation-related lncRNAs were included in the LASSO (least absolute shrinkage and selection operator) regression algorithm to develop a predictive model. KM (Kaplan-Meier) survival analysis and ROC (receiver operating characteristic) curves were applied to validate the constructed model. A battery of biological cell assays was conducted to confirm the cancer-promoting effects of RP11-468E2.5 in the model. Results: A classifier containing five pseudouridine-related lncRNAs was developed to stratify PCa patients on BCR and named the "ψ-lnc score." KM survival analysis showed patients in the high ψ-lnc score group experienced BCR more than those in the low ψ-lnc score group. ROC curves demonstrated that ψ-lnc score outperformed other clinical indicators in BCR prediction. An external dataset, GSE54460, was utilized to validate the predictive model's efficacy and authenticity. A ceRNA (competitive endogenous RNA) network was constructed to explore the model's potential molecular functions and was annotated through GO (Gene Ontology) and KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analyses. RP11-468E2.5 was picked for further investigation, including pan-cancer analysis and experimental validation. Preliminarily, RP11-468E2.5 was confirmed as a tumor promoter. Conclusion: We provide some evidence that pseudouridylation in lncRNA played a role in the development of PCa and propose a novel prognostic classifier for clinical practice.