Construction and validation of cuproptosis-related lncRNA prediction signature for bladder cancer and immune infiltration analysis

Bladder cancer (BC) is a common urologic tumor with a high recurrence rate. Cuproptosis and long noncoding RNAs (lncRNAs) have demonstrated essential roles in the tumorigenesis of many malignancies. Nevertheless, the prognostic value of cuproptosis-related lncRNA (CRLs) in BC is still unclear. The public data used for this study were acquired from the Cancer Genome Atlas database. A comprehensive exploration of the expression profile, mutation, co-expression, and enrichment analyses of cuproptosis-related genes was performed. A total of 466 CRLs were identified using Pearson's correlation analysis. 16 prognostic CRLs were then retained by univariate Cox regression. Unsupervised clustering divided the patients into two clusters with diverse survival outcomes. The signature consists of 7 CRLs was constructed using the least absolute shrinkage and selection operator (LASSO) Cox regression analyses. Survival curves and receiver operating characteristics showed the prognostic signature possessed good predictive value, which was validated in the testing and entire sets. The reliability and stability of our signature were further confirmed by stratified analysis. Additionally, the signature-based risk score was confirmed as an independent prognostic factor. Gene set enrichment analysis showed molecular alteration in the high-risk group was closely associated with cancer. We then developed the clinical nomogram using independent prognostic indicators. Notably, the infiltration of immune cells and expression of immune checkpoints were higher in the high-risk group, suggesting that they may benefit more from immunotherapy. In summary, the prognostic signature might effectively predict the prognosis and provide new insight into the clinical treatment of BC patients.