Abstract
Bladder cancer (BC) is a highly prevalent cancer form of the genitourinary system; however, the effective biomarkers are still ambiguous and deserve deeper investigation. Long non-coding RNA (lncRNA) occupies a prominent position in tumor biology and immunology, and ferroptosis-related genes participate in regulatory processes of cancer. In this study, 538 differentially expressed ferroptosis-related lncRNAs were identified from the The Cancer Genome Atlas database through co-expression method and differential expression analysis. Then, the samples involved were equally and randomly divided into two cohorts for the construction of gene model and accuracy verification. Subsequently, a prediction model containing five ferroptosis-related lncRNAs was constructed by LASSO and Cox regression analysis. Furthermore, in terms of predictive performance, consistent results were achieved in the training set, testing set, and entire set. Kaplan-Meier curve, receiver operating characteristic area under the curve, and principal component analysis results verified the good predictive ability of model, and the gene model was confirmed as an independent prognostic indicator. To further investigate the mechanism, we explored the upstream of five lncRNAs and found that they may be modified by m6A to increase or decrease their expression in BC. Importantly, the low-risk group displayed higher mutation burden of tumors and lower Tumor Immune Dysfunction and Exclusion score, which may be predicted to have a higher response rate to immunotherapy. Interestingly, the patients in the high-risk group appeared to have a higher sensitivity to traditional chemotherapeutic agents through pRRophetic analysis. In general, our research established a five-ferroptosis-related lncRNA signature, which can be served as a promising prognostic biomarker for BC.