Abstract
The main purpose of this study was to construct two nomograms to predict all-cause mortality (ACM) and cause-specific mortality (CSM) in non-muscle-invasive bladder cancer (NMIBC) patients after transurethral resection of bladder tumors (TURBTs). We selected NMIBC patients who underwent TURBT between 2004 and 2017 from the Surveillance, Epidemiology, and End Results database. The patients were randomly divided into a training set and a validation set at a ratio of 7:3. The independent influencing factors of ACM and CSM in the training set were determined by univariate and multivariate Cox regression analyses. We then integrated those independent influencing factors to construct nomograms. These prediction nomograms were further verified in the validation set. The C-index, calibration curve, receiver operating characteristic (ROC) curve and decision curve analysis (DCA) curve were used to evaluate the identification, calibration, predictive ability and clinical effectiveness of the nomograms. A total of 28,086 cases were ultimately included in this study, which were divided into a training set (19,661 individuals) and a validation set (8425 individuals). Nine variables, including age at diagnosis, race, marital status, tumor grade, T stage, tumor size, number of tumors, and primary site, were obtained via multivariate Cox regression of the training set and used to construct two nomograms prediction model. The C-index values for the ACM nomogram were 0.743 and 0.741 for the training and validation sets, respectively. Moreover, the corresponding values of the C-index for the CSM nomogram were 0.785 and 0.786, respectively. The ROC curves, calibration curves, and DCA curves showed good predictive performance. The nomograms can assist clinicians in identifying high-risk populations and devising more individualized treatment strategies for NMIBC patients.