A Nomogram Model to Predict Recurrence of Non-Muscle Invasive Bladder Urothelial Carcinoma After Resection Based on Clinical Parameters and Immunohistochemical Markers

Objective This study aims to establish a nomogram model by combining traditional clinical parameters with immunohistochemical markers to predict the recurrence of non-muscle invasive bladder urothelial carcinoma (NMIBUC) after resection. Methods In total, 504 patients were included in this study. Of these patients, 353 underwent transurethral resection of bladder tumor (TURBT) in the Yongchuan Hospital of Chongqing Medical University and were identified as a training cohort. Univariate and multivariate Cox regression analyses were used to determine the risk factors associated with recurrence in the training cohort and to establish a nomogram model. A total of 151 patients who were hospitalized in the Second Affiliated Hospital of Chongqing Medical University (validation cohort) were used for further validation. The calibration curve was generated for internal and external model validation. The clinical practicability of this model was further verified by comparing the consistency index (C-index) among various models. Results The mean follow-up time of the training cohort was 45.6 months (range 4-90). In total, 146 patients relapsed in training cohort. After univariate analysis, multivariate analysis further confirmed tumor grade (p=.034), immediate postoperative instillation therapy (p=.025), Ki67 (p=.047), P53 (p=.038) and CK20 (p=.049) as independent risk factors for recurrence, and these factors were included in the nomogram model. The model more accurately predicted recurrence compared with other models based on the highest C-index of 0.82 (95% CI, 0.78-0.86) in the training cohort and 0.80 (95% CI, 0.77-0.83) in the validation cohort. Conclusions This proposed nomogram model based on traditional clinical parameters and immunohistochemical markers can more accurately predict postoperative recurrence in patients with NMIBUC.