Development and validation of a novel nomogram model for predicting the survival of patients with T2-4a, N0-x, M0 bladder cancer: a retrospective cohort study
Yu Xia,Xi Liu,Binbin Ma,Tao Huang,Danfeng Xu,Chenhui Zhao
2023-01-01
Abstract:Objective: Recent developments in bladder cancer treatment strategies have significantly improved the prognosis of clinically curable muscle invasive bladder cancer (MIBC) patients. Here, the prognostic factors of T2-4a, N0-x, M0 MIBC patients were investigated using the Surveillance, Epidemiology, and End Results (SEER) database and a novel nomogram model was established for prognosis prediction.Methods: The data of 7,292 patients with T2-4a, N0-x, M0 MIBC were retrieved from the SEER database (2000-2020) and randomly classified into a training set (n = 5,106) and validation set (n = 2,188). Kaplan-Meier analysis was used to calculate cancer-specific survival (CSS) and overall survival (OS) rates of patients, and differences between survival curves were analyzed using the log-rank test. Cox regression analysis was used to screen and incorporate patient prognosis-affecting independent risk factors into the nomogram model. Consistency index (C-index) values and areas under the time-dependent receiver operating characteristic curve (AUC) were used to evaluate the discriminatory ability, and the calibration curve was used to assess the calibration of the model. Its predictive performance and American Joint Committee on Cancer (AJCC) stage were compared using decision curve analysis (DCA).Results: The 1-, 3-, and 5-year CSS and OS rates of patients with T2-4a, N0-x, M0 MIBC were 76.9%, 56.0%, and 49.9%, respectively, and 71.3%, 47.9%, and 39.5%, respectively. Cox regression analysis showed that age, marital status, race, pathological type, tumor size, AJCC stage, T stage, N stage, surgery of primary tumor, regional lymph node dissection, radiation, and chemotherapy were independent prognostic risk factors of both CSS and OS (P < 0.05). The C-index and AUC of the nomogram model constructed based on the training and validation sets were both > 0.7, and calibration curves for predicting the 1-, 3-, and 5-year survival were consistent with the ideal curve. The nomogram model showed a higher net benefit with DCA than AJCC stage analysis.Conclusion: The nomogram model could accurately predict the prognosis of patients with T2-4a, N0-x, M0 MIBC. It may help clinicians perform personalized prognosis evaluations and formulate treatment plans.