Development and Internal Validation of a Nomogram Predicting Overall Survival Based on Log ODDS of Positive Lymph-Nodes for Post Radical Cystectomy Patients in Muscle Invasive Carcinoma of Bladder.

Abstract

To develop and validate a nomogram based on LODDS (Log ODDS of positive lymph-nodes) for prediction of overall survival (OS) in post radical cystectomy (RC) patients of muscle invasive bladder cancer (MIBC). Data was retrospectively collected from 282 cases of MIBC that underwent RC from 2011 to 2017 at our institute. Significant independent predictors were identified using Cox regression model and incorporated into a nomogram to predict 1, 2, and 4-year OS. Multivariate analysis showed that Neo-Adjuvant Chemo-Therapy (NACT) (P< .001), LODDS (P< .001), T-stage (Pi=.001), CCI (Charlson Comorbidity Index) (P=.034) and grade (P=.003) were independent predictors of OS. The C-index of nomogram (0.740) was higher than that of the American Joint Committee on Cancer (AJCC) staging system (0.614). The bias-corrected calibration plots showed that the predicted risks were in excellent accordance with the actual risks. The results of NRI, IDI, and DCA exhibited superior predictive capability and higher clinical use of the nomogram. A simple, easy to use nomogram to predict OS in cases of MIBC has been constructed. To best of our knowledge, LODDS has been incorporated for the first time. It has superior predictive ability and higher clinical use than AJCC system. It would help the clinicians for better patient counselling, planning follow-up strategies and designing a clinical trial for newer adjuvant therapy (eg immunotherapy) in post radical cystectomy patients of MIBC.