Prognostic nomograms based on log odds of positive lymph nodes for patients with renal cell carcinoma: A retrospective cohort study

The aim of the current study is to build prognostic nomograms for patients with renal cell carcinoma (RCC) and compare the predictive performance with the American Joint Committee on Cancer (AJCC) staging system. A total of 9453 patients were identified (2005–2015) from the Surveillance Epidemiology and End Results (SEER) database. Propensity-score matching (PSM) was conducted to reduce selective bias. The matched cohort was further divided equally into the development and the validation cohort. Nomograms based on log odds of positive lymph nodes (LODDS) were formulated to predict individualized cancer-specific survival (CSS) and overall survival (OS) for RCC. Then, the performance of nomograms was internally and externally validated via the concordance index (C-index) and calibration plots. Decision curve analysis (DCA) was used to compare the clinical practicable between nomograms and AJCC staging system. LODDS was identified as an independent prognostic indicator for CSS and OS using univariate and multivariate Cox regression analyses. Two nomograms incorporating LODDS were formulated. The C-indices of the nomograms for predicting CSS and OS were 0.7561 (95% CI, 0.7356–0.7766) and 0.7140 (95% CI, 0.6936–0.7343) in the development cohort, which was higher than C-index of the AJCC staging system. The results were reproducible in the validation cohort. Moreover, internal and external calibration plots showed that the nomograms-predicted was consistent with the actual observation. Additionally, DCA demonstrated that the nomograms were superior to the AJCC staging system with obtaining more clinical net benefit. LODDS could be considered as a reliable prognostic factor for patients with RCC. Two nomograms were able to more accurately and applicable than the AJCC staging system for predicting CSS and OS. •Log odds of positive lymph nodes (LODDS) is a reliable prognostic indictor for renal cell carcinoma (RCC).•Two novel nomograms have been formulated to predict the 1-, 3- and 5-year cancer-specific survival for RCC.•Propensity score matching was performed to improve reliability of our nomograms by reducing the selection bias.•Decision curve analysis proved that our nomograms have consistently reliability and clinically practicality.