Prognostic indicators for survival in renal cell carcinoma with venous thrombus and development of predictive nomograms

BACKGROUND: Previous predictive models of prognosis of patients with renal cell carcinoma (RCC) and venous tumour thrombus (VTT) didn’t included patients have not undergoing radical nephrectomy (RN). We analysed both patients receive RN or not to investigate the prognostic factors of survival for patients with RCC and VTT comprehensively. METHODS: The clinical data of patients with RCC and VTT diagnosed from 2000–2018 in the Surveillance Epidemiology and End Results (SEER) database were downloaded and compared with the clinical data of patients with VTT admitted to the Department of Urology of the Tongji Hospital (TJH) from 2004–2020. The matched cases were divided into a training set and a validation set. The training set was used to establish nomograms based on key prognostic factors. The reliability of the nomograms for predicting the survival of patients in the training set, those in the validation set and TJH patients and was evaluated by C-indexes, ROC curves and calibration curves. RESULTS: Multivariate Cox regression analysis identified nine prognostic factors for overall survival (OS): age, tumour size, histologic classification, nuclear grade, location of VTT, N stage, M stage, surgery, and systemic treatments (P<0.001). Nomograms for OS and cancer specific survival (CSS) were established based on key prognostic factors obtained from the multivariate analysis. The C-indexes of the nomogram for predicting OS in the training set, validation set, TJH cohort were 0.762 (95% CI: 0.746–0.778), 0.718 (95% CI: 0.687–0.749), and 0.819 (95% CI: 0.745–0.893), respectively. The calibration curves are all close to a straight line with a slope of 1. Based on the ROC curves, the nomograms had greater areas under the curve (AUCs) than the tumor, node and metastasis (TNM) staging system in predicting the 3-year OS and CSS. All three validations showed that the nomograms established based on key prognostic factors have reliable accuracy in predicting the survival of both TJH and SEER patients who developed RCCs with VTT. CONCLUSIONS: Beside the location of VTT, the tumour size can also predict the survival of patients with RCC and VTT. Nomograms based on key prognostic factors can predict the survival of patients from both America and central China with reliable accuracy.