Construction of a large sample Chinese patient-based survival nomogram for patients with clear cell renal cell carcinoma / 中华泌尿外科杂志

ABSTRACT

Objective:To construct an effective survival nomogram for patients with clear cell renal cell carcinoma (ccRCC) using a large sample sized Chinese dataset, which can be used to predict individual 3- and 5-year overall survival (OS) precisely.Methods:The data of 672 ccRCC patients received operation diagnosed at Xijing Hospital from January 2012 to December 2016 were retrospectively analyzed. There were 467 males and 205 females. Their median age was 56 years old (ranging 23-83 years old). There were 327 patients with tumor on the left kidney and 345 patients with tumor on the right kidney. Clinical stageⅠ, Ⅱ, Ⅲ, Ⅳ were 584, 47, 19 and 22 cases, respectively. At the time of diagnosis, 504 patients were asymptomatic and 168 patients were symptomatic. Preoperative alkaline phosphatase was 80 (41-240) U/L. Preoperative serum albumin was 44.8 (30.5-59.8) g/L. Preoperative neutrophil absolute value/lymphocyte absolute value (NLR) was 2.25 (0.81-9.89). Preoperative platelet count was 205 (82-589)×10 9/L. Preoperative creatinine was 97 (55-230) μmol/L. Radical nephrectomy was performed in 420 (62.5%) patients and partial nephrectomy was performed in 252 patients. Cox multivariate analysis was used to determine the independent predictors of the postoperative OS. Then, the nomogram was constructed using R software, which integrates all independent predictors according to the coefficients in the multivariate analysis. Moreover, the performance of the nomogram was evaluated using the consistency index (C-index) and the calibration plots. Results:Cox multivariate analysis results showed that age at diagnosis ( P<0.001), clinical TNM stage ( P<0.001), preoperative NLR ( P=0.012), preoperative alkaline phosphatase ( P=0.002) and preoperative albumin ( P<0.001) were the independent predictors of postoperative OS in ccRCC patients. The nomogram established by integrating these five factors had a good discriminatory ability (C-index=0.819, 95% CI 0.813-0.825), and the calibration plots showed that excellent agreements between the nomogram prediction and the actual observation were achieved. Conclusions:Based on a large sample sized Chinese dataset, this study established an effective survival model for patients with ccRCC and good performance of the nomogram was demonstrated by internal validation. Our nomogram can help urologists to predict individual 3- and 5-year OS accurately for Chinese ccRCC patients.