Abstract
Objective: To investigate the impact of preoperative serum C-reactive protein (CRP) on clinicopathological features and prognosis in patients with upper tract urothelial cancer (UTUC) after radical nephroureterectomy (RNU). Patients and Methods: Data of 265 patients from three German centers who underwent RNU for UTUC without neoadjuvant chemotherapy between 1990 and 2012 were evaluated. Mean follow-up was 37 months (interquartile range 9-48). CRP was analyzed as a categorical and continuous variable for the prediction of recurrence-free survival (RFS), disease-specific survival (DSS) and all-cause survival (ACS) using uni- and multivariate Cox regression analyses. Results: The optimal cutoff for CRP was calculated by the Youden index at 0.90 mg/dl. Elevated CRP was significantly associated with pT3/4 and pN+ in a preoperative model including age, gender, tumor multifocality, tumor localization and the Eastern Cooperative Oncology Group Performance Status. In a multivariable Cox regression model adjusted for features significant in univariable analysis, categorized and continuous CRP levels were both independent predictors for RFS [hazard ratio (HR) 1.18, p = 0.050; HR 1.03, p = 0.012] and DSS (HR 1.61, p = 0.026; HR 1.06, p = 0.001). Continuous CRP was an independent predictor for ACS (HR 1.05, p = 0.036). Conclusions: Elevated preoperative CRP is significantly associated with aggressive tumor biology and an independent predictor for poor survival after RNU. Preoperative serum CRP represents an easily obtainable and cost-effective marker in UTUC and may help in counseling patients with regard to operative management and/or adjuvant or neoadjuvant therapies.