Abstract

451 Background: Renal cell carcinoma (RCC) demonstrates heterogeneous behavior. Approximately 30% to 40% of patients with clinically localized RCC will later metastasize. Current tools are imperfect for predicting who will have distant spread of disease. C-reactive protein (CRP) levels and the Mayo Clinic SSIGN scores are widely use to predict outcomes of clear cell RCC. We investigated the predictive value of the combined use of pre-operative CRP and SSIGN score for cancer specific survival. Methods: Patients undergoing nephrectomy for clinically localized RCC from 2005 to 2010 were studied. Inclusion criteria required clear cell histology, no nodal or metastatic disease at the time of surgery, two years follow-up time, and a recorded preoperative CRP level. SSIGN score was based on postoperative data and was categorized into scores 0-2, 3-5, and ≥ 6. A cut-point of 20 mg/l was used to classify elevated preoperative CRP. Univariate analysis identified variables associated with cancer specific survival (CSS). Variables significant in univariate analysis at p < 0.05 were entered into a multivariate Cox regression model using forward stepwise regression to identify significant predictors of CSS. Results: Study criteria identified 284 patients. Of these patients, 62% were men and 13% had preoperative CRP values ≥ 20 mg/L. No patients had T4 disease, while 22% had T3, 7% had T2, and 70% had T1 disease. SSIGN scores of 0-2, 3-5, and ≥ 6 were assigned to 57%, 30%, and 13% of patients, respectively. On univariate analysis CRP ≥ 20 mg/L was a statistically significant predictor of CSS (p < 0.0004, HR 7.07, CI 95% 2.42, 20.68). Multivariate Cox Regression found preoperative CRP (p = 0.033, HR 3.65, CI 95% 1.11, 11.97) to be an independent predictor of CSS despite controlling for age at surgery (p < 0.05) and SSIGN score (p < 0.05), which were also independent predictors of CSS. Conclusions: In patients with clear cell RCC, pre-operative CRP levels in combination with SSIGN score provides additional information for predicting CSS.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call