Predictive Factors Of High-grade Pathology For Clear-cell Renal Cell Carcinoma After Renal Surgery

Objective: To investigate predictive factors of high-grade pathology for clear-cell renal cell carcinomain Chinese patients after renal surgery on preoperative imaging.To evaluate whether R.E.N.A.L nephrometry score could predict tumor pathology.Methods and materials: We retrospectively investigated the medical records of 288 patients who underwent renal surgery in the First Affiliated Hospital of Fujian Medical University for clear-cell renal cell carcinoma on preoperative imaging between January 2012 and December 2015.All patients had preoperative imaging data(CT/CTA) and had R.E.N.A.L nephrometry score.Clinical and imaging parameters were compared between low fuhrman grade and high fuhrman grade.We also improved the correlation between the clinical and imaging parameters and fuhrman grade.Results: In the 288 patients, 197 tumors(68.4%) were low fuhrman grade,91 tumors(31.6%) were high fuhrman grade, On univariate analysis, the clinical symptoms, cystic necrosis, regional lymphnode enlargement, distant metastasis, c T stage, TNM stage, operation method, largest tumor diameter(cm), nearness of the tumor to the collecting system or sinus, indicidual anatomic descriptor components significantaly differed between high and low-grade tumors.On multiple logistic regression analysis, largest tumor diameter was associated with high pathology significantly(OR=1.51; 95%CI, 1.04-2.21;P=0.03), The HFG risk of radius(2 points) were 11.59 times to radius(1 points)(95%CI, 1.87-71.87;P=0.008), The HFG risk of radius(3 points) were 17.56 times to radius(1points)(95%CI, 1.29-239.77; P=0.032, The HFG risk of exophytic/endophytic properties of the tumor(2 points) were 5.10 times to exophytic/endophytic properties of the tumor(1 points)(95%CI, 2.01-12.94;P=0.001), The HFG risk oflocation relative to the polar lines(3 points) were 26.85 times to location relative to the polar lines(1 points)(95%CI, 7.38-97.70; P<0.001).The fuhrman grade has no obvious correlation with the nearness of the tumor to thecollecting system or sinus, anterior/posterior and total nephrometry score.ROC curve was drawn with multivariate analysis regression model, largest tumor diameter(cm), R, E and L. Aear under the curve(AUC) of the ROC were 0.935(95%CI, 0.904-0.966), 0.901(95%CI, 0.866-0.937), 0.868(95%CI, 0.825-0.911), 0.511(95%CI,0.442-0.581), 0.842(95%CI, 0.791-0.892).Conclusion:The current study is the first in China to use preoperative imaging anatomical data to objecticy the relationship between tumor anatomy and pathology. Largest tumor diameter(cm), radius, exophytic/endophytic properties of the tumor, location relative to the polar lines were significant predictive factors of high-grade pathology.It can help clinician choose an appropriate operation method.