Correlation Between Functional Imaging And Pathological Features Of Clear Cell Renal Cell Carcinoma

Part Ⅰ Differentiation of low-and high-grade clear cell renal cell carcinoma: tumor size versus CT perfusion parametersPurpose: To compare the utility of tumor size and CT perfusion parameters for differentiation of low-and high-grade clear cell renal cell carcinoma(RCC).Materials and methods: CT perfusion imaging and Fuhrman pathological grading of clear cell RCC were performed in 120 patients(fifteen grade 1,eighty-two grade 2,nineteen grade 3,and four grade 4).Equivalent blood volume(Equiv BV),permeability surface-area product(PS),and blood flow(BF)of tumors were measured.Maximum tumor diameter was measured at the maximum tumor area.Grade 1 and 2 were defined as low-grade group(n=97),meanwhile high-grade group(n=23)included grade 3 and 4.Comparisons of tumor size,CT perfusion parameters,and CT perfusion parameters/tumor size ratio of the two different groups were performed.Regression analysis of the each index was performed.Correlations between tumor size,CT perfusion parameters,and Fuhrman grade(grade 1,2,3,and 4)were assessed.Results: High-grade clear cell RCC had a significantly higher tumor size than low grade(P < 0.001).PS,PS/size ratio,Equiv BV /size ratio,BF/size ratio were significantly lower in high-grade than in low-grade clear cell RCC(P < 0.05).However,using regression analysis,tumor size and PS were statistically significant indexes for differentiating high-grade from low-grade clear cell RCC.Tumor size positively correlated with Fuhrman grade(r = 0.219,P = 0.016),but PS negatively did(r =-0.263,P = 0.004).There was no significant correlation between tumor size and CT perfusion parameters(P > 0.05).Conclusions: Tumor size and PS of clear cell RCC had weak correlation with Fuhrman grade.Tumor size and PS were significantly independent indexes for differentiating high-grade from low-grade clear cell RCC.Part Ⅱ Application of intravoxel incoherent motion in diagnosis of pathological grade of clear cell renal cell carcinoma:correlation with microvessel densityPurpose: To evaluate the utility of IVIM-DWI for differentiation pathological grading of clear cell renal cell carcinoma and the correlation with microvessel density.Materials and methods: 82 patients with surgical pathology-proven clear cell renal cell carcinoma(CCRCC)were included(14 grade 1,47 grade 2,15 grade 3,and 6 grade 4).Grade 1 and 2 were defined as low-grade group(n=63),meanwhile high-grade group(n=21)included grade 3 and 4.IVIM-DWI was performed with 9 b-values(0,20,50,100,200,400,600,800 and 1000 s/mm~2).True diffusivity(D),pseudo-diffusion coefficient(D*)and perfusion fraction(f)of the biexponential DWI were calculated for the CCRCC.Microvessel density(MVD)was quantified by CD34 staining.Comparisons of IVIM-DWI parameters of the two groups were performed.Regression analysis of each index was performed.Correlations between MVD,IVIM-DWI parameters and Fuhrman grade were assessed.Results: D*、D and MVD were statistically significant differences between the two groups.The D* value of the low-grade group was 51.1± 7.6mm~2 /s and that of the high-grade was 44.3± 9.5mm~2 /s(P = 0.007).D values were 1.2± 0.3mm~2 /s and 1.3± 0.3mm~2 /s(P = 0.015)respectively.MVD scores were 81.5±8.8 and 67.5±12.8 respectively.D*,D and MVD were all correlated with pathological grade of CCRCC.D* value was positively correlated(r = 0.346,P < 0.05),D value was negatively correlated(r =-0.315,P < 0.05)and MVD was positively correlated(r = 0.617,P > 0.05).ROC analysis showed the areas under the ROC curve of D and D* were 0.7 and 0.7.The thresholds of D and D* were 1.2 mm~2 /s and 44.4 mm~2 /s,the sensitivity was 70.0% and 60.7% and the specificity was 70% and 90%.D* value were positively correlated with MVD(r = 0.585,P < 0.05),but D value negatively did(r =-0.337,P < 0.05).Conclusions: IVIM-DWI imaging can provide information on different pathological grades of patients with clear cell renal cell carcinoma,and can be used for non-invasive assessment of microvessel density.Part Ⅲ Prediction models for clear cell renal cell carcinoma Fuhrman grade: based on CT imaging features and the RENAL scoreObjective: To evaluate the prediction models for the Fuhrman grade of clear cell renal cell carcinoma(CCRCC)based on CT imaging features and the RENAL score.Methods: The CT imaging features and the RENAL score of 100 cases of low-grade(I + II)and 80 cases of high-grade(III + IV)CCRCC based on Fuhrman pathological grade were analyzed.All patients were divided into training set and validation set according to the time sequence of scanning.Prediction models of CCRCC Fuhrman grade were established by using CT imaging features and the RENAL score in the training set and were verified in validation set.ROC was used to evaluate the predictive value of the prediction model.Results: In the training set,the tumor size,the corticomedullary PAC and the RENAL score were statistically different between the two groups of tumors.The tumor size of low-grade CCRCC was 29 mm and that of high-grade CCRCC was 39.47 mm(P < 0.05).The corticomedullary PAC of high grade CCRCC was 5.24±0.37 and that of low grade CCRCC was 4.83±0.26(P < 0.05).The RENAL score in the high-grade group was 8.28±1.17,The low grade group was 5.67±1(P < 0.05).Multivariate logistic regression analysis showed that the corticomedullary PAC[OR 7.09(95% CI 2.41 to 20.85)] and the RENAL score [OR 11.71(95% CI 4.19 to 32.73)]were the independent predictors of fuhrman grade of CCRCC.ROC analysis showed the diagnostic efficiency of the prediction model more than 90%.Conclusions: The prediction model using the corticomedullary PAC and the RENAL score is effective for CCRCC of Fuhrman grade.