Clear Cell Renal Cell Carcinoma: The Value Of Sex-specific Abdominal Visceral Fat Measured On CT For Prediction Of Fuhrman Nuclear Grade

PurposeTo investigate whether sex-specific abdominal visceral fat composition on CT predict Fuhrman nuclear grading of clear cell renal cell carcinoma(ccRCC).Materials and MethodsOne hundred seventy-one patients(123 males and 48 females)from four hospitals(multicentre group)and 159 patients(109 males and 50 females)from the cancer imaging archive(TCIA-KIRC group)with pathologically proven ccRCC(multicentre: 124 low grade and 47 high grade;TCIA-KIRC: 79 low grade and 80 high grade)were retrospectively included.Abdominal fat was segmented into subcutaneous fat area(SFA)and visceral fat area(VFA)on CT using Image J software.The total fat area(TFA)and relative VFA(rVFA)were then calculated.Clinical characteristics(age,sex,waist circumference and maximum tumour diameter)were also assessed.Univariate and multivariate logistic regression analyses were performed to identify the association between general or sex-specific visceral fat composition and Fuhrman grade.The ROC curve is drawn to evaluate its prediction efficiency.Results1.Females with high-grade ccRCC from the multicentre group had a higher rVFA(42.4% vs.31.3%,p=0.001)than those with low-grade ccRCC after adjusting for age.There was no significant difference in males(p=0.070).2.The rVFA remained a stable and independent risk factor for females high-grade ccRCC in both the univariate(multicentre: OR 1.205,95% CI 1.074–1.352,p=0.001;TCIA-KIRC: OR 1.171,95%CI 1.016–1.349,p=0.029)and multivariate(multicentre: OR 1.095,95% CI 1.024–1.170,p= 0.003;TCIA-KIRC: OR 1.103,95% CI 1.024–1.187,p=0.010)models.3.ROC curves were plotted based on univariate(rVFA)and multivariate(age,maximum tumour diameter,waist circumference,rVFA)logistic regression analysis.The AUC values of univariate and multivariate analysis were(0.859,p=0.0001;0.880,p<0.0001),the AUC values of univariate and multivariate models in TCIA-KIRC group were(0.774,p=0.0012;0.798,p=0.0005).ConclusionSex-specific visceral fat composition has different values for predicting high-grade ccRCC and could be used as an independent predictor for females with high-grade ccRCC.