A novel tumor grading scheme for chromophobe renal cell carcinoma: prognostic utility and comparison with Fuhrman nuclear grade.

Chromophobe renal cell carcinoma (RCC) is a histologic subtype of RCC that portends a favorable prognosis. It is controversial whether the Fuhrman nuclear grade of chromophobe RCC has prognostic utility. Irregular nuclei, prominent nucleoli, and nuclear pleomorphism are inherently present in chromophobe RCC. Hence, the Fuhrman nuclear grade is higher even though the majority of these tumors have a favorable outcome. In this study, the prognostic utility of a novel 3-tiered tumor grading system in which the innate nuclear atypia of chromophobe RCC was discounted, herein referred to as chromophobe tumor grade from a series of 124 chromophobe RCC, was compared with Fuhrman nuclear grade. Chromophobe tumor grade is based on the assessment of geographic nuclear crowding and anaplasia. The Fuhrman nuclear grade distribution between the tumors was grade 1 (1%), grade 2 (19%), grade 3 (74%), and grade 4 (6%), whereas the chromophobe tumor grade distribution was grade 1 (74%), grade 2 (16%), and grade 3 (10%). Neither Fuhrman nuclear grade nor chromophobe tumor grade was significantly associated with patient's age or sex and chromophobe RCC cell types, but both showed a significant association with tumor size. Both Fuhrman nuclear grade and chromophobe tumor grade showed statistically significant positive associations with broad alveolar growth, necrosis, vascular invasion, and with pathologic stage; however, all these associations tended to be dictated by tumors with sarcomatoid change. When tumors with sarcomatoid change were excluded, a strong positive association persisted between chromophobe tumor grade and pathologic stage. In contrast, there was no such association between Fuhrman nuclear grade and stage in nonsarcomatoid chromophobe RCCs. Characterizing aggressive chromophobe RCC with aggressive behavior with the time from surgery to first occurrence of metastasis, local recurrence, or death owing to disease, we found that both Fuhrman nuclear grade and chromophobe tumor grade were highly associated with adverse outcome. However, as with the pathologic stage, only a significant association between chromophobe tumor grade and outcome was retained among nonsarcomatoid chromophobe RCCs. Multivariable Cox regression analysis also tended to support chromophobe tumor grade rather than Fuhrman nuclear grade as an independent predictor of adverse outcome, controlling for other univariably significant risk factors [estimated relative hazard=3.68 (P=0.026) vs. 1.86 (P=0.42)]. In conclusion, the novel chromophobe tumor grading system proposed herewith provides superior prognostic value to that of the Fuhrman nuclear grade in chromophobe RCC and will potentially help stratify patients of chromophobe RCC who are at a greater risk of disease progression.