Perinephric and renal sinus fat infiltration in pT3a renal cell carcinoma: possible prognostic differences

To evaluate the influence of perinephric (PN) and renal sinus (RS) fat infiltration on cancer-specific survival beyond other prognostic factors, as the Tumour-Node-Metastasis (TNM) classification system defines stage T3a renal cell carcinoma (RCC) as infiltration of perirenal fat and/or direct infiltration of the adrenal gland. Perirenal fat invasion is differentiated into RS and PN fat infiltration, but not further classified for the prognosis. From 1990 to October 2007 106 patients with advanced RCC (T3a) were followed prospectively at one academic centre; all had a radical nephrectomy. To identify prognostic effects of PN, RS or RS + PN fat infiltration, univariable and multivariable Cox proportional hazard regression models were applied, including lymph node status, metastases, presence of sarcomatoid features and tumour necrosis, Fuhrman's grade, Karnofsky performance status, and tumour size. PN fat invasion alone was present in 58, RS in 21, and PN + RS in 27 patients. The median follow-up was 2.9 years; 49 patients died from RCC. In univariable and multivariable analyses RS fat infiltration was an unfavourable prognostic factor (adjusted hazard ratio, HR, 2.24, P = 0.019). Univariable analysis of RS + PN fat infiltration showed the worst prognostic effect (HR 3.25, P < 0.001). In multivariable analysis this combination was an independent prognostic factor (HR 2.75, P = 0.007), as was the presence of metastasis (HR 5.64, P < 0.001). In this group of RS + PN fat infiltration the 5-year cancer-specific survival was 31%. Univariable and multivariable analyses showed that the combination of RS and PN fat infiltration is an independent unfavourable prognostic marker. We recommend that perirenal fat infiltration should be further differentiated into RS fat or PN infiltration in the TNM classification. This will better stratify patient prognosis and might allow those in need of adjuvant therapy to be identified.