Introduction

Positive nodal status and the number of positive nodes are well known independent predictors of survival in renal clear cell carcinoma (ccRCC) patients. However, no study has ever tested whether the location of nodal metastases does affect cancer specific survival (CSS) in ccRCC patients.

Methods

Among 2,884 patients treated with nephrectomy at two European Tertiary Care Centers, 419 (14.4%) underwent open extended retroperitoneal lymph node dissection (LND) defined as the removal of hilar, side-specific (paraaortic or pre-retrocaval) and interaortocaval nodes. Cox regression analyses were used to assess the effect of the area involved (hilar vs. side-specific vs. interaortocaval) and the number of anatomical areas affected by nodal disease (1 vs. 2 vs. 3 areas) on CSS. Multivariable analyses were adjusted for age, pathologic T stage, metastases at diagnosis and Fuhrman grade.

Results

ccRCC patients who were selected for nephrectomy and extended LND (n=419) showed pT1-pT2 in 37.4%, pT3 in 56.1% and pT4 in 6.4% of patients. Mean tumor size was 9 cm (median 8.3, range 1-23). Overall, 95 patients (22.9%) showed nodal disease at final pathology. Mean number of nodes removed was 15 (range 3-58). Hilar nodes vs. paraaortic/precaval vs. interaortocaval were found positive in 11% vs. 18% vs. 12% of the cases, respectively. In 46 (11%), 26 (6.2%) and 23 (5.5%) cases 1, 2 or all retroperitoneal nodal areas were affected, respectively. Among patients with 1 positive nodal site, 26% of patients were positive only in the interaortocaval area and 54.3% only in side-specific station. Among patients with 2 positive nodal areas, 3.8% had hilar and interaortocaval areas involved but not side-specific one, and 53.8% had side-specific and interaortocaval areas involved but not hilar one. Mean follow up period of 75.1 months. CSS at 1 and 2 years resulted 58% and 40% vs. 56% and 28% vs. 37% and 30% for patients with 1, 2 or 3 areas affected by nodal disease (p=0.5), respectively. At MVA, the number of nodal stations involved by disease did not affect CSS (all p>0.5). Conversely, the presence of nodal disease in the interaortocaval area resulted an independent predictor of CSS (Hazard Ratio 1.8, p=0.05).

Conclusions

When ccRCC patients harbour nodal disease, its spreading is not systematic and can occur at any nodal station without involving the others. However, the number of anatomical areas involved by nodal invasion does not affect CSS. Conversely, presence of interaortocaval positive lymph nodes is an independent predictor of CSS in RCC patients.

Funding

None