Objective: The most commonly used staging system for renal cell carcinoma (RCC) is the tumor-node-metastasis (TNM) system. In the most recent TNM edition, lymph node (LN) involvement is defined as pN0, pN1, or pN2, depending on the number of metastatic LNs (none, 1, or >1). This study evaluated the prognostic value of this classification and tried to improve its clinical impact by considering an additional parameter, that is, LN density (ratio between number of positive LNs and total number of LNs retrieved).
Methods: All pathologic reports of radical nephrectomies performed for RCC in two urologic centers between November 1983 and December 1999 were reviewed. For each patient, complete clinical and pathologic data, number of LNs removed, location and number of positive LNs, and LN density were recorded. The Kaplan-Meyer method and the log-rank test were used to calculate cause-specific survival rates and to compare survival curves, respectively.
Results: A total of 735 patients underwent radical nephrectomy. Lymphadenectomy was performed in 618 cases, and the rate of positive LNs was 14.2%. The 5-yr cause-specific survival rate of pN+ patients was 18%, with no statistically significant difference between pN1 and pN2. The average number of LNs removed was 13 (range, 1-35). The median number of LNs involved was 3 (range, 1-18). LN density ranged between 3.7% and 100% (median, 22.9%). The number of LNs removed had no impact on survival in pN+ patients. The only significant unfavorable prognostic factors were >4 LNs involved (p = 0.02) and LN density >60% (p = 0.01).
Conclusion: The results show that in RCC the current TNM stratification of positive LNs is not significantly correlated with prognosis. From our data it appears that classification as < or =4 or >4 LNs involved, supported by LN density, better reflects the impact of the disease on survival.