RESUMO
BACKGROUND: The hypothesis tested in this study was whether patients with stage III metastatic melanoma confined to their sentinel lymph nodes (SLNs) had a more favorable prognosis than patients who had SLN and non-SLN (NSLN) metastases. METHODS: Patients were identified who were clinically negative in their regional basins but with lymphatic mapping were found to have positive SLNs (331 patients). All patients subsequently underwent a complete lymph node dissection of the lymphatic basin involved, and the total number of metastatic SLNs and NSLNs were documented. RESULTS: As the regional metastatic disease involves NSLNs, disease-free survival (DFS) and overall survival (OS) decreases. For patients with a total of 2 nodes positive, those with disease confined to the SLNs had a significant better prognosis (DFS and OS: P < .00001) than those in whom 1 SLN and 1 non-SLN was involved. This difference was apparent for those patients with N2 and N3 disease (2 or more nodes positive in their regional basin). A multivariate regression analysis that included Breslow thickness, ulceration, number of positive nodes, and NSLN positivity showed that NSLN positivity (P = .0019) was the most powerful predictor of DFS and OS. CONCLUSIONS: When metastatic melanoma overwhelms the SLN and involves NSLNs, the biologic behavior changes to portend a worse survival, regardless of the total node count positive. These data make the argument that the current N staging system should be changed to incorporate SLN vs NSLN involvement.