RESUMO
BACKGROUND: Nodal staging with sentinel node biopsy (SNB) and completion lymph node dissection (CLND) provides prognostic information to patients with melanoma and their physicians. It is not known whether the timing of CLND is associated with survival outcome and/or CLND tumour load. This study investigated whether CLND timing is associated with CLND tumour load, disease-free survival (DFS) and/or melanoma-specific survival (MSS). METHODS: A retrospective cohort of patients with SNB-positive melanoma from nine European Organisation for Research and Treatment of Cancer (EORTC) Melanoma Group centres undergoing surgery between 1993 and 2009 were examined. Patients were selected based on availability of CLND and follow-up data. The CLND interval was defined as the number of days between diagnosis and CLND. Patient and tumour characteristics were collected. Five-year DFS and MSS rates were calculated. Cox and logistic regression analysis were performed, adjusting for known prognostic/predictive indicators. RESULTS: A total of 784 patients were included in the study. Their median age was 51 (i.q.r. 40-62) years, and 418 patients (53·3 per cent) were men. Median Breslow thickness was 3·0 (i.q.r. 2·0-5·0) mm, and 148 patients (18·9 per cent) had a residual tumour load. Median CLND interval was 84 (i.q.r. 65-105) days. Five-year DFS and MSS rates were not significantly different for patients operated on with a median CLND interval of less than 84 days and those with an interval of at least 84 days (DFS: 54·2 versus 53·3 per cent respectively; MSS: 66·9 versus 65·1 per cent). In a multivariable Cox model, CLND interval was not a significant prognostic indicator. CLND interval was negatively correlated with identification of positive non-sentinel nodes, but following adjustment for known risk factors this effect was no longer found. CONCLUSION: The time interval between diagnosis of melanoma and CLND did not influence CLND tumour load, DFS or MSS.