Regional Node Basin Recurrence in Melanoma Patients: More Common After Node Dissection for Macroscopic Rather than Clinically Occult Nodal Disease

Abstract

Background Recommended treatment for patients with sentinel lymph node (SLN)-positive melanoma has recently changed. Randomized trials demonstrated equivalent survival with close observation versus completion lymph node dissection (CLND), but increased regional node recurrence. We evaluated factors related to in-basin nodal recurrence after lymphadenectomy (LND) for SLN-positive or macroscopic nodal metastases. Methods An institutional database and the first Multicenter Selective Lymphadenectomy Trial (MSLT-I) were analyzed independently. Exclusions were multiple primaries, multi-basin involvement, or in-transit metastases. Patient demographics, primary tumor thickness and ulceration, lymph nodes retrieved, and use of adjuvant radiotherapy were analyzed. Multivariate analyses were performed to determine factors predicting in-basin nodal recurrence (significance p \u2009≤\u20090.05). Results The retrospective cohort (577 patients) showed an in-basin failure rate of 6.6% after CLND for a positive SLN and 13.1% after LND for palpable disease ( p \u2009=\u20090.001). This recurrence risk persisted after adjustment for patient, tumor, and LND factors [hazard ratio (HR) 2.32; p \u2009=\u20090.004]. In the MSLT-I cohort (326 patients), the failure rate after CLND following SLNB was 6.2%, but 10.1% after LND for palpable recurrence in observation patients. After adjustment for other factors, macroscopic disease was associated with an increased risk of recurrence after LND (HR 2.24; p \u2009=\u20090.05). Conclusion After LND for melanoma, in-basin recurrence is infrequent, but a clinically significant fraction will fail. Failure is less likely if dissection is performed for clinically occult disease. Further research is warranted to evaluate the long-term regional control and quality of life associated with nodal basin observation, which has now become standard practice.