Object: Reports on the surgical treatment of brain metastases from melanoma in a large group of patients are sparse. The goal of this paper is to review the surgical experience in a series of 91 patients with brain metastases from primary melanoma treated at a single institution.
Methods: Seven hundred eighty patients underwent resection of brain metastases at Memorial Sloan-Kettering Cancer Center between 1974 and 1994. The records of 91 (11.7%) of these patients who had melanoma were retrospectively reviewed. The median time from diagnosis of the primary melanoma to diagnosis of the brain lesion was 14.1 months. The overall median length of survival following craniotomy was 6.7 months. Fifteen patients with resected multiple metastases had shorter median survival times than 76 patients with a single lesion (5.4 months compared with 7.8 months, p = 0.12). In eight patients with cerebellar metastases the median length of survival was significantly shorter than that found in patients with supratentorial lesions (2 compared with 7 months, p = 0.03). There was no difference in length of survival between 49 patients who underwent postoperative whole-brain radiation therapy (WBRT) and 29 patients who did not (9.5 compared with 8.3 months, p = 0.67). The incidence of brain metastasis recurrences in WBRT-treated and untreated patients was similar (56% and 45.7%, respectively). Only the presence of infratentorial metastases (p = 0.0013) and unresected recurrence of brain metastases (p = 0.0003) had an impact on outcome according to a Cox regression analysis. Five patients (5.5%) died within 31 days of surgery. Overall survival rates at 1, 2, 3, and 5 years were 36.3, 18.7, 13.2, and 6.6%, respectively.
Conclusions: Although melanoma metastatic to the brain carries a foreboding prognosis, patients who do not display preoperative neurological deficits, harbor a single lesion situated supratentorially, and have no lung or visceral metastases may derive significant palliative benefit from surgical resection of brain metastases.