Temozolomide (TMZ) as prophylaxis for melanoma brain metastases (BrM): Results from a phase III, multicenter study.

20014 Background: Almost 170,000 patients developed CNS metastases in 2007 and melanoma is the 4 leading cause. Some authors reported a possible role for TMZ in preventing BrM. In 2002, the IMI started a randomized study with the primary end point of evaluating whether TMZ is able to reduce the expected incidence of BrM at 12 months from 30 % to 12 % (α= 5%, β= 80%, two tailed test).,150 pts with untreated metastatic melanoma were randomly assigned (1:1) to receive cisplatin 75 mg/m day 1, and IL-2 3 MU X 2/d on days 9-18 plus TMZ 200 mg/m/d on days 1-5, or DTIC 800 mg/m on day 1 every 28. (arm CTI=75 pts, arm CDI=75). Brain CT or MRI were planned every four months in the first 2- years and every six months after.,147 patients (CTI=73 pts, CDI=74) were valuable. The two arms were well balanced for sex, PS, LDH serum level, metastatic site. The overall incidence of BrM was 31% in CTI and 40% in CDI arm (p=ns) with a median time to BrM of 15 vs 13 mos respectively. The 1-year BrM progression free survival was 57% in CTI vs 53% in CDI. Considering only responsive pts (13 in CTI and 16 in CDI), the overall incidence of BrM was 55% with a median time to BrM of 23 mos in CTI and 21 in CDI arm. The median PFS and OS for all patients were 4 and 8.4 mos in CTI and 4 and 9.5 mos in CDI.,The BrM incidence was higher than previously reported, especially in responsive patients. The relative 22% reduction of BrM occurrence obtained using a TMZ-based CT was less than planned and therefore did not translate in a significant benefit. No significant financial relationships to disclose.