RESUMEN
PURPOSE: To assess the response rate and toxicity of the alkylphosphocholine analogue, perifosine, in patients with metastatic or recurrent malignant melanoma. PATIENTS AND METHODS: Patients had histologically proven, unidimensionally measurable disease which was incurable by standard therapy. Prior adjuvant immunotherapy was allowed but patients had not received prior chemotherapy. Perisfosine was given orally as a loading dose of 900 mg on day 1 followed by a maintenance dose of 150 mg po on days 2-21 in a 28 day cycle. The loading dose was 300 mg on day 1 of all subsequent cycles. Tumour response was assessed every 2 cycles. RESULTS: 18 patients were accrued over 7 mos. No objective responses occurred in the 14 evaluable patients. Three patients (21%) achieved stable disease after 2 cycles and 11 had progression. Seventeen patients were evaluable for toxicity. Grade 3 or 4 non-hematologic toxicities included: diarrhea (12%), arthralgia (12%), nausea (6%), headache (6%), and fatigue (6%). No grade 3 or 4 hematological or biochemical toxicity were observed. Seventy-seven percent of patients received >or=90% of planned cycle 1 dose intensity and 58% received >or=90% of planned dose for cycle 2+. Four patients required dose reductions; treatment was delayed in 5 patients; and 5 patients missed doses because of toxicity. CONCLUSIONS: Perifosine can be safely administered when given as an initial loading dose followed by daily maintenance therapy over 28 days. Gastrointestinal toxicity is common but generally of low grade. Hematological toxicity is minimal. No objective responses were observed. No further development of single-agent perifosine is recommended in malignant melanoma.