OBJECTIVES: I. Evaluate the clinical efficacy of leuprolide, a gonadotropin-releasing hormone agonist (GnRHa), in treating infants with vision-endangering or large, disfiguring hemangiomas.

II. Assess the impact of GnRHa on growth and development during infancy. III. Assess the safety of GnHRa in these patients.

PROTOCOL OUTLINE: This is a randomized, placebo-controlled study. Patients are stratified according to gender and by position of the lesion (periorbital vs nonperiorbital).

All patients receive oral prednisone daily for 3 weeks. Patients are then randomized to receive either placebo or leuprolide IM every 3 weeks, while continuing oral prednisone. Tumors are assessed at 1, 3, and 6 weeks. If the tumor is not responding, the leuprolide will be administered every 2 weeks. Tumors are reassessed at 3 and 6 months, at which point the treatment is stopped. Responding patients are observed every 3 weeks for 3 months. If the tumor begins to grow again, leuprolide may be administered for another 3 months. Patients whose tumors grow rapidly during treatment may crossover to the alternate therapy, repeat the leuprolide or prednisone therapy, or undergo surgical excision.

• Drug: leuprolide
• Drug: prednisone

• Presence of hemangioma meeting at least one of the following criteria: Vision-threatening because of induced astigmatism or occlusion of the visual axis or proptosis Severe anatomic distortion compromising function of an organ or creating an unacceptable cosmetic outcome Other complications, e.g., Kasabach-Merritt consumptive coagulopathy, high-output heart failure, etc.
• No lesions that are clearly regressing before therapy
• No vascular malformations other than juvenile hemangiomas