We prospectively evaluated the effect of interferon alfa-2a on subfoveal choroidal neovascularization. Eight patients with recurrent subfoveal choroidal neovascularization after surgical excision and 12 patients with subfoveal choroidal neovascularization without previous surgical excision received interferon alfa-2a (3.0 to 6.0 million U/m2 of body surface area), every other night for an average of 12 weeks. Mean follow-up was nine months. Visual acuity improved in two of 20 (10%), remained stable in seven of 20 (35%), and worsened in 11 of 20 (55%) patients. The fluorescein angiogram improved in two of 20 (10%), remained stable in three of 20 (15%), and deteriorated in 15 of 20 (75%) patients. All patients had side effects, which led to the discontinuation of therapy in five of 20 (25%) patients. Side effects included flulike symptoms, leukopenia, thrombocytopenia, increased liver enzymes, alopecia, fever, nausea, and suicidal depression. Interferon alfa-2a failed to improve visual acuity or the fluorescein angiographic appearance of subfoveal neovascular membranes in 90% of cases and was associated with marked side effects. We discourage the widespread use of interferon alfa-2a in the treatment of choroidal neovascularization.