Radiotherapy for subfoveal choroidal neovascularization in age-related macular degeneration: a randomized clinical trial

PURPOSE: To report results of 18-month follow up of external beam radiation therapy with photons for subfoveal classic or occult choroidal neovascularization (CNV) in age-related macular degeneration (ARMD). DESIGN: Randomized clinical trial. METHODS: A total of 161 patients with subfoveal CNV in ARMD were recruited in a prospective double-masked study. The posterior pole of the afflicted eye was given 1 Gy (4 x 0.25 Gy) in the control group and 8 Gy (4 x 2 Gy) or 16 Gy (4 x 4 Gy) in the treatment groups. At the time of treatment, and 6, 12, and 18 months post treatment, best-corrected visual acuity (BCVA), reading ability, and CNV size were measured. RESULTS: At the completion of the study 150 (93.2%), 139 (86.3%), and 137 (85.1%) patients were followed for 6, 12, and 18 months, respectively. The mean number of lines lost in the BCVA was -1.69, -2.2, and -3.23 in the 1 Gy group; -0.94, -1.25, and -1.73 in the 8 Gy group; -0.51, -0.67, and -1.93 in the 16 Gy group. The difference was significant after 12 months (P =.016 for 8 Gy vs. 1 Gy; P =.006 for 16 Gy vs. 1 Gy), and 18 months (P =.011 for 8 Gy vs. 1 Gy; P =.05 for 16 Gy vs. 1 Gy). The patients with classic CNV, or with an initial distance visual acuity >or=20/100, benefited more from treatment. A significant difference was not found between control group and treatment groups in the reading ability and in the CNV size. No radiation-associated side effects were reported thus far. CONCLUSION: The number of lines lost in the BCVA was less in the 8 Gy and 16 Gy treatment groups than in the control group during the complete follow up examination. Radiation therapy with 8 Gy and 16 Gy, without showing any difference in efficacy, resulted in a near stabilization of the BCVA in patients with subfoveal classic or occult CNV in ARMD. Further studies are necessary to determine the significance of repeated radiotherapy series with a dose of 8 Gy to improve the effect on the CNV size and thereby to prolong stabilization of distance visual acuity.