Electrophysiological evaluation and 18-month follow-up of two regimens with aflibercept for neovascular age-related macular degeneration.

Purpose: To compare two aflibercept treatment regimens and the electrophysiological outcome concerning cone and rod function in age-related macular degeneration (nAMD) over 18 months.

Methods: 41 patients with treatment-naïve nAMD were randomized 1:1 to either arm 1 or 2. Arm 1 received three consecutive monthly aflibercept injections, followed by bimonthly treatment until week 52. Thereafter, a treat-and-extend (TAE) regimen was applied. Arm 2 was treated according to a TAE protocol throughout the 18-month follow-up. We assessed visual acuity (VA), central retinal thickness (CRT), injection rate and interval, and evaluated cone and rod function with full-field and multifocal electroretinography (ffERG, mERG).

Results: There were no statistically significant differences in mean baseline VA, lesion type, age, gender, or symptom duration between the two arms. During the 18-month follow-up, mean VA improved in arm 1 (n = 19) from 63.5 ± 10.5 to 69.1 ± 9.2 letters; p = 0.098; and in arm 2 (n = 20) from 66.8 ± 13.6 to 73.9 ± 9.0 letters; p = .002. In both arms, mean CRT was significantly reduced; p < 0.000. At month 18, we found no significant difference in the number of injections or injection intervals between groups. Arm 1 had received 11.3 ± 1.7 injections vs. 10.9 ± 2.0 in arm 2. The mean injection interval was 9.2 ± 3.4 weeks vs. 9.5 ± 3.1, with 52% (n = 10) on the maximum 12-week interval in arm 1, and 50% (n = 10) in arm 2. The combined rod-cone a-wave amplitude significantly decreased over time; p = 0.043. The isolated rod b-wave amplitude showed a statistically significant decline; p = 0.026. The overall mERG amplitude and implicit time remained unchanged over time; p = 0.878 vs. p = 0.922. The central ring 1 mERG amplitude improved; p = 0.041, with an unaffected implicit time.

Conclusions: After 18 months, both treatments arms have received a similar number of injections at comparable intervals. Electrophysiological evaluation shows no signs of toxicity concerning cone function. But ffERGs for the combined and isolated rod response have declined, possibly reflecting either toxic effects of the drug to rods or the natural course of the disease itself.