Macular Atrophy in Neovascular Age-Related Macular Degeneration with Monthly versus Treat-and-Extend Ranibizumab: Findings from the TREX-AMD Trial

Purpose To compare the enlargement rate of macular atrophy (ERMA) in eyes treated with ranibizumab monthly or using a treat-and-extend (TREX) regimen for neovascular age-related macular degeneration (AMD) or fellow control eyes, as well as analyze risk factors for macular atrophy (MA) development and progression. Design Eighteen-month, multicenter, randomized, controlled clinical trial. Participants Sixty patients with treatment-naive neovascular AMD in 1 eye randomized 1:2 to monthly or TREX ranibizumab. Methods Patients' study and fellow eyes were followed for 18 months using spectral-domain optical coherence tomography (SD OCT) and fundus autofluorescence (FAF) imaging. The MA was quantified on FAF images using Heidelberg Region Finder software (Heidelberg Engineering, Heidelberg, Germany), with suspected areas of atrophy confirmed by SD OCT and infrared reflectance imaging. For eyes without baseline MA yet developed MA by 18 months, intervening visits were assessed to determine the first visit at which MA appeared to define progression rates. Foveal choroidal thickness (FCT), subretinal hyperreflective material (SHRM), and pigment epithelial detachment (PED), were assessed at baseline to determine whether they influenced MA progression. Main Outcome Measures Mean ERMA at 18 months. Relationship between visual acuity and MA, and the baseline risk factors for ERMA were also assessed. Results The final analysis cohort included 88 eyes in 3 groups: monthly (n = 19), TREX (n = 30), and control fellow eyes (n = 39). Mean ERMA over 18 months was 0.39±0.67 (monthly), 1.1±1.9 (TREX), and 0.49±1 mm 2 (control, P  = 0.12). Mean ERMA per group among the 40.9% (n = 36) of baseline patients with MA was 0.9±1, 1.9±2.2, and 1±1.3 mm 2 , respectively ( P  = 0.31). The incidence rate of MA in the 3 groups was 40%, 0%, and 8.3%, respectively. Mann–Whitney U test revealed a statistically significant association between baseline FCT (127±46 vs. 155±55 μm, P  = 0.01) and SHRM thickness (106±131 vs. 50±85 μm, P  = 0.02) on MA. In eyes with no baseline MA, presence of SHRM, SHRM, and PED thickness, and presence of baseline hemorrhage were all significant predictors of new MA development ( P  = 0.04, 0.01, 0.04, 0.004, 0.01, respectively). Conclusions Ranibizumab did not show a statistically significant influence on new MA development in eyes with neovascular AMD, whether dosed monthly or per TREX regimen. The FCT, SHRM thickness, and hemorrhage at baseline were all significant predictors of new MA.