Purchase this article with an account.
Delu Song, Gui-Shuang Ying, Juan E Grunwald, Ebenezer Daniel, Maureen G Maguire, Yafeng Li, Daniel F Martin; Incidence and Features of Geographic Atrophy in Fellow Eyes in the Comparison of Age-related Macular Degeneration Treatment Trials (CATT). Invest. Ophthalmol. Vis. Sci. 2019;60(9):1144.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To describe the incidence and characteristics of geographic atrophy (GA) in fellow eyes of patients with unilateral choroidal neovascularization (CNV) treated with anti-VEGF injections in the Comparison of Age-related Macular Degeneration (AMD) Treatment Trials (CATT).
Participants with untreated neovascular AMD in their study eye were randomly assigned to ranibizumab or bevacizumab (monthly or PRN). After 2 years of treatment, participants were released from the treatment protocol and recalled for follow up examination at 5 years. Patients at baseline with no GA in either eye and no CNV in the fellow eye were included. Bilateral, stereoscopic color fundus photographs were assessed for GA by two masked graders at baseline, years 1, 2 and 5. Proportional odds logistic regression was used to identify factors associated with the number of eyes developing GA.
Among 599 eligible participants, the cumulative incidence in fellow eyes vs study eyes is 0.05 vs 0.07 by year 1, 0.10 vs 0.13 by year 2, 0.33 vs 0.34 by year 5 (p=0.12). For participants assigned to ranbizumab treatment, GA incidence in the fellow eye was lower than that of study eye (0.05 vs 0.10 by year 1, 0.11 vs 0.16 by year 2, 0.32 vs 0.40 by year 5, p=0.009), while GA incidence in the fellow eye was similar for both drugs used for the study eye (p=0.91) (Figure 1). Among 196 (32.7%) participants with incident GA in either eye during follow-up, GA was bilateral in 76 (38.8%) participants and detected in both eyes at the same visit in 48 (63.2%) participants. In multivariate analysis, female gender (OR=1.69, p=0.009), number of eyes with pseudodrusen (OR=1.60, p<0.0001) and number of T risk alleles in Toll-like receptor 3 (OR=1.42, p=0.03) were associated with higher number of eyes with GA (Table 1).
Among CATT participants without CNV in the fellow eye at baseline, one third of participants developed GA in their fellow eye by 5 years, similar to the GA incidence in the study eye. Among participants with ranibizumab treatment in study eyes, GA incidence was lower in the fellow eye than the study eye. Female gender, pseudodrusen, and risk allele C in Toll-like receptor 3 were risk factors for GA.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
This PDF is available to Subscribers Only