Risk factors for the incidence of geographic atrophy during anti-VEGF treatment for neovascular age-related macular degeneration

Session Details

Session Title: Free Paper Session 14: AMD IV

Session Date/Time: Friday 08/09/2017 | 16:30-18:00

Paper Time: 17:12

Venue: Room 111

First Author: : I.Mantel SWITZERLAND

Co Author(s): :    A. Dirani   M. Zola   P. Parvin   S. De Massougnes   C. Bergin                 

Abstract Details


To investigate baseline and treatment factors associated with the incidence of geographic atrophy (GA) in neovascular age-related macular degeneration treated with Ranibizumab or Aflibercept.


Post hoc 2-year analysis of GA incidence in two analog prospective clinical research protocols using the same variable dosing treatment regimen with either Ranibizumab or Aflibercept.


Both prospective clinical protocols (Ranibizumab or Aflibercept) used the Observe-and-Plan interval based, variable dosing regimen. A total of 204 eyes completed the 2 year study duration. All eyes were graded for the presence or absence of GA at baseline and at 2 years on multimodal imaging. GA was defined as a dark zone on fundus autofluorescence, increased visibility of the choroidal vessels on fluorescein angiography or colour photography, sharply demarcated, and higher reflectivity of the choroid on SD-OCT with absent RPE line. The minimum diameter of GA area was 250um. In this study only eyes with no GA at baseline were included. In bilateral cases, only the right eye was included, to avoid possible bias due to inter eye correlations. Baseline and treatment factors were investigated in univariate and multivariate analysis (including drug and number of injections).


De novo GA developed in 63 (incidence 42%) out of 149 eyes (149 patients) included into the analysis (mean age 79.6 years). The univariate analysis indicated that GA incidence was associated with reduced visual acuity (p 0.003), thinner choroid (p 0.016), the presence of retinal angiomatous proliferation (p 0.000); reticular pseudodrusen (p 0.0017), retinal depigmentation (p 0.000) and intraretinal cysts (p 0.000), but the absence of subretinal fluid (p 0.0044), and fewer injections over the course of treatment (p 0.009). Additional factors (p<0.2) included into the multivariate analyses were age, hyperpigmentation, pigment epithelium detachment, the subretinal tissue complex, and the drug type. After multivariate logistic regression, the following factors were observed to be independently associated with GA incidence: lower baseline visual acuity (p 0.0006), lower number of anti-VEGF injections (p 0.011), the presence of depigmentation (p 0.0004), reticular pseudodrusen (p 0.0005) and retinal angiomatous proliferation (p 0.0011). The drug showed no significant association (p 0.21).


A range of ocular factors were shown to be associated with the incidence of GA in treated neovascular AMD. Here the drug type was not shown to significantly contribute to GA incidence. GA incidence was associated with lower retreatment numbers. The presentation of GA may be indicative of lower disease activity, requiring less aggressive treatment to attain control. Monthly retreatment, if required by the disease activity, is not a risk factor for developing atrophy. This is not entirely converse to reports about elevated GA risk in fixed monthly regimen, which represents overtreatment in most cases.

Back to previous
EURETINA, Temple House, Temple Road, Blackrock, Co Dublin. | Phone: 00353 1 2100092 | Fax: 00353 1 2091112 | Email: euretina@euretina.org

Privacy policyHotel Terms and Conditions Cancellation policy