September 2016
Volume 57, Issue 12
Open Access
ARVO Annual Meeting Abstract  |   September 2016
Identification of subgroups in geographic atrophy using cluster analysis
Author Affiliations & Notes
  • Marc Biarnes
    Institut de la macula, Barcelona, Spain
    Barcelona Macula Foundation, Barcelona, Spain
  • Jordi Mones
    Institut de la macula, Barcelona, Spain
    Barcelona Macula Foundation, Barcelona, Spain
  • Footnotes
    Commercial Relationships   Marc Biarnes, None; Jordi Mones, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science September 2016, Vol.57, 53. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Marc Biarnes, Jordi Mones; Identification of subgroups in geographic atrophy using cluster analysis. Invest. Ophthalmol. Vis. Sci. 2016;57(12):53.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Purpose : It is currently not known if specific fundus features in patients with geographic atrophy secondary to age-related macular degeneration (GA) can be used to identify subgroups of patients with the disease that experience different rates of growth. We analyzed if there are specific subgroups in GA as seen on multimodal imaging with different prognosis in terms of progression.

Methods : We included patients aged 50 and above from a prospective, longitudinal natural history study of patients with GA with a follow-up ≥6 months whose aim was to determine risk factors associated with progression of the disease, the GAIN study. The defining features used to characterize the fundus appearance were: high number of soft drusen (detected by fundus photography -FP-); reticular pseudodrusen (RPD, as seen on infrared or spectral domain optical coherence tomography -OCT-); presence of foveal atrophy (on FP); high autofluorescence and grey color of atrophy (on fundus autofluorescence -FAF-); and subfoveal choroidal thickness (on OCT). Cluster analysis, a data-driven method to identify similar observations when group membership is not known a priori, was used to generate the different subgroups. These groups were then compared in terms of their defining features (Kruskal-Wallis test).

Results : Out of 211 screened patients, 79 eyes of 79 patients were finally included. Cluster analysis suggested an optimal number of 3 subgroups, which showed statistically significant differences between them in all defining fundus features (p≤0.03), except increased FAF. Growth rate also differed between phenotypes (1.26, 1.36 and 3.00 mm2/year in subgroups 1, 2 and 3, respectively; p=0.0001). Subgroup 1 was characterized by slow growth and the presence of foveal atrophy, high load of soft drusen and low prevalence of RPD. Subgroups 2 and 3 showed different growth rates, a thin subfoveal choroid and non-foveal involvement of atrophy; however, soft drusen and RPD were more common in subgroup 3, the one with larger growth.

Conclusions : We identified 3 major subgroups of patients with GA who differed in terms of atrophy growth and fundus features. These subgroups can help the clinician to provide an individualized prognosis and can be used for proper patient eligibility in clinical trials. They may also offer insights into disease pathogenesis.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.