Predictors of Geographic Atrophy Growth and Visual Decline in a longitudinal Asian cohort
In brief
Geographic atrophy expanded 0.83 mm2 in one year in an Asian cohort
In 155 eyes, geographic atrophy grew by an average of 0.83 mm2 over 12 months; larger lesions, multiple atrophy spots and subretinal deposits were linked to faster growth. Photoreceptor damage extending beyond established atrophy added only modest predictive information, while visual acuity changed little, leaving the clinical value of these imaging markers uncertain.
- Journal
- Ophthalmology. Retina (Q1)
- Published
- 23 September 2026
- Study design
- Prospective / inception cohort
- Evidence level
- Level 2, Moderate (CEBM 2b)
- Authors
- Kelvin Y C Teo, Lisa Maria George, Ali Forouhari, Dun Jack Fu, Sobha Sivaprasad, Beau Fenner, et al.
- PMID
- 42777948
- DOI
- 10.1016/j.oret.2026.09.020
Why clinicians should know about it
- Picked for Ophthalmology (paper of the day, 28 September 2026): Longitudinal GA growth and visual decline in Asian AMD cohort
Abstract
PURPOSE: To characterize GA growth and visual outcomes in Asian eyes with geographic atrophy (GA) secondary to age-related macular degeneration and to evaluate the incremental prognostic value of deep learning-derived optical coherence tomography (OCT) biomarkers. DESIGN: Prospective longitudinal observational cohort study. PARTICIPANTS: One hundred fifty-five eyes of 139 patients with GA and 12-month follow-up. METHODS: Participants underwent baseline and month-12 multimodal imaging with fundus autofluorescence (FAF) and spectral-domain OCT. GA area was quantified on FAF. A validated deep learning algorithm segmented retinal pigment epithelium layer loss (RPEL), photoreceptor degeneration (PRD), hypertransmission (HTR), and retinal pigment epithelium and outer retinal atrophy (RORA). Layer-specific excess metrics beyond RORA were derived. Multivariable mixed-effects models were constructed to identify predictors of GA growth and month-12 best-corrected visual acuity (VA), with nested models assessing incremental value beyond conventional demographic and imaging features. MAIN OUTCOME MEASURES: Twelve-month change in GA area and month-12 VA. RESULTS: Mean baseline GA area was 3.89±4.24 mm2 and increased by 0.83±0.92 mm2 at month 12 (P<0.01). In lesions with baseline area ≥2.5mm2, one year growth rate was 1.32 ± 0.96 mm2. Baseline GA area, multifocality, and subretinal drusenoid deposits (SDD) were independently associated with faster growth. When analysed by baseline GA area, SDD were associated with increased growth in eyes with smaller baseline GA area (<2.5mm2). PRD-RORA excess was independently associated with GA growth and provided incremental predictive value beyond conventional predictors (ΔR2=0.02, P=0.04). Predictors differed by baseline lesion size. Mean VA decreased minimally from 0.68±0.56 at baseline to 0.72±0.57 logMAR at 1-year (P<0.01). Central 1-mm OCT occupancy measures provided greater explanatory value for month-12 VA than 6-mm OCT occupancy. CONCLUSIONS: Photoreceptor degeneration extending beyond established atrophy provided modest incremental prognostic information beyond conventional features evaluated. Central 1-mm OCT features showed limited association with month-12 VA and may serve as complementary structural correlates of visual function.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.