A new study published in Eye investigated the associations between visual acuity (VA) and intermediate age-related macular degeneration (iAMD) subclassifications.
Give me some background.
Most patients only notice visual impairment when AMD has progressed to the advanced forms of either geographic atrophy (GA) or neovascular AMD (nAMD)—which can coexist—making VA not a sensitive functional measure until advanced stages.
However: Regulatory agencies require interventions for AMD to achieve an endpoint of at least 5-letter gain in eyes with baseline VA of 20/100 or better, and a responder is defined as a ≥ 15-letter change relative to baseline.
- Consequently: Baseline VA is a strong determinant of final visual outcome, an important consequence of transition to both advanced forms.
Now talk about the study.
In this multicenter retrospective and prospective study, investigators analyzed baseline data from the INTERCEPT-AMD study of 805 patients (mean age: 75.8 years, 35% male, 983 eyes) with iAMD in at least one eye.
Note: INTERCEPT-AMD (NCT05698316) is a large database of imaging data of iAMD that aids researchers in studying AMD progression.
- This project is a collaborative effort by investigators across sites in Europe that are members of the European Vision Institute Clinical Research Network (EVICR) to pool datasets for secondary analysis.
The subclassifications of iAMD were organized as:
- iAMD with no evidence of incomplete retinal pigment epithelial and outer retinal atrophy (iRORA) or subretinal drusenoid deposits
- iAMD with subretinal drusenoid deposits (SDDs) with no iRORA
- iAMD with iRORA with no SDD
- iAMD with iRORA and SDD
AMD characteristics of the study cohort?
The concordance in subtypes of iAMD in eyes with bilateral iAMD was 78.1% irrespective of age.
Only 7.2% of the cohort with iAMD had a VA of < 70 letters.
- Meaning: Trial designs need to consider whether the aim is to assess a 15-letter gain or loss; the baseline VA in most iAMD eyes is high, so an outcome of preventing a 15-letter loss is more appropriate.
Findings?
Eyes with iRORA and SDD had lower VA relative to eyes with no iRORA and no SDD (odds ratio [OR]: 0.98, 95% confidence interval [CI]: 0.96-1.09, P = 0.03).
- Further: The VA in the better seeing eye was significantly higher than in the worse-seeing eye.
When the iAMD eyes were classified based on the condition of the fellow eye into iAMD / iAMD, iAMD / nAMD, and iAMD / GA, there were significant differences in mean VA comparing the better- and worse-seeing eyes in bilateral iAMD within these categories.
Anything else?
Further: Eyes with nAMD in the fellow eye had reduced odds of iRORA in the study eye (adjusted OR: 0.60, 95% CI: 0.40-0.91, P = 0.015).
Increased age (sex adjusted OR: 1.07, 95% CI: 1.05-1.09, P < 0.001) and female gender (age adjusted OR: 0.75, 95% CI: 0.56-1.01, P = 0.057) were associated with SDD.
- Note: This suggests that if trial cohorts are skewed to older age-groups (i.e., 75 years or above), a high proportion are likely to have SDD.
Eyes with iRORA were three times more likely to have GA in the fellow eye (adjusted OR: 3.30, 95% CI: 2.00-5.45, P < 0.001).
Expert opinion?
“Including the worse seeing eye of bilateral iAMD group in prevention trials may result in a more homogeneous trial cohort as the VA of the worse-seeing eye is more likely to be closer to the iAMD in fellow eyes with advanced AMD,” the study authors explained.
They added that using structural markers to determine VA—such as greater outer retinal and photoreceptor thickness and volume and smaller ellipsoid zone defect area in the central 1 mm zone at baseline—as screening tools may help identify the subgroup of iAMD that is likely to benefit from a gain in VA despite a low baseline VA.
Limitations?
These included:
- The sample was obtained from retinal clinics and was consequently biased towards those undergoing treatment for nAMD
- There were minimal data on other risk factors aside from age, gender, and disease status of the fellow eye
- Other ocular imaging prognostic markers were not individually graded in the study
- VA could have been influenced by other factors not collected by clinicians
- The grading of iAMD subclassifications were performed at the site level and a centralized reading center was not utilized, which may have introduced variability in classification
- The use of AREDS supplements was not recorded
Tie it all together.
These findings suggest that baseline age, presence of SDD, and / or iRORA and fellow eye status need to be considered in future clinical trials evaluating preventative or treatment options for iAMD.