Independent re-assessment of a high risk OCT feature based grading scale for early and intermediate AMD using Hawk and Harrier fellow eyes

Abstract

Purpose : Identification of early or intermediate (e/i)AMD eyes that are at high risk of conversion to late-stage AMD could help to enrich study populations and enable feasible clinical studies in this population. Nassisi et al. reported on a grading scale that enriched for high-risk AMD eyes based on presence of any of four OCT risk features: drusen volume ≥0.03 mm³ within a 3 mm fovea-centered circle, hyper-reflective foci, subretinal drusenoid deposits or hypo-reflective foci within drusen. The aim of this study was to determine the conversion rate to late stage AMD, of an enriched pre specified sub population of the fellow eyes of HAWK and HARRIER trials (NCT02307682, NCT02434328) compared to the overall study population at 2 years. Two independent reading centers (RC1/ RC2) were used to perform the analysis.

Methods : After data curation and removal of eyes with late AMD, 893 and 862 eligible OCT images at baseline were graded by RC1 and RC2 respectively, for the presence of at least one aforementioned risk feature (constituting the risk set). Corresponding OCT images at month-24 were then graded for progression to late AMD. Two definitions of late AMD were assessed, presence of: 1) complete RPE and outer retinal atrophy (cRORA) or exudative macular neovascularization (exMNV), or 2) cRORA or exMNV or new incomplete RPE and outer retinal atrophy (iRORA) or treatment with an anti vascular endothelial growth factor.

Results : Late AMD definition-1 corresponded to 15.5% and 19.8% conversion in the overall set and 22.6% and 27.8% of conversion in the risk-set as graded by RC1 and RC2, respectively. Whereas definition-2 corresponded to 27.3% or 30.2% conversion in the overall set and 39.8% and 41.7% of conversion to late AMD at month-24 in the risk set by RC1 and RC2, respectively. Conversion rates were found to be similar between the two reading centers.

Conclusions : Including iRORA and treatment with anti-VEGF in addition to cRORA or exMNV in the definition of late AMD increases the risk set conversion to ~40% making it feasible for design of clinical trials with the end point of conversion within a 2 year timeframe.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.

 

2-Year conversion rates in the overall set and the risk set in Hawk and Harrier fellow eyes as per late AMD definition-1 and definition-2 (blue boxes) as determined by two independent reading centers compared to Nassisi et al. 2019