Comprehensive Assessment of Geographic Atrophy - Optical Coherence Tomography Classification

Leo Sheck
minute read

This post was written by Dr Aaron Yap, Ophthalmology Fellow and Dr Leo Sheck. Part 1 and part 3 can be found here. Information on consultation with Dr Sheck can be found here.

Geographic atrophy is the end stage manifestation of age related macular degeneration (AMD) that is characterised by loss of retinal pigment epithelium (RPE) and photoreceptors. Optical coherence tomography (OCT) is a fast, widely available and minimally invasive method of visualising the retina in greater detail. In this article, we will discuss the most recent literature surrounding OCT characterisation of geographic atrophy

OCT definitions of Atrophy

An international group of experts gathered as part of the Classification of Atrophy Meetings (CAM) to develop a consensus classification system for atrophy. The CAM group described 4 terms to describe atrophy in the context of AMD based on OCT images. To avoid confusion, GA will still be used to refer to complete RPE and outer retinal atrophy (cRORA) in the absence of choroidal neovascularisation (CNV). Nascent GA refers to incomplete RPE and outer retinal atrophy (iRORA) in the absence of CNV.

  1. Complete RPE and outer retinal atrophy (cRORA)
  • Region of hypertransmission of at least 250 mm in diameter in any lateral dimension
  • RPE disruption at least 250 mm in diameter
  • Overlying photoreceptor degeneration characterised by loss of the interdigitation zone, ellipsoid zone, and external limiting membrane and thinning of the outer nuclear layer. 
  • Exclusion criteria: Signs of RPE rip (scrolled RPE)
  1. Incomplete RPE and outer retinal atrophy (iRORA)
  • Region of hypertransmission, with or without basal laminar deposits 
  • Overlying photoreceptor degeneration, i.e. subsidence of the inner nuclear layer (INL) and outer plexiform (OPL), disruption of the external limiting membrane (ELM) and ellipsoid zone (EZ)
  1. Complete outer retinal atrophy
  • Absent EZ and interdigitation zone (IZ) and outer retinal thinning with intact RPE
  • Intermittent hypertransmission
  1. Incomplete outer retinal atrophy.
  • EZ disruption with regressing subretinal drusenoid deposits (SDD) and outer retinal thinning
  • Intact ELM, RPE layer and no hypertransmission.

Velaga discovered that OCT-derived measurements of GA correlate well with areas of hypoautofluorescence in FAF images. This is important because most recent studies of GA have transitioned to the use of FAF imaging, including the pegcetacoplan randomised control trials. Whilst junctional patterns on FAF correlate to the rate of GA progression, there are OCT biomarkers that are predictive of progression. It is worth noting that manual correction of  OCT segmentation errors was required in this study and may not be practical in the clinical setting. In another post, we will cover the role of artificial intelligence in OCT interpretation and prognostication.

OCT risk factors of late AMD

Certain biomarkers on OCT infer a greater risk of progression to atrophy or macular neovascularisation (MNV). This knowledge facilitates individualised counselling of the patient and provides a research framework to evaluate the effectiveness of early GA interventions.

Hyperreflective foci

Photo credit: CAM report 5, Jeff

Intraretinal hyperreflective foci are characterised by discrete, well-circumscribed, punctate lesions that match the reflectivity of the RPE. In several studies, hyperreflective foci are associated with increased risk for progression from early to advanced AMD.

Subretinal Drusenoid Deposits

Photo credit: Lei

Subretinal drusenoid deposits (SDDs), also known by reticular pseudodrusen, medium-reflective hyper-reflective mounds or cones, either at the level of the ellipsoid zone or between the ellipsoid zone and the RPE surface.

Drusen characteristics. Photo credit: CAM report 5, Jeff

Drusen associated with an acquired vitelliform lesion (AVL) progressing to cRORA (bottom left).

Drusen with hyporeflective cores and hyperreflective crystalline deposits on near-infrared imaging and OCT.

Drusenoid pigment epithelial detachment (PED) appears as a smooth, dome-shaped elevation of the RPE, with moderate, homogeneous internal reflectivity.

Lei conducted a retrospective, single-centre study to develop an OCT-based scoring system for determining the risk of progression to late AMD (atrophy or MNV formation). Based on the OCT risk factors outlined above and status of the fellow eye, 138 eyes were categorised into four groups. Logistic regression analysis showed risk of progression to late AMD was 16.4 times higher for an eye assigned to highest risk than for an eye in the second lowest risk category. However, it was subject to selection bias as higher risk patients tended to have higher visit numbers and hence, were more likely to be recruited.

HARBOR was a phase 3, multicentre randomised control trial evaluating the efficacy and safety of different doses and treatment regimens of intravitreal ranibizumab in treatment-naïve patients with neovascular AMD. Nassisi performed a subgroup analysis on the 501 fellow eyes of participants to ask the same question - which OCT features predict progression to late AMD (cRORA or MNV). At 2 years, they found hypereflective foci posed a hazard risk (HR) of 5.21, hyporeflective foci within drusen had an HR of 2.42, subretinal drusenoid deposits had an HR of 1.95, and drusen volume of >0.03 mm3 had an HR of 1.46. It is important to note that this cohort consisted of patients with CNV in the fellow eye, which is an independent risk factor; therefore, the results may not be generalisible to other study populations.

Risk factors for atrophy in choroidal neovascularisation. Photo credit: Sadda

From the HARBOR dataset, Sadda then posed another question - which OCT features in eyes receiving treatment for neovascular AMD predict future development of atrophy. Their definition of atrophy included lesions >125 microns and was divided into definite atrophy (all 3 CAM criteria) and questionable atrophy (2 criteria present). Higher central drusen volume, type 3 MNV, lower choroidal thickness, presence of nascent atrophy, reticular pseudodrusen, and increased central foveal thickness were all associated with an increased risk of macular atrophy. As we observed in type 1 membranes, subretinal fluid was protective against atrophy development. The choice treatment regimen did not influence the rate of atrophy development.

Lastly, Pasricha utilised the Age-Related Eye Disease Study (AREDS 2) dataset which captures the long term, prospective follow up of 488 eyes with intermediate AMD. Interestingly, they compared OCT features in the area of an eye where GA developed with a similar region in the same eye where GA did not develop. They found that GA precursor regions were, at various times before GA development, more likely to have features such as greater drusen volume, choroidal hypertransmission, hyperreflective foci, hyperreflective drusen substructures, and disruption of photoreceptor layer and RPE. Further to this, they were able to demonstrate the temporal sequence of these signs, noting that photoreceptor layer disruption and outer retinal layer thinning preceded other signs, thus proposing an order whereby photoreceptors are lost first, followed by the RPE, with choroidal hypertransmission occurring later.

Clinical Implications

Certain OCT biomarkers are highlighted as precursors to atrophy development in patients with AMD. We can take a step further than simply classifying AMD into “dry” vs “wet” and start to have better informed discussions with our patients about their risk of disease progression based on these biomarkers. With new treatments for GA on the horizon, these biomarkers may serve as eligibility criteria for recruitment into clinical trials. Familiarising ourselves with these features now will only help to improve our management of this common sight-threatening disease.

About Dr Leo Sheck

Dr Sheck is a RANZCO-qualified, internationally trained ophthalmologist. He combined his initial training in New Zealand with a two-year advanced fellowship in Moorfield Eye Hospital, London. He also holds a Doctorate in Ocular Genetics from the University of Auckland and a Master of Business Administration from the University of Cambridge. He specialises in medical retina diseases (injection therapy), cataract surgery, ocular genetics, uveitis and electrodiagnostics.