Seeing the difference: Multimodal imaging for AMD and GA

A timely and accurate diagnosis of retinal disease gives patients the best chance of achieving an optimal visual outcome. This is particularly true now, with the availability of drug treatments for geographic atrophy (GA) and inherited retinal diseases (IRDs) caused by biallelic mutations in the RPE65 gene. As more developments emerge, it is important that eye care providers recognize the phenotypical similarities between macular atrophy associated with age-related macular degeneration (AMD) and IRD-associated atrophy.

Markakis et al. undertook a retrospective clinical medical records review of patients diagnosed with AMD between 1995 and 2023 from a large multidisciplinary private ophthalmic practice in Australia to identify cases of patients diagnosed with GA without drusen.¹ They further assessed the cases for potentially missed IRD with macular atrophy and then presented flagged cases to specialists in AMD and IRD to establish a most likely diagnosis.¹ Of the 1136 cases reviewed, the possible rate of misdiagnosis observed was 1.9%, with 1.0% representing potentially missed IRDs, they reported.¹

The numbers are small, they noted, but an incorrect diagnosis can lead to inappropriate treatment. In their conclusion, the authors noted that a multimodal approach facilitated diagnosis.¹ Consistent with clinical practice, optical coherence tomography (OCT) was subjectively determined to be the modality that most often assisted in the identification and initial characterization of drusen, and fundus autofluorescence was determined to be the modality used most frequently to distinguish between AMD and IRD.

They also noted age at which macular atrophy occurred, family history of central vision loss, lifestyle factors, symmetry of fundus findings, nasal involvement, and accurate distinction between drusen and retinal flecks as important differentiating characteristics.¹ Ultimately, this allows for targeted genetic testing to be recommended.

Spotting drusen: Personal experience

When assessing a patient for the first time, I look for characteristics associated with macular degeneration (i.e., drusen). Is there focal pigment clumping? Are the clumps the shape and size to be expected with drusen? Other conditions may also have yellow deposits in the retina that are more oblong and irregular, like a fish or small bone, but drusen appear rounder.

Identifying signs of progression will help establish whether a patient truly has GA. I will discuss with the patient that there is a possibility that they have the condition, but to be sure, we need to monitor with imaging. I might have them come back in 3 months if I am concerned, or 6 months if less so. At that point, I can tell whether there are signs of progression. If the lesion looks the same every visit with no indication of macular degeneration, it is unlikely to be GA.

The right tools

Fundus autofluorescence (FAF) allows visualization of the typical hallmark hyperautofluorescence pattern along the border of a GA lesion. The technique uses the natural fluorescence of lipofuscin in the retinal pigment epithelium (RPE) and can also help in differentiating IRDs, particularly by identifying RPE changes and atrophy (Figure 1).

Several different patterns of hyperautofluorescence have been described, such as patchy, diffuse, banded, or focal. Importantly, in the FDA studies that led to the approval of pegcetacoplan injection (Syfovre; Apellis Pharmaceuticals Inc) and avacincaptad pegol intravitreal solution (Izervay; Astellas Pharma Inc), a finding of hyperautofluorescence was required for enrollment.^2,3 When it is present, the condition is more likely to be GA.

Fluorescein angiography (FA) is a useful tool for differentiating IRDs from GA, which can look similar in some cases. Stargardt disease is one of the more common of these. The condition often begins in childhood or adolescence, but its characteristic fundus appearance of flecks and localized atrophy can sometimes be mistaken for GA in older patients. Stargardt disease can be identified by “choroidal silence,” or a very dark choroid that is visible on FA (along with irregular deposits), which is due to lipofuscin masking choroidal fluorescence (Figure 2).

Pattern dystrophy is a less common group of disorders that present with a variety of pigmentary changes affecting the macula. When these dystrophies progress to significant RPE atrophy, the resulting fundus picture may mimic GA. A family history and detailed imaging are valuable tools in making the correct diagnosis. Other IRDs are much less common; yet, once they are sought, they can be found. Some of these conditions exhibit a phenotype characterized by choroidal atrophy. They do not typically have drusen, but they can have large, circular lesions that resemble GA. Genetic testing should also be performed in these situations (Figure 3).

Benefits of multimodal imaging

To diagnose and monitor AMD and GA, clinicians should use OCT and have the ability to acquire FAF images. I use a specific multimodal imaging platform (Spectralis; Heidelberg Engineering). Other widefield cameras have a filter that can produce an FAF image. It is imperative to see the pattern of hyperautofluorescence along the lesion’s edge.

Multimodal imaging enables identification of the edge of RPE atrophy in relation to the retinal ellipsoid zone (EZ). FDA studies now accept the loss of that zone as an end point to show that a drug is preserving photoreceptors. Managing GA treatment has shown that patients with a significant loss of EZ are rapid progressors. If RPE atrophy and EZ atrophy are the same on both sides of the lesion, that is likely to be a slow progressor, or it may not even be AMD. In that situation, I would monitor for change with follow-up in 3 to 6 months (Figure 4).

With the help of multimodal imaging, the data show that there are real benefits of GA treatment. We can slow down disease progression, and the effect increases over time. Patients in their second year experience a better effect than in the first year, and this improvement continues into the third year. GA is a challenging disease to treat. However, the evidence of slowed progression, along with the ability to demonstrate this to patients through images, is winning over previously skeptical eye care providers.

Conclusion

Conditions cannot be treated if they are not diagnosed, and they cannot be diagnosed if they are not seen. Advances in therapeutic options for retinal conditions once thought untreatable require that imaging capabilities are up to the task. Multimodal OCT can help identify and differentiate types of macular atrophy, enabling timely and accurate treatment and preservation of vision.

Jeremiah Brown Jr, MD, MS, FASRS

E: jbmd@retinaconsultantstexas.com

Brown founded the Brown Retina Institute in 2010 to establish a practice devoted to diagnosing, treating, and researching diseases of the retina and vitreous. In 2022, he joined Retina Consultants of Texas to expand his research goals. Brown is a past president of the San Antonio Society of Ophthalmology and has served as a voting member on the FDA Ophthalmic Devices Panel. Brown receives research funding from Apellis Pharmaceuticals, Character Biosciences, Genentech, Opthea Ltd, Outlook Therapeutics, and REGENXBIO; serves on the advisory boards for Allergan (now AbbVie), Genentech, and Outlook Therapeutics; and is also on the speakers’ bureaus for Apellis Pharmaceuticals, Genentech, and Iveric Bio.

References

1. Markakis D, Britten-Jones AC, Guymer RH, et al. Retrospective audit reviewing accuracy of clinical diagnosis of geographic atrophy in a single centre private tertiary retinal practice in Australia. Sci Rep. 2025;15(1):8528. doi:10.1038/s41598-025-90516-z

2. Syfovre. Prescribing information. Apellis Pharmaceuticals Inc; 2021. Accessed November 6, 2025. https://pi.apellis.com/files/PI_SYFOVRE.pdf

3. Izervay. Prescribing information. Astellas Pharma US, Inc; 2023. Accessed November 6, 2025. https://www.astellas.com/content/dam/astellas-com/global/en/documents/izervay_pi.pdf