Using Retinal Images to Predict Risk of Eye and Systemic Diseases

Routine ophthalmic procedure may help in preventive and diagnostic medicine

Different colored lines showing thickness layers on a black background

Optical coherence tomography shows thickness of nine retinal layers across all individuals studied. Image: Seyedeh Maryam Zekavat and Saman Doroodgar Jorshery, et al., Science Translational Medicine

At a glance:

  • Study finds thickness of retinal layers can help predict a person’s risk of developing eye, neuropsychiatric, heart, metabolic, and lung diseases.
  • Researchers identify more genes that determine retinal health, which could inform development of therapies for eye disease.
  • Delving further into the link between retinal thickness and disease could help ophthalmologists work with other specialists to better prevent disease and personalize treatment.

By combining retinal imaging, genetics, and big data, physician-researchers from Harvard Medical School, Mass Eye and Ear, Massachusetts General Hospital, and the Broad Institute of MIT and Harvard have found that they can estimate how likely a person is to develop eye and systemic diseases in the future.

They found significant associations between the thinning of different retinal layers and increased risk of developing ocular, cardiac, pulmonary, metabolic, and neuropsychiatric diseases and identified genes associated with retinal layer thickness.

Their findings are published in Science Translational Medicine.

Get more HMS news here

“We showed that retinal images could be used to predict the future risk of both ocular disease and systemic disease,” said Seyedeh Maryam Zekavat, HMS clinical fellow in ophthalmology at Mass Eye and Ear, a postdoctoral scholar at the Broad Institute, and co-first author of the study with Saman Doroodgar Jorshery, an HMS postdoctoral fellow at Mass General and the Broad Institute.

“This could potentially help with disease prevention — if we know from someone’s retinal image that they are at high risk of developing glaucoma or cardiovascular disease in the future, we could refer them for follow-up screening or preventive treatment,” he said.

Because of its position behind the transparent structures of the eye, the retina is easy to visualize and image noninvasively, and retinal imaging is already a routine procedure in ophthalmology.

The new study uncovers possibilities for preventive medicine and cross talk between ophthalmology and other areas of medicine.

Previous studies have shown links between retinal health and health conditions, such as aging, cardiometabolic diseases such as diabetes and hypertension, and neurological diseases such as dementia, stroke, and multiple sclerosis.

“We’ve come to realize recently that there is a lot more information that we can get from our retina images than we thought was possible,” said senior author Nazlee Zebardast, HMS assistant professor of ophthalmology and director of glaucoma imaging at Mass Eye and Ear.

“It’s really exciting to be able to see that these images, which are obtained without having to do any sort of invasive procedure, are associated with so many systemic conditions, both at a genetic level as well as an epidemiologic level,” she said.

To identify associations between retinal health and disease risk, and to identify genes associated with retinal health, the researchers analyzed data from 44,823 UK Biobank participants who underwent optical coherence tomography (OCT) imaging of the retina, genotyping, and baseline measurements of health in 2010 and were then followed for disease development over an average of 10 years.

Unlike previous studies that searched for genes associated with overall retinal health, this study delved deeper into the role of the different cell layers that make up the retina.

“Each layer of the retina is made up of different types of cells with diverse structures and functions, and we show that the thicknesses of these different layers are associated with different conditions,” says Zebardast, who is also an associated scientist at the Broad Institute.

The study also provides insight into the genes and biological pathways that determine retinal health, which could be leveraged to develop future therapies, the researchers say. Altogether, the team identified 259 genetic loci associated with retinal thickness.

One insight from this work is the finding that multiple systemic health conditions including poor heart, metabolic, lung, and kidney function are linked to thinning of the photoreceptor segment of the retina, though further research would be needed to confirm causality. Future studies should also aim to replicate the study’s methods in more diverse populations and different age groups, since participants in the UK Biobank were predominantly white and 40 to 70 years old at baseline.

Retinal OCT imaging is already a standard clinical procedure in ophthalmology at Mass Eye and Ear and elsewhere, but the authors say that their results suggest its use could be widened.

Further work on the connection between ocular and cardiometabolic health will enable understanding of its clinical utility, and the researchers are extending this line of research along with co-author Pradeep Natarajan, HMS associate professor of medicine, director of preventive cardiology at Mass General, and an associate member in the Cardiovascular Disease Initiative at the Broad Institute.

“Patients come to us for their eye health, but what if we could tell them more than that,” said Zebardast. “What if we could use someone’s retinal images to tell them, you seem to have a high risk of having high blood pressure, so maybe you should get screened or maybe your primary care doctor should know about that.”

The authors have developed an online user interface for all of their findings on the Ocular Knowledge Portal to enable researchers to explore associations between retinal layer thickness, disease, and genetics.

Funding, authorship, disclosures

Additional authors include Franziska G. Rauscher, Katrin Horn, Sayuri Sekimitsu, Satoshi Koyama, Trang T. Nguyen, Maria C. Costanzo, Dongkeun Jang, Noël P. Burtt, Andreas Kühnapfel, Yusrah Shweikh, Yixuan Ye, Vineet Raghu, Hongyu Zhao, Marzyeh Ghassemi, Tobias Elze, Ayellet V. Segrè, Janey L. Wiggs, Lucian Del Priore, Markus Scholz, and Jay C. Wang.

This study was supported by the National Eye Institute; Hassenfeld Scholar Award from Mass General; National Heart, Lung, and Blood Institute; and LIFE Leipzig Research Center for Civilization Diseases.

Natarajan reports research grants from Allelica, Apple, Amgen, Boston Scientific, Genentech/Roche, and Novartis; personal fees from Allelica, Apple, AstraZeneca, Blackstone Life Sciences, Creative Education Concepts, CRISPR Therapeutics, Eli Lilly & Co., Foresite Labs, Genentech/Roche, GV, HeartFlow, Magnet Biomedicine, Merck, and Novartis; scientific advisory board membership of Esperion Therapeutics, Preciseli, and TenSixteen Bio; scientific cofounder of TenSixteen Bio; equity in MyOme, Preciseli, and TenSixteen Bio; and spousal employment at Vertex Pharmaceuticals, all unrelated to the present work. Wiggs has received grant support from Aerpio and served as a consultant for Allergan, Avellino, Editas, Maze, and Regenxbio outside of the present work, and has served as a consultant for Zeiss. Other authors report no conflicting relationship.

Adapted from Broad Institute and Mass Eye and Ear news releases.