SAN FRANCISCO – A large observational registry study of almost 100,000 eyes has found that the diabetes drug semaglutide, a GLP-1 agonist recently approved for weight loss, does not worsen the progression of potentially vision-threatening diabetic retinopathy in the long term in patients taking the drug. However, the researchers said, the findings do not obviate the need for providers to have a conversation about the potential risks to vision posed by the drug.
“In patients that have either no or early or relatively nonadvanced diabetic retinopathy, the absolute risk of having a worsening in their retinopathy is variable,” Zeeshan Haq, MD, a retina specialist at Retinal Consultants of Minnesota, told this news organization. Dr. Hag presented the findings Nov. 3 at the annual meeting of the American Academy of Ophthalmology.
“Based on this preliminary evidence and what we know so far, it suggests that there is a risk of worsening, but it’s quite low for most patients, and so a conversation needs to be had between anyone considering prescribing the drug, such as a general practitioner or a nurse practitioner, and that patient’s optometrist or comprehensive ophthalmologist or retina specialist.”
Methodology and results
Dr. Haq reported on a retrospective case series of 96,462 eyes from the Intelligent Research in Sight (IRIS) registry. Patients had type 2 diabetes and began taking injectable semaglutide between January 2013 and December 2021.
The study evaluated eyes with three levels of retinopathy:
- No retinopathy or background retinopathy (71.8%).
- Mild or moderate nonproliferative diabetic retinopathy (NPDR) (18.4%).
- Severe NPDR or proliferative diabetic retinopathy (PDR) (9.8%).
In eyes with no or background retinopathy, 1.3%, 1.2%, 1.6%, and 2.2% experienced a worsening in status of the condition at 3, 6, 12, and 24 months, respectively.
In eyes with mild or moderate NPDR, 2.4%, 3%, 3.4%, and 3.5% showed worsening retinopathy at the respective time intervals.
Improvement of retinopathy rather than worsening was evaluated in the eyes with severe NPDR or PDR. At 3, 6, 12, and 24 months, improvement was observed in 40%, 37.8%, 47.7%, and 58.7% of these eyes, respectively.
Most patients were aged 51-75 years (77.2%), female (55.0%), and White (63.8%).
The study found low rates of the following complications across the same time intervals: vitreous hemorrhage (from 0.1% to 0.15%); traction retinal detachment (0.02% to 0.05%); and neovascular glaucoma (0.03% to –0.04%), Dr. Haq reported.
Dr. Haq noted that understanding the possible consequences that semaglutide has on vision is important as the drug becomes more widely available for both diabetes and weight control. The Centers for Disease Control and Prevention reports that 37.3 million people in the United States have diabetes; 28.7 million cases have been diagnosed, and 8.5 million are undiagnosed.
Clinical implications
“Any patient in the United States with diabetes has to undergo screening for diabetic eye disease, and so they’re usually plugged into the eye-care system,” Dr. Haq said. “But if they’re going to be starting this drug and they don’t have any existing diabetic retinopathy, the discussion should be had between their doctor and the eye-care provider, and if they do have a history of DR, an evaluation with the eye-care provider should probably happen upon starting the drug.”
Vaidehi Dedania, MD, a retina specialist at NYU Langone Health in New York, said the findings underscore the importance of counseling patients who are taking semaglutide about potential vision outcomes.
“We know that when patients get rapid control of their diabetes, their diabetic retinopathy can worsen in the short term, although it always ends up doing better in long term anyway,” Dr. Dedania said in an interview. “We always educate our patients that if they get control of the diabetes to not feel discouraged if their diabetic retinopathy worsens despite getting good control, because we know in the long run it always get better.”
The new findings, however, may have masked some worsening of retinopathy because of how the researchers categorized the condition. “It’s hard to assess changes within a designation because they’re so broad,” Dr. Dedania said.
She also noted potential limitations with the IRIS database itself. “The data collected from it are not always as complete as you might need for the purpose of understanding this, so that’s a limitation,” she said. While the high number of patients is a strength of the study, she added, “I still think the limitations are pretty significant.”
Dr. Haq has disclosed no relevant financial relationships. Dr. Dedania has relationships with Genentech/Roche and Regeneron Pharmaceuticals.
A version of this article first appeared on Medscape.com.