The Link between GLP-1 Drugs and Diabetic Retinopathy Is Not So Clear | AAO 2024

Ophthalmologists are seeing increased referrals for diabetic retinopathy in patients who are taking anti-glucagon-like peptide 1 (GLP-1) for diabetes or weight loss, according to a presentation at the annual meeting of the American Academy of Ophthalmology (AAO), which took place this weekend in Chicago. But the research about the impact of GLP-1 drugs on diabetic eye complications has been mixed.

This class of medications, including semaglutide (Ozempic for diabetes and Wegovy for weight loss and tirzepatide (Mounjaro for diabetes and Zepbound for weight loss), “have captivated society is because of their unparalleled ability to help patients in their weight loss journey,” Ehsan Rahimy, M.D., adjunct clinical assistant professor, ophthalmology, Stanford Medicine in Palo Alto, Calif., said during the presentation.

These drugs are not going away, he said. In a sampling of his own clinic, Rahimy found that 40% of patients were talking a drug in the GLP-1 class. “These are transformative medications that have a huge potential for morbidity and mortality benefit for our patients,” he said. “But need to know who to monitor more closely for the systemic health risk for diabetic retinopathy complications. These are the patients at highest risk of progression, longer duration of disease and increased decline.”

Clinical trials for GLP-1 therapies and other trials have found a link between these drugs and diabetic retinopathy complications. The prescribing information for Ozempic/Wegovy and Mounjaro/Zepbound including warnings about diabetic retinopathy complications in patients with type 2 diabetes.

Both Ozempic’s and Wegovy’s label includes data about a two-year trial that showed that rapid improvement in glucose control had been associated with worsening of diabetic retinopathy. Mounjaro/Zepbound prescribing information says tirzepatide has not been studied in patients with diabetic retinopathy but cautions that rapid improvement in glucose control has been associated with a temporary worsening of diabetic retinopathy.

In the semaglutide trials, “they reported a higher rate of retinopathy complications including, vitreous hemorrhage, blindness or conditions requiring treatment with an intravitreal agent or photocoagulation. This was noted in 3% of patients on semaglutide compared with 1.8% in the placebo group,” Rahimy said during his presentation. “This was enough to include this as a warning on the label.”

One study in particular, the SUSTAIN-6 trial, published in 2016 and found that semaglutide was associated with complications of retinopathy, including vitreous hemorrhage and decreased vision in some patients, with a hazard ratio 1.76. (A hazard ratio greater than 1 indicates increased risk of harm to patients.)

He pointed that out subsequent studies found that have many of the patients who had these complications had preexisting diabetic retinopathy or even proliferative disease at the time of entry into the study.

“Follow-up studies have excluded these patients,” he said. “Overall, with these trials there has been a relatively short follow-up period, on average, about 20 months. That’s not enough to really track long term impact of these medications.”

Rahimy also highlighted the work of he and his colleagues at Stanford to assess aggregate data in electronic medical records of patients with diabetic retinopathy who were receiving either GLP-1 agonists or SLT2 inhibitors. They assessed 5,500 patients in each of the two groups. They found an increased uptick in diabetic retinopathy or diabetic macular edema over a 36 month period.

But other studies, including one trial done by Aleksandra Rachitskaya, M.D., a vitreoretinal surgeon at Cleveland Clinic Cole Eye Institute and colleagues and published in the November/December 2024 issue of Ophthalmology Science, found that there was no association between use of semaglutide and worsening diabetic retinopathy. This was a retrospective study of 1,000 patients with type 2 diabetes who were treated at the Cole Eye Institute between 2012 and 2023. About 700 of these patients took GLP-1 receptor agonists (e.g., semaglutide, dulaglutide, exenatide). A control group of about 300 patients took SGLT-2 inhibitors (e.g., canagliflozin, dapagliflozin, empagliflozin), drugs that are comparable to GLP-1 receptor agonists at lowering blood glucose levels.

Initially, the team at Cole Eye Institute found that the percentage of patients in the study with worsening retinopathy after taking GLP-1 receptor agonists was higher. But the research team reviewed the medical records of each patient diagnosed with worsening eye disease and found coding discrepancies and not actual worsening disease.

Rachitskaya said in a news release that physicians and endocrinologists should not hesitate to prescribe these medications if they are also monitored by an ophthalmologist. “There may be patient-specific factors that contribute to eye disease, especially in patients who have a dramatic improvement in glycemic control,” she said.

Rahimy suggested during the AAO session that activation of GLP-1 receptors in human retinal stabilize vascular endothelium and may confer anti-inflammatory and neuroprotective benefits. A review article published in January 2024 found that GLP-1 agonists reduce systemic inflammation by modulating immune cell signaling and reducing pro-inflammatory cytokines.

“What we are likely seeing in the real world is the early worsening of diabetic retinopathy phenomena,” he said. “This is well described, dating back to the DCCT trial [Diabetes Control and Complications Trial], which was published in 1998 and found that insulin therapy resulted in worse diabetic retinopathy outcomes for patients. We saw this again with bariatric surgery. Patients getting gastric bypass or gastric banding had this paradoxical worsening of their diabetic retinopathy.”

Rahimy said future trials need to specifically assess risk of GLP-1 therapies and diabetic retinopathy outcomes.