Research recently published in JAMA Ophthalmology investigated whether diabetic patients’ use of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) may increase their risk for developing neovascular age-related macular degeneration (nAMD).
Let’s start with GLP-1 RAs.
What they are: A class of medications typically used to treat type 2 diabetes mellitus (T2D) and obesity (see their other recommended uses, including for managing cardiovascular risk and chronic kidney disease).
- What they do: Mimic the action of the GLP-1 hormone (which helps to increase natural insulin release) to trigger insulin release, reduce glucose (lower serum glucose levels and overall production in the liver), and manage patients’ metabolism.
- Plus: They also reportedly contain anti-inflammatory effects to potentially treat inflammatory diseases.
Examples of FDA-approved GLP-1 RAs available on the U.S. market include Wegovy, Rybelsus, and Ozempic—all of which have semaglutide as the main ingredient, although Wegovy is only approved for weight loss.
- See here for other examples.
So how does this trace back to the eye?
While GLP-1 RAs have their benefits, researchers noted that such use also comes with adverse events—including in the eye.
Their supporting evidence: Results from two phase 3a trials (PIONEER 6 and SUSTAIN 6; both of which were exploring GLP-1 RAs’ cardiovascular effects among diabetic patients) that found an increased risk for diabetic retinopathy.
And beyond this: Investigators reference prior research that posited the rapid reduction in blood glucose levels triggered by GLP-1 RAs leads to a hypoxic state of the retina that promotes further abnormal angiogenesis.
Yikes. Now let’s talk about this new research.
A population-based, retrospective cohort study was conducted in Ontario, Canada, over a nearly 3-year period (January 2020 to November 2023), followed by a 3-year follow-up period.
The specifics: Researchers used linked administrative health and demographic data sourced from the health records of diabetic patients (aged 66+) via specific International Classification of Disease (ICD) diagnostic codes (ICD-9 and ICD-10).
- Note: All patients had a minimum follow-up period of 12 months following their initial diagnosis.
So how many patients were included in this?
A 1:2 matched cohort enrolled 139,002 patients (46.6% women), with 46,334 (33%) exposed to GLP-1 RAs and 92,668 (66%) unexposed matched patients.
- GLP-1 RA exposure was for 6+ months.
And what was measured?
Investigators assessed the development (including how long it took) of nAMD in all patients during the follow-up period.
- A 1:2 propensity score matching between the groups was utilized to increase the compatibility of the two groups
- A Cox proportional hazards model compared the incidence of new cases of nAMD in the exposure group vs the unexposed.
Now to the findings.
The overarching result: The use of GLP-1 RAs among diabetic patients was linked to a 2-fold higher risk of nAMD development compared to diabetic patients who were not exposed.
Get into the specifics.
The mean (SD) follow-up time for the exposed versus unexposed group was 2.4 (1.0) years and 2.5 (0.9) years, respectively, while the mean (SD) history of diabetes was 6.22 (2.49) years.
Among the GLP-1 RA group, the following variations were used:
- Semaglutide (97.5%)
- Lixisenatide (2.5%)
And the nAMD diagnoses?
Out of the 46K+ patients exposed to GLP-1 RA, 0.2% were newly diagnosed by the end of November 2023.
- Compare this to patients in the non-GLP-1 RA group at 0.1%
And in an univariable Cox proportional hazards model, exposed diabetic patients had a “substantially higher incidence of a diagnosis of nAMD” compared to unexposed patients.
- The data: 0.2% vs 0.1%; difference, 0.1%; 95% confidence interval [CI], 0.08%-0.12%; hazards ratio [HR], 2.11; 95% CI, 1.58-2.82
Did any other factors play a role in nAMD development?
Yup—increased age and a history of a cerebrovascular incident were linked to a substantially increased risk for nAMD (see the adjustment model data).
So what else to take away from this?
The study authors noted that, while the risk for nAMD develop was generally low among both diabetic patient groups, “risk was highest (more than 3 times as high) among those with the longest exposure to GLP-1 RAs.”
Next: limitations of the study.
Among several noted:
- The analysis didn’t stratify by the type of GLP-1 RA prescribed, so the findings couldn’t be generalized to specific GLP-1 RAs
- The analysis also could not account for the exact GLP-1 RA dose, route of administration, or frequency of administration
- Other variables—such as smoking or sun expsoure—couldn’t be factored into the final analysis, resulting in the potential for “residual confounding”
And the big-picture outlook?
These findings are the latest in a number of recent studies that have raised issues (and concern) over the potential ocular safety of systemic GLP-1 RA use among diabetic patients, according to the authors.
- They emphasized: “It is also important to note that these findings demonstrate associations rather than cause and effect.”
Interestingly enough: These findings are also drastically different from other recent research Glance reported on earlier this year, which found RLP-1 RAs may actually reduce the risk of non-exudative AMD (and protect against glaucoma development).