Semaglutide for diabetes tied to increased NAION risk

Research published earlier this month in JAMA Ophthalmology examined the use of semaglutide, intended for type 2 diabetes (T2D) and to treat obesity, and its potential association to the development of nonarteritic anterior ischemic optic neuropathy (NAION).

Let’s start with some background.

First up: NAION, an ocular condition where blood flow to the optic nerve is either reduced or blocked.

This second most common form of optic neuropathy—and also known as optic nerve stroke—is the leading cause of acute vision loss in adults aged 50+.

Next: Semaglutide, which belongs to a class of medications referred to as glucagon-like peptide-1 (GLP-1) receptor agonists that prompts the body to produce more insulin as well as reduce appetite and initiate a feeling of fullness (leading to weight loss).

And this is FDA approved, right?

Yup. Semaglutide was approved in 2017 and 2022 for T2D and obesity, respectively.

Two products currently available on the U.S. market, both manufactured by Novo Nordisk:
- Ozempic (semaglutide injection) 0.5 mg, 1, mg, 2 mg
  - Indicated for:
    - Improving blood sugar in adults with T2D
    - Reducing the risk of major cardiovascular events in T2D patients
- Wegovy (semaglutide injection) 2.4 mg
  - Indicated for:
    - Reducing the risk of major adverse cardiovascular events in adult patients with cardiovascular disease and who are either obese or overweight
  - Reducing excess body weight and maintaining weight reduction in adults and pediatrics aged 12+

What’s the ocular connection to this drug?

Since its initial approval, semaglutide has been found to worsen existing diabetic retinopathy (DR) in diabetic patients following a drop in their blood glucose (blood sugar control).

Other effects may include blurred vision and macular complications.

While these symptoms have been noted as being manageable for the majority of diabetic patients—providing they receive regular eye exams and keep their ophthalmologists updated on medication use—new research sought to dive deeper into these associations.

Which brings us to this study?

Indeed. Researchers from Massachusetts (Mass) Eye and Ear conducted a retrospective, matched cohort study of 16,827 neuro-ophthalmic patients from one Boston, Massachusetts-based clinical practice to investigate the potential and suspected association between the use of semaglutide among diabetic patients with the risk of NAION.

Participant selection:
- Patients referred for neuro-ophthalmic conditions (excluding NAION) between Dec. 1, 2017, and Nov. 30, 2023, identified via electronic health records (EHR)

How were participants evaluated?

Based on two separate groups:

Group 1: Analyzing hazard ratio (HR) of NAION in regards to T2D
- 710 patients (median age: 59; 52% female) diagnosed with T2D and prescribed either:
  - Cohort 1: Semaglutide (n = 194)
  - Cohort 2: Non-GLP-1 receptor agonist (RA) antidiabetic medications (n = 516)
    - Included insulin and analogues; metformin; sulfonylureas; α-glucosidase inhibitors, thiazolidinediones, dipeptidyl peptidase-4 inhibitors, and sodium-glucose transport protein 2 inhibitors.
Group 2: Analyzing HR of NAION in regards to weight
- 979 patients (median age: 47; 72% female) with body mass index of 25-29.9 or > 30 (overweight or obese, respectively) prescribed either:
  - Cohort 1: Semaglutide (n = 361)
  - Cohort 2: Non-GLP-1 RA weight-loss medications (n = 618)
    - Included bupropion, naltrexone, orlistat, topiramate, phentermine, and setmelanotide

How were these two cohorts balanced?

A 1:2 nearest-neighbor propensity score was matched to account for such variables as:

Demographic factors
NAION-related comorbidities
Semaglutide contraindications

See here for the complete list.

And what was measured?

The study’s primary outcome was the development of the first NAION event.

Note: For those who experienced NAION during the study—and had been prescribed semaglutide—investigators confirmed that the medication had been dispensed.

Now these findings; How many T2D patients developed NAION?

Out of the 710 participants included in the T2D cohort, NAION developed in 17 and 6 patients from the semaglutide and non-semaglutide cohorts, respectively.

Further, at 36 months, the cumulative incidence of NAION was

8.9% (95% Confidence interval [CI], 4.5% to 13.1%) for the semaglutide cohort
1.8% (95% CI, 0% to 3.5%) for the non-semaglutide cohort

Moving on to the obese and overweight group…

Out of the 979 participants included in this group, NAION developed in 20 and 3 patients from the semaglutide and non-semaglutide cohorts, respectively.

Further, at 36 months, the cumulative incidence of NAION was

6.7% for the semaglutide cohort
0.8% for the non-semaglutide cohort

And overall for both groups?

Based on Cox proportional HR analysis, T2D patients in the semaglutide cohort demonstrated a higher risk of NAION (HR = 4.28).

The obese and overweight group demonstrated a higher risk for NAION than the T2D group (HR = 7.64).

So what does this mean?

The study authors concluded their main finding to be that the prescribed use of semaglutide is associated with an increased risk for developing NAION.

However, they stipulated that “the pathogenesis of NAION has not been fully elucidated,” indicating the need for further research.

Go on …

While both T2D and obese or overweight patients prescribed semaglutide demonstrated a substantially higher risk for NAION vs those prescribed other medications, the authors noted that this risk does not seem “to be due to difference in baseline characteristics between the cohorts.”

Important finding: Survival analyses from both groups determined that NAION may have the greatest risk of developing within the first year of starting semaglutide prescription use.

Any potential limitations?

Quite a few …

The findings may not be fully generalized to other settings, as it involved specialized neuro-ophthalmology services
Retrospective study design didn’t allow for potential biases such as which patients were prescribed semaglutide or referred/evaluated in the study’s neuro-ophthalmology clinic
Nonadherence could not be determined, potentially leading to an inaccurate estimate of semaglutide-associated risk