Metabolic medicine
What the FDA’s GLP-1 Eye Safety Warning Means for Clinical Trials
The FDA requires GLP-1 labels to warn of potential NAION risk. Review the trial and Sentinel evidence and what sponsors can do to prepare study sites.
By Lampgen Clinical Research

4 min read · estimated
What the source reported
On November 9, 2026, the FDA required manufacturers to add a warning about a potential increased risk of non-arteritic anterior ischemic optic neuropathy (NAION) to the Warnings and Precautions section of GLP-1 receptor agonist labels. The U.S. action covers medicines with GLP-1 activity, including semaglutide, the dual GIP/GLP-1 agonist tirzepatide, and oral orforglipron. For trial teams, the immediate question is whether a report of sudden vision change can move quickly from the participant to a clinician and whether the study can capture what happened. [1, 2]
What the FDA found in trials and routine care
The FDA pooled 126 randomized trials involving 134,253 patients and approximately 289,000 person-years of exposure. It identified 10 probable NAION events among 78,217 participants receiving a GLP-1 drug and five among 56,036 participants receiving placebo or a non-GLP-1 glucose-lowering drug. All five comparator events occurred in placebo recipients. The pooled rates were 6.4 and 3.8 events per 100,000 person-years, respectively. That is a difference of 2.6 events per 100,000 person-years in the pooled analysis. With just 15 events, the estimates cannot tell a clinician how much risk a particular drug adds for an individual patient. [1]
In the FDA’s retrospective Sentinel study of adults with type 2 diabetes who started treatment between 2019 and 2024, the comparison drug mattered. Relative to DPP-4 inhibitors, the GLP-1 group had a hazard ratio of 1.43 (95% CI 1.02–2.01). Relative to SGLT2 inhibitors, the hazard ratio was 1.19 (95% CI 0.91–1.54). The latter interval included 1.0. These estimates come from a commercially insured diabetes population. The underlying conditions and access to diagnosis can differ in obesity-only populations or in patients outside that network. [1]
Earlier observational work also found that the size of the association varied with study design and comparator. In 2025, the European Medicines Agency’s safety committee classified NAION as a very rare side effect of semaglutide specifically. The FDA’s newer warning has a broader scope and describes a potential risk. The FDA did not identify a mechanism in the animal data it reviewed. [1, 3, 4]
What patients and clinicians need to recognize
NAION occurs when blood flow to part of the optic nerve decreases. It can cause sudden, painless loss of vision or a new blind spot in one eye, with lasting impairment in some cases. Diabetes, high blood pressure, sleep apnea, and kidney disease can also raise baseline risk, which complicates interpretation of reports in people taking GLP-1 medicines. [1]
The FDA advises patients to contact a health care professional immediately if these symptoms occur. It advises clinicians to arrange prompt ophthalmology evaluation when NAION is suspected and to consider discontinuing a GLP-1 drug after NAION occurs unless the benefits clearly outweigh the risk. Treatment decisions depend on the individual patient and their clinician. [1]
What this means for trial teams
Review the safety plan and participant materials. For relevant ongoing studies, sponsors and investigators should assess whether the new information changes the protocol, safety collection plan, investigator brochure, or consent materials. Under ICH E6(R3), teams should promptly communicate new information relevant to a participant’s willingness to continue, document the discussion, and assess whether reconsent is needed. [5]
Rehearse the first call. At site training, ask who answers after hours, how that person reaches the investigator, and how the site arranges prompt ophthalmology assessment. Record symptom onset, which eye was affected, exposure timing, clinical review, the referral outcome, and the specialist’s findings when available. The CRA can check that the source, study, and safety records reconcile. The investigator directs care and the sponsor safety team handles regulatory reporting under the applicable requirements. [5]
Check who completes the referral. With Lampgen’s SIGNAL™ framework, compare time from first report to clinical review and completed referral across sites. Look for barriers such as transportation, interpreter availability, and access to an after-hours specialist. Use those measures to assess the response process. A delayed exam may leave the case unresolved and distort the picture of how often events are confirmed. [6, 7]
Sources
- FDA, GLP-1 receptor agonists and potential NAION risk, Drug Safety Communication, November 9, 2026
- Elaine Chen, STAT reporting on the FDA warning, November 9, 2026 (public excerpt)
- Cai et al., semaglutide and NAION across 14 databases, JAMA Ophthalmology, 2025
- European Medicines Agency, PRAC review of NAION and semaglutide, June 2025
- FDA, ICH E6(R3) Good Clinical Practice final guidance, September 2025
- Lampgen Clinical Insights, trial participation burden and patient access
- Lampgen Clinical Research, a worked SIGNAL™ monitoring case