FDA NAION warning now covers every GLP 1 drug after a review of 126 trials
The FDA told makers of all GLP 1 receptor agonists to add a possible higher risk of a rare optic nerve stroke to their labels. Its pooled trial data show about 2.6 extra cases per 100,000 person years.
What's new
The FDA required makers of all GLP 1 receptor agonists to add a possible higher risk of NAION, a rare optic nerve stroke, to the Warnings and Precautions section of their labels.
Why it matters
Millions take these drugs, and NAION can cause permanent vision loss. The absolute risk is small, but it now applies to the whole class, not only semaglutide.
What's next
Labels will be updated, and the FDA asks clinicians and patients to report suspected cases through MedWatch as further studies refine the estimate.
The FDA NAION warning announced on November 9 now applies to every GLP 1 receptor agonist sold in the United States, from exenatide to orforglipron. The agency told drugmakers to add the possible higher risk of non arteritic anterior ischemic optic neuropathy, a sudden loss of blood flow to the optic nerve, to the Warnings and Precautions section of each label. The FDA calls the condition rare, and its own numbers show a small absolute risk.
What the FDA NAION warning changes on the label
The safety communication lists twelve products, including Ozempic, Rybelsus and Wegovy for semaglutide, Mounjaro and Zepbound for tirzepatide, Trulicity, Victoza, Saxenda, the insulin combinations Xultophy and Soliqua, generic exenatide and the 2026 oral drug Foundayo. The label change covers the whole class, not one molecule. Europe took a narrower step in 2025, when its safety committee added the condition to semaglutide alone as a very rare side effect, as covered in our report on the EU semaglutide eye warning. The FDA says it reached its view after finding reports of the condition for every GLP 1 drug now on the market. The agency also says it found no mechanism in animal studies that would explain how these drugs could harm the optic nerve.
NAION usually shows up as painless vision loss in one eye, often noticed on waking, and the damage can be permanent. The FDA lists the warning signs patients should act on at once: sudden blurred or hazy central vision, blind spots, tunnel vision or colours that look wrong. It also names the background risk factors that overlap heavily with the people who take these drugs, namely smoking, diabetes, high blood pressure, sleep apnea, anemia and kidney disease. A person who has had the condition in one eye faces a higher risk in the other. The agency says clinicians should refer suspected cases to an eye specialist quickly and consider stopping the drug unless the benefits clearly outweigh the risk.
The numbers behind the class warning
The FDA pooled 126 randomized controlled trials covering 134,253 patients, of whom 78,217 took a GLP 1 drug and 56,036 took placebo or another comparator. Fifteen probable NAION events turned up across 288,879 person years of follow up: ten in the GLP 1 groups, five on placebo and none on other glucose lowering drugs. That works out to 6.4 cases per 100,000 person years on GLP 1 drugs against 3.8 on comparators, a gap of about 2.6 extra cases for every 100,000 people treated for a year. With so few events, the agency did not publish a confidence interval for that comparison on its page.
That works out to 6.4 cases per 100,000 person years on GLP 1 drugs against 3.8 on comparators, a gap of about 2.6 extra cases for every 100,000 people treated for a year.
A second analysis used the FDA Sentinel system to follow adults with type 2 diabetes who started a GLP 1 drug, an SGLT 2 inhibitor or a DPP 4 inhibitor between 2019 and 2024. Rates ran from 26 to 32 cases per 100,000 person years. Against DPP 4 inhibitors, GLP 1 users had a hazard ratio of 1.43, with a 95% confidence interval of 1.02 to 2.01, which counts as a higher risk. Against SGLT 2 inhibitors the hazard ratio was 1.19 with an interval of 0.91 to 1.54, which does not reach significance. The table below sets out the main figures from the FDA safety communication.
| Data set | Population | Main figure |
|---|---|---|
| Pooled trials | 126 trials, 134,253 patients | 6.4 vs 3.8 cases per 100,000 person years |
| Sentinel cohort | Type 2 diabetes, 2019 to 2024 | hazard ratio 1.43 vs DPP 4 inhibitors |
| Sentinel cohort | Same population | hazard ratio 1.19 vs SGLT 2 inhibitors, not significant |
| Published background rate | Adults over 50 | 2.3 to 10.2 cases per 100,000 people a year |
How to read the GLP 1 eye risk
Ziyad Al-Aly, a clinical epidemiologist at Washington University in St. Louis, wrote in a piece for The Conversation that the evidence now points the same way from several directions. He listed the two FDA analyses, observational studies from independent teams and the fact that two regulators reached similar views on their own. He also argued that the total number of people likely to be affected is small, and that patients using these drugs for an approved condition on medical advice should not be put off by the warning alone. His point was that awareness, not alarm, is the right response: patients should know the symptoms and seek care quickly if they appear.
The size of the GLP 1 eye risk is still uncertain. The FDA itself says its adverse event reports were hard to interpret, because the conditions that raise NAION risk are common in people prescribed these drugs. Most published work has focused on semaglutide, and no study yet compares individual drugs head to head for this outcome, so it remains unknown whether some carry more risk than others. Background rates also vary widely by age. The warning therefore sits alongside earlier concerns rather than closing the debate, and it adds weight to legal claims already described in our story on NAION lawsuits and vision loss, which the label change does not decide.
What patients and prescribers are asked to do
The practical advice attached to the FDA NAION warning is short. Before starting a GLP 1 drug, prescribers should explain the eye risk, and patients should report any sudden change in vision in one eye straight away rather than waiting for a routine appointment. People with a past episode of NAION carry a higher risk of it coming back, so they need a careful talk about benefits and risks before they start or continue treatment. The FDA does not tell anyone to stop a drug on their own. It frames stopping as a decision for the patient and clinician to make together once NAION has occurred, weighing heart, kidney and weight benefits against the eye risk.
Drugmakers now have to update their labels, and the agency is asking doctors and patients to report suspected cases through its MedWatch program, which will add to the record over time. Further observational studies and longer trials with eye checks may narrow the estimate of extra cases, or show whether the risk differs by dose, by drug or by length of use. For readers following the rest of the safety picture, the Peptide Zone safety desk tracks label changes and side effect reviews as they come out. None of this changes the approved uses of these medicines, which still include diabetes, weight management, heart and kidney protection and sleep apnea.
Conclusion
The FDA NAION warning turns a semaglutide question into a label for the whole GLP 1 class. The agency's trial data suggest about 2.6 extra cases per 100,000 person years, a small absolute risk for a condition that can still cause permanent vision loss. Sudden painless vision loss in one eye on any of these drugs calls for prompt eye care, and patients should decide with their prescriber whether to continue treatment afterwards. This is research reporting, not medical advice.