FAERS comparison puts tirzepatide's optic neuropathy signal far below semaglutide's

Researchers analyzing the FDA's adverse event database found a far stronger reporting signal for optic ischemic neuropathy with semaglutide than with tirzepatide, and no signal at all for three older GLP-1 drugs. The study, by a team at the Postgraduate Institute of Medical Education and Research in Chandigarh, India, was published June 8, 2026, in Diabetes, Obesity and Metabolism ^[1].

Optic ischemic neuropathy refers to a loss of blood flow to the optic nerve that can cause sudden, often permanent vision loss in one eye. The best-known form, non-arteritic anterior ischemic optic neuropathy (NAION), has been under scrutiny as a possible semaglutide risk and has already prompted "formal regulatory review and action by the European Medicines Agency," the authors write ^[1].

What the study found

The team pulled reports from each drug's approval date through the third quarter of 2025 using OpenVigil 2.1, a tool for querying the FDA Adverse Event Reporting System (FAERS). They looked only at reports where the drug was listed as the primary suspect, and used the standardized medical term "optic ischaemic neuropathy" ^[1].

Semaglutide — the molecule in Ozempic, Wegovy and Rybelsus — had 355 reports, with a reporting odds ratio (ROR) of 94.45 (95% confidence interval 83.02–107.45) and a proportional reporting ratio of 93.77 (82.48–106.61) ^[1].

Tirzepatide, sold as Mounjaro and Zepbound, had 49 reports and an ROR of 2.94 (2.20–3.93). Liraglutide, the molecule in Victoza and Saxenda, had 13 reports and an ROR of 4.58 (2.65–7.91). Both crossed the threshold for a statistical signal, but at a small fraction of semaglutide's magnitude ^[1].

No signal was identified for dulaglutide, exenatide or lixisenatide ^[1].

Among semaglutide-linked reports, the mean age was 59 years. Disability was documented in 18% of cases and hospitalization in 11% ^[1].

The authors' main conclusion was about the class as a whole: "The absence of signals with other glucagon-like peptide-1 receptor agonists argues against a uniform class effect" ^[1]. In other words, based on these data, they do not see evidence that every incretin drug carries the same eye risk.

What these numbers do and don't mean

A reporting odds ratio compares how often an event shows up for one drug against how often it shows up for all other drugs in the database. It is a screening tool. It cannot tell you how many people out of 1,000 users will develop optic nerve damage, because FAERS has no denominator — nobody knows how many people took each drug, and reporting is voluntary and incomplete.

The authors say so directly, calling for "prospective studies with neuro-ophthalmologist-confirmed diagnoses" to "establish causality and quantify absolute risk" ^[1]. That means the diagnoses behind these 355 and 49 reports were not independently verified by eye specialists as part of this analysis.

One factor the published abstract does not address is whether media attention and regulatory activity around semaglutide and NAION inflated the number of semaglutide reports relative to other drugs — a known phenomenon in pharmacovigilance. Whether the study's full text adjusts for that is not clear from the material available here.

Why it matters for patients

For someone weighing semaglutide against tirzepatide, or already on one of them, this study adds comparative context to a safety question that has so far been discussed mostly around semaglutide alone. It suggests the reporting patterns for the two drugs are very different in size, and that older GLP-1 drugs like dulaglutide and exenatide show no signal at all ^[1].

But the study does not establish that semaglutide causes optic nerve damage, and it does not tell anyone their personal odds. The outcome reports that did exist were serious: nearly one in five semaglutide cases involved documented disability ^[1]. Sudden vision changes in one eye are the kind of symptom that gets evaluated urgently regardless of what medication a person takes.

What happens next

The European Medicines Agency has already reviewed and acted on the semaglutide-NAION question ^[1]. The sources here do not describe any current or planned FDA action on tirzepatide labeling, nor do they say whether any regulator is reviewing the tirzepatide data. The authors' stated next step is prospective research with confirmed diagnoses ^[1]; no timeline for such studies is given.