What the research found

Researchers investigated whether reports linking semaglutide to nonarteritic anterior ischemic optic neuropathy (NAION)—a form of vision loss—represent a genuine drug effect or reflect increased awareness and reporting bias. They analysed FDA adverse event reports spanning 2018–2026 and applied genetic methods to test whether activating GLP-1 receptors actually causes the eye condition.

The FDA database showed a dramatic spike in NAION reports after July 2024, when media coverage intensified, jumping from roughly 8 cases per 10,000 users annually (pre-2024) to over 200 per 10,000 in early 2026. Notably, reports of NAION with other GLP-1 drugs remained stable, and the timing pattern of symptom onset didn't match what you'd expect from cumulative drug toxicity. Genetic analysis using Mendelian randomization—which tests causal relationships using DNA variants—found no evidence that GLP-1 receptor activation directly causes NAION-related eye changes, while confirming that GLP-1 drugs do lower glucose as expected.

Why it matters for you

If you're using semaglutide or considering it, the distinction between notoriety bias and real toxicity is significant. Notoriety bias means the signal reflects heightened reporting and concern after publicity, not a new safety problem. The data suggest NAION reports surged after media coverage, not before—a classic pattern of awareness-driven reporting rather than a toxicity signal emerging from the drug itself.

For users tracking biomarkers and optimising health, this highlights the importance of contextualising safety signals. The genetic evidence is reassuring: it shows no mechanistic link between GLP-1 receptor activation and eye damage, even though GLP-1 drugs clearly work on their intended target (glucose control). That said, anyone with pre-existing risk factors for eye disease (like diabetes, hypertension, or previous optic nerve issues) should discuss semaglutide use with their doctor, as individual risk may differ from population-level findings.

Caveats

  • Observational reporting data: FDA adverse event reports capture spontaneously reported cases, not all adverse events; reporting patterns are influenced by media attention and provider awareness, making temporal trends difficult to interpret in isolation.
  • Genetic methods have limits: Mendelian randomization tests association, not individual response; a null genetic result doesn't rule out toxicity in specific subgroups or at very high doses.
  • No direct mechanistic study: The research doesn't assess semaglutide's effects on ocular blood flow, inflammation, or eye tissue in humans—only statistical and genetic inference.
  • Early-stage for some analyses: The data extend into 2026, making very recent findings provisional.