August 2026

the ozempic blindness lawsuits just crossed 200 cases — and the label hasn't moved

As of August 3, the federal multidistrict litigation consolidating Ozempic vision-loss claims — MDL No. 3163, the GLP-1 NAION Products Liability Litigation — had 200 cases filed, up from 86 in May. The MDL sits before Judge Karen Marston in the Eastern District of Pennsylvania, the same judge managing the separate Ozempic gastroparesis MDL. A special master for discovery was recently appointed to manage evidence production as the litigation moves toward its pretrial phase. The central legal theory across all 200 cases is failure to warn: plaintiffs argue that Novo Nordisk had reason to know semaglutide was associated with nonarteritic anterior ischemic optic neuropathy — a rare form of sudden optic nerve stroke that can cause permanent vision loss — and did not disclose that risk on the drug's label. As of August 2026, the FDA has not required Novo Nordisk to add a NAION warning to the Ozempic or Wegovy prescribing information. The label has not changed. That fact is not incidental to the lawsuits — it is the lawsuits.

The actual molecule

semaglutide is one of the best-characterized drugs in modern medicine — researchers still can't explain the eye signal

Semaglutide is a GLP-1 receptor agonist, a synthetic analog of glucagon-like peptide-1, the gut hormone that signals satiety, triggers meal-stimulated insulin release, and modulates food intake through hypothalamic pathways. Its pharmacology is among the most extensively documented of any recent drug: the [PubMed literature on semaglutide](https://pubmed.ncbi.nlm.nih.gov/?term=semaglutide) spans large-scale weight-management trials, cardiovascular outcome studies, and decade-long safety surveillance programs. The [current DailyMed prescribing information](https://dailymed.nlm.nih.gov/dailymed/search.cfm?query=semaglutide) documents every known interaction, contraindication, and labeled warning — none of which mention optic neuropathy. The proposed mechanism linking GLP-1 agonism to NAION remains speculative. GLP-1 receptors are expressed in retinal tissue, and researchers have hypothesized that agonist-driven vascular changes could affect perfusion to the optic nerve head — which in patients with already-narrowed vessels could tip toward ischemia. That hypothesis is biologically plausible and has not been tested in a controlled trial. The clinical programs that earned semaglutide its FDA approvals were cardiovascular and metabolic outcome studies; they were not designed to detect a rare ophthalmologic event, and they enrolled populations with elevated baseline vascular risk regardless of drug exposure.

The legal pitch

the failure-to-warn theory depends on when the risk was knowable — that is what the mdl will decide

Failure-to-warn pharmaceutical litigation has a precise structure: the plaintiff needs to establish that the manufacturer knew or had reason to know about the risk, failed to disclose it, and that the failure caused the injury. The first published signal that framed the semaglutide-NAION question as a research problem — rather than scattered case reports — was a 2024 retrospective cohort study at Massachusetts Eye and Ear. Researchers pulled records from 16,827 neuro-ophthalmic patients and found that those with diabetes taking semaglutide were over four times more likely to develop NAION over three years than diabetic patients taking other medications. [That study was published in JAMA Ophthalmology and is indexed at PubMed](https://pubmed.ncbi.nlm.nih.gov/38958939/). Novo Nordisk disputed the methodology. The authors acknowledged the sample was small and limited to a single academic center. Plaintiff attorneys began filing anyway. The central question in MDL 3163 is whether the 2024 Massachusetts study — plus the case reports and biological plausibility behind it — constituted a signal Novo Nordisk should have disclosed before it appeared in peer review, or whether the science was still too early to require label action.

What the data says

three published studies report three different risk estimates — they all show a signal and they don't agree on its size

The Massachusetts Eye and Ear study found a roughly four-fold increased risk in diabetic semaglutide users. A 2025 multicenter replication using the OHDSI collaborative network — 37.1 million adults with type 2 diabetes across 14 databases — [found a statistically significant but substantially smaller increase in NAION incidence associated with semaglutide exposure](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11843465/) than the original 4x figure. A large Danish national cohort of 424,152 people with type 2 diabetes [found once-weekly semaglutide approximately doubled the five-year risk of NAION](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11657653/) compared to non-users. The picture is three peer-reviewed studies, three different risk magnitudes — roughly 2x, 4x, and somewhere between. All three detected a signal. None established causality. The absolute background rate of NAION is low — typically 2 to 10 cases per 100,000 person-years. Doubling or quadrupling a small baseline produces a striking relative risk difference while the absolute number of people affected stays modest. That tension between a real, consistent directional signal and a genuinely contested magnitude is not spin from either side — it is the scientific reality the courts are going to have to work with.

Approved — PeptideFactCheck stance

the drug's benefit record is enormous — the vision signal is real, contested, and currently unaddressed in the label

Semaglutide holds PeptideFactCheck's Approved evidence tier, and that designation reflects a clinical record that most drugs never reach: FDA approval for type 2 diabetes and chronic weight management, a completed cardiovascular outcome trial in over 3,000 patients, the STEP weight-management program across multiple large randomized trials, and nearly a decade of postmarket safety data. Approved means regulatory certainty for labeled uses. It does not mean the label is complete or final. Labels are updated when evidence crosses a disclosure threshold — and as of August 2026, neither the FDA nor Novo Nordisk has determined that the NAION signal has crossed it. Three studies and 200 lawsuits later, that determination has not changed. The lawsuits will eventually force a reckoning: courts will hear expert testimony, review internal documents about when Novo Nordisk first saw optic event reports, and decide whether the failure-to-warn theory holds. A verdict or settlement favoring plaintiffs would pressure the FDA to update the label. An outcome favoring defendants could define the signal as background noise. [The FDA's GLP-1 safety communications](https://www.fda.gov/drugs/postmarket-drug-safety-information-patients-and-providers/fdas-concerns-unapproved-glp-1-drugs-used-weight-loss) track the current official posture. For now, the label says what it says. Source trail before certainty.

Editorial boundary

What this page will not do

It will not provide dosing, cycling, sourcing, injection, or personal medical instructions. The job is to classify claims and explain mechanisms.