Semaglutide and diabetic retinopathy: what did the trial show?

What is diabetic retinopathy?

First, know the safety point: sudden vision loss or rapidly worsening sight warrants urgent assessment, whatever its cause. Diabetes can damage tiny blood vessels in the retina, the light-sensitive layer at the back of the eye. Early retinopathy may cause no symptoms, which is why regular eye examinations matter even when vision seems normal. A clinician can explain the appropriate screening schedule and treatment for an individual. NAION affects a different part of the eye and has a separate warning.31

What did SUSTAIN-6 measure?

The manufacturer-funded SUSTAIN-6 trial followed 3,297 adults with type 2 diabetes at high cardiovascular risk for 104 weeks. Participants were randomly assigned a weekly semaglutide injection or placebo, alongside usual care. A first diabetic-retinopathy complication was recorded in 50 of 1,648 people assigned semaglutide (3.0%) and 29 of 1,649 assigned placebo (1.8%). The difference was about 1.3 percentage points using the unrounded counts. These numbers describe the trial population and time period, not an individual prediction.45

Does that mean 3% became blind?

No. The trial’s combined outcome included several kinds of complication, including bleeding inside the eye and need for specific eye treatments as well as diabetes-related blindness. It was not designed as a detailed study of retinal progression and did not routinely assess visual acuity. Combining events is useful for detecting a signal, but the percentage for the combined outcome is not a percentage who became blind.45

Why might the complication rate differ?

Many participants with complications already had diabetic retinopathy. A later analysis suggested that a large early improvement in blood-glucose control could contribute to temporary worsening in some people with existing disease. This was a post hoc analysis—an investigation after the trial—and does not establish the mechanism or exclude other explanations. Improving glucose control remains important for diabetes care; the question is how the care team can account for eye health and monitoring, not whether someone should change treatment independently.

The US Ozempic label reports that among trial participants with a history of diabetic retinopathy at baseline, complications occurred in 8.2% assigned Ozempic and 5.2% assigned placebo. Among those without a known history, the corresponding figures were 0.7% and 0.4%. These are results for this trial and its combined outcome, not a prediction for an individual. The label says the effect of long-term glucose control with semaglutide on retinopathy complications has not been studied.256

What can I discuss with my care team?

Mention known diabetic eye disease, previous eye treatment, changes in sight and when you last had an eye examination. The EU Ozempic information recommends monitoring people with diabetic retinopathy and says rapid improvement in glucose control may temporarily worsen it. Follow your local eye-screening and care-team advice. If sight is suddenly lost, partly lost or rapidly worsening, urgent assessment is needed rather than waiting for a routine appointment.

The EU Ozempic information advises particular caution and close monitoring for people with diabetic retinopathy who also use insulin. It says semaglutide 2 mg is not recommended for people with type 2 diabetes and uncontrolled or potentially unstable diabetic retinopathy because there is no experience in that group. This is EU product-specific information; discuss your own product and care with your clinician rather than changing it yourself.231

How does this fit with benefits and other products?

The study enrolled a particular high-risk diabetes population and used a particular injection; its eye result should not be treated as the risk for every person or all medicines. The trial also studied cardiovascular outcomes: in that population semaglutide reduced the composite of cardiovascular death, nonfatal heart attack or nonfatal stroke versus placebo. A treatment decision considers both benefits and harms. See the semaglutide profile and how product labels differ.42