Quick Answer
Rybelsus (oral semaglutide), a GLP-1 medication, has not been shown to slow Alzheimer's disease. In the large phase 3 EVOKE and EVOKE+ trials, oral semaglutide did not slow decline in thinking or daily function compared with placebo in people who already had early Alzheimer's disease. The trials also did not delay progression from mild cognitive impairment to dementia.
Separate observational studies have found fewer dementia or Alzheimer's diagnoses among some people with type 2 diabetes who use GLP-1 medications. Those studies can identify an association, but they cannot prove that the medication prevented dementia. No GLP-1 medication is approved to prevent or treat dementia.
Key Points
- The phase 3 treatment evidence is negative: EVOKE and EVOKE+ found no clinical benefit from oral semaglutide in early Alzheimer's disease.
- Biomarker changes were not a treatment benefit: Some inflammation- and Alzheimer's-related laboratory markers improved, but cognition and function did not decline more slowly.
- Prevention is a different question: Observational data suggest lower rates of dementia diagnoses in some GLP-1 users, but randomized prevention trials have not established a protective effect.
- Do not assume a class effect: A result involving semaglutide does not prove that every GLP-1 medication will have the same result.
- Medication decisions should follow approved uses: Do not start or continue a GLP-1 medication because you expect it to protect memory.
The Most Important Update: EVOKE and EVOKE+ Were Negative
The earlier excitement around GLP-1 medications and Alzheimer's came from animal experiments, small studies, and health-record analyses. Researchers then tested semaglutide, a GLP-1 medication, in two large randomized trials.
EVOKE and EVOKE+ enrolled 3,808 adults ages 55 to 85 with biomarker-confirmed Alzheimer's disease at the mild cognitive impairment or mild dementia stage. Participants received once-daily oral semaglutide, up to 14 mg, or placebo in addition to standard care.
After 104 weeks, semaglutide was not better than placebo at slowing decline on the Clinical Dementia Rating–Sum of Boxes, a measure that combines cognition and everyday function. It also did not improve the key daily-function outcome. A pooled analysis found no delay in progression to dementia among participants who began with mild cognitive impairment. Because the clinical outcome was negative, the extension period was discontinued.
Semaglutide did reduce several fluid biomarkers related to inflammation, tau, and neurodegeneration. That is scientifically interesting, but it did not translate into slower clinical decline. A biomarker moving in a favorable direction is not the same as a person thinking, functioning, or living independently for longer.
Prevention and Treatment Are Not the Same Question
Treatment and prevention are separate research questions:
- Treatment research asks whether a medication slows decline after Alzheimer's disease is already present.
- Prevention research asks whether a medication lowers the chance of developing dementia in the first place.
EVOKE and EVOKE+ tested treatment, not prevention, and the treatment result was negative. The trials do not prove that GLP-1 medications have no possible effect before Alzheimer's begins. But they also do not support calling semaglutide an Alzheimer's therapy or a proven “brain-protective” drug.
For prevention, the most encouraging findings still come from observational research. A 2024 target-trial emulation using electronic health records from people with type 2 diabetes found that semaglutide use was associated with fewer first-time Alzheimer's diagnoses than several other diabetes treatments. A separate Veterans Affairs analysis found that GLP-1 receptor agonist use was associated with lower risks of several neurocognitive outcomes.
Those findings are hypotheses to test, not proof of prevention. People prescribed different diabetes drugs may differ in health, weight, access to care, prescribing history, or other ways that researchers cannot fully measure. Even careful statistical adjustment cannot reproduce random assignment.
What About Other GLP-1 Medications?
The negative EVOKE results are specifically about oral semaglutide in people with early symptomatic Alzheimer's disease. Researchers are still studying whether another drug, dose, disease stage, or patient group could produce a different result.
Liraglutide has also been tested in Alzheimer's disease. In the 204-person phase 2b ELAD trial, liraglutide did not meet its primary brain-metabolism outcome. One secondary cognitive measure favored liraglutide, but measures of daily living and overall clinical status did not differ significantly. Secondary findings from a relatively small trial are not enough to establish an Alzheimer's treatment.
For now, no member of the GLP-1 drug class has demonstrated a clear clinical benefit for treating Alzheimer's disease.
Why the Biology Sounded Promising
There were reasonable scientific reasons to run these trials. GLP-1 signaling affects metabolism, appetite pathways, blood vessels, and inflammatory processes. Diabetes, obesity, high blood pressure, and cardiovascular disease are also connected with dementia risk.
But plausible biology does not guarantee a clinical benefit. Claims that semaglutide simply crosses the blood-brain barrier and directly protects memory centers are especially misleading. In a rodent distribution study, semaglutide reached selected brain regions but did not broadly cross the blood-brain barrier. More importantly, animal findings cannot establish that a medication preserves memory in people.
The EVOKE trials illustrate why clinical outcomes matter: laboratory markers changed, yet patients did not experience slower cognitive and functional decline.
What This Means If You Take a GLP-1 Medication
If You Have Type 2 Diabetes or Obesity
Use semaglutide or another GLP-1 medication for the condition your clinician prescribed it to treat—not for an expected dementia benefit. Current FDA prescribing information for semaglutide does not include dementia prevention or Alzheimer's treatment.
That does not erase the established benefits a particular GLP-1 medication may offer for approved metabolic or cardiovascular uses. It means those benefits should not be converted into an unsupported brain-protection claim.
If You Are Worried About Memory Changes
Do not self-treat with a GLP-1 medication. Memory problems can have many causes, including medication effects, sleep problems, depression, vitamin deficiencies, thyroid disease, and neurodegenerative disease. A clinician can review symptoms, medicines, and next steps.
If You Want to Reduce Dementia Risk
There is no guaranteed way to prevent Alzheimer's disease. The National Institute on Aging describes evidence for risk-reduction strategies as promising but not definitive. Managing high blood pressure, staying physically active, addressing diabetes and other cardiovascular risks, avoiding smoking, and maintaining social connection support overall health and may help reduce dementia risk.
What We Still Do Not Know
- Whether a GLP-1 medication can prevent dementia in people who do not yet have cognitive symptoms
- Whether any observed association is caused by the drug itself, weight loss, improved diabetes control, or differences between patient groups
- Whether another GLP-1 medication or a different treatment stage would produce a meaningful clinical result
- Which patients, if any, might benefit cognitively
- Whether longer follow-up would change prevention findings
The Bottom Line
The evidence now has an important split. Observational studies associate GLP-1 medication use with fewer dementia diagnoses in some people with type 2 diabetes, but that does not prove prevention. In the phase 3 EVOKE and EVOKE+ treatment trials, oral semaglutide did not slow early Alzheimer's disease.
Do not view semaglutide or another GLP-1 medication as an Alzheimer's treatment or proven dementia-prevention strategy. Use these medications for approved indications after an individualized discussion with a healthcare professional.
Frequently Asked Questions
Did semaglutide slow Alzheimer's disease in phase 3 trials?
No. In EVOKE and EVOKE+, oral semaglutide did not slow decline in cognition and daily function compared with placebo after 104 weeks. It also did not delay progression to dementia in participants who entered with mild cognitive impairment due to Alzheimer's disease.
Do lower dementia rates in GLP-1 users prove prevention?
No. Observational studies can show an association, but they cannot fully rule out differences between people who received different medications. Randomized prevention trials would be needed to establish whether a GLP-1 medication lowers dementia risk.
Did semaglutide improve Alzheimer's biomarkers?
Some fluid biomarkers related to Alzheimer's disease and inflammation improved in EVOKE and EVOKE+, but the changes did not lead to slower cognitive or functional decline. Biomarker improvement alone is not proof that a treatment helps patients.
Are GLP-1 medications approved for dementia?
No. GLP-1 medications are not FDA-approved to prevent dementia or treat Alzheimer's disease. Their approved uses vary by product and can include type 2 diabetes, chronic weight management, and certain cardiovascular or kidney-risk indications.
Should I take a GLP-1 medication to protect my brain?
Not on that basis. Medication choice should depend on an approved indication, your health history, possible benefits and harms, and a discussion with your healthcare professional—not an unproven expectation of memory protection.
This article is for informational purposes only and does not replace medical advice. GLP-1 medications are not approved for preventing or treating dementia. Talk with a qualified healthcare professional about memory concerns and medication decisions.
