GLP-1 for Alzheimer’s disease looked promising in animal and health-record studies, but the largest trials have been negative. In EVOKE and EVOKE+, two phase 3 trials of 3,808 people with early Alzheimer’s, oral semaglutide 14 mg did not slow decline compared with placebo after 104 weeks.[1] In Parkinson’s, results are split: lixisenatide slowed motor decline slightly in one trial, while exenatide failed in a phase 3 trial.[4][3]
No GLP-1 drug is approved for either disease. For background on the drug class, see the GLP-1 basics hub.
Bottom line
| Trial | Disease | Drug | Size and length | Result |
|---|---|---|---|---|
| EVOKE / EVOKE+ (2026)[1] | Early Alzheimer’s | Oral semaglutide 14 mg | 3,808 people, 104 weeks (primary) | No benefit |
| ELAD (2026)[2] | Mild-moderate Alzheimer’s | Liraglutide | 204 people, 52 weeks | Primary outcome negative |
| Exenatide-PD3 (2025)[3] | Parkinson’s | Exenatide 2 mg weekly | 194 people, 96 weeks | No benefit |
| LIXIPARK (2024)[4] | Early Parkinson’s | Lixisenatide daily | 156 people, 12 months | Small slowing of motor decline |
Why the idea took hold
GLP-1 receptors are found in the brain. In lab and animal studies, GLP-1 drugs reduced inflammation, improved how brain cells use energy, and protected nerve cells. Type 2 diabetes and insulin resistance are also linked to a higher risk of dementia. Health-record studies then found that people with diabetes who took GLP-1 drugs had less dementia than those on other diabetes drugs.[6] That combination made semaglutide a leading candidate for Alzheimer’s prevention research.
Alzheimer’s: the key trials
EVOKE and EVOKE+ (Lancet, 2026). These Novo Nordisk-funded trials ran at 566 sites in 40 countries. They enrolled people aged 55 to 85 with amyloid-confirmed mild cognitive impairment or mild dementia due to Alzheimer’s. Participants took oral semaglutide up to 14 mg daily or placebo.
The main measure was the Clinical Dementia Rating-Sum of Boxes (CDR-SB), a standard scale of memory, thinking and daily function. After 104 weeks, scores worsened by about the same amount in both groups: 2.3 points on semaglutide and 2.3 on placebo in EVOKE, and 2.2 versus 2.1 in EVOKE+. Neither difference was close to significant. Side effects were consistent with semaglutide’s known profile. Both trials were discontinued.[1]
ELAD (Nature Medicine, 2026). In 204 people with mild to moderate Alzheimer’s and no diabetes, daily liraglutide for 52 weeks did not change brain glucose metabolism, the primary outcome. One secondary test of executive function favored liraglutide, but daily living and dementia rating scores did not differ.[2]
Parkinson’s: the key trials
LIXIPARK (NEJM, 2024). In 156 people diagnosed with Parkinson’s in the past 3 years, daily lixisenatide for 12 months kept motor scores (MDS-UPDRS part III) essentially unchanged (-0.04 points), while scores on placebo worsened by 3.04 points. That difference of about 3 points was statistically significant. Nausea affected 46% of people on lixisenatide.[4] Lixisenatide is an older, short-acting GLP-1 drug developed for type 2 diabetes.
Exenatide-PD3 (Lancet, 2025). This UK phase 3 trial followed an encouraging earlier phase 2 study. In 194 people with Parkinson’s, weekly exenatide 2 mg for 96 weeks did not slow motor decline: off-medication scores worsened by 5.7 points on exenatide and 4.5 on placebo. Exenatide was safe and well tolerated.[3]
Pooled analysis (2026). A meta-analysis of four randomized trials (667 people) found no difference between GLP-1 drugs and placebo in motor scores, non-motor symptoms, cognition or quality of life. Weight loss was more common on GLP-1 drugs.[5]
What’s established vs preliminary
Established:
- Oral semaglutide does not slow early Alzheimer’s disease.[1]
- Exenatide does not slow Parkinson’s progression over two years.[3]
Preliminary:
- Whether GLP-1 drugs lower the chance of developing dementia in people with diabetes. One 2026 target-trial emulation of 1,320 people with diabetes and mild cognitive impairment linked GLP-1 drugs with a 26% lower dementia risk than DPP-4 inhibitors (hazard ratio 0.74), but the authors called it hypothesis-generating.[6]
- Whether lixisenatide’s small motor effect holds up in a larger, longer trial.[4]
- Whether injectable or newer drugs such as tirzepatide, which may reach the brain differently, would perform better.
Limitations
- Timing. EVOKE treated people who already had symptoms and amyloid buildup. Prevention may require starting years earlier, which no trial has tested.[1]
- Brain access. It’s unclear how much oral semaglutide reaches the brain.
- Observational bias. People prescribed GLP-1 drugs may differ in health, income and access to care from those on older drugs.[6]
- Small Parkinson’s trials. All Parkinson’s trials so far have had fewer than 200 participants.[5]
What this means if you already take a GLP-1
These results don’t change the reasons people take GLP-1 drugs today: blood sugar control, weight, and heart and kidney protection. The EVOKE trials found side effects in line with semaglutide’s known profile, so there is no new brain-related safety concern.[1] But there is also no trial evidence to start a GLP-1 drug, or to stay on one, to protect memory or slow Parkinson’s. If memory or movement problems are a worry, ask about a full assessment rather than relying on a weight-loss or diabetes drug.
What’s next
Smaller studies continue. An Imperial College London phase 2 trial of oral semaglutide in 60 people with Alzheimer’s is listed on ClinicalTrials.gov as not yet recruiting, with completion planned for late 2027.[7] We found no active phase 3 GLP-1 trial in Alzheimer’s or Parkinson’s as of October 2026.
GLP-1 drugs still have proven benefits for heart and kidney health in the right patients; see our page on GLP-1 heart outcomes. Mood questions are covered in GLP-1 drugs and depression.
Questions to discuss with a clinician
- I have diabetes and a family history of dementia: does that change which diabetes drug makes sense for me?
- Are there clinical trials for memory or Parkinson’s that I might qualify for?
- If I’m already on a GLP-1 drug, could weight loss affect my Parkinson’s symptoms or balance?
- What treatments have proven benefit for my diagnosis?
