GLP-1s for Parkinson's and Neurodegeneration: What the 2025 Phase 3 Trial Changed

The Exenatide-PD3 trial published in The Lancet in February 2025 was negative. What it means for GLP-1 neuroprotection, and which open questions remain.

For several years, GLP-1 medications looked like a potential breakthrough in neurodegenerative disease. Small early trials suggested slowed motor progression in Parkinson's disease. Epidemiologic studies showed reduced dementia incidence among diabetic patients on GLP-1s. Animal data consistently demonstrated neuroprotective effects across multiple disease models. For patients with progressive, incurable neurologic conditions, the possibility of a repurposed, widely-available medication was genuinely exciting.

Then in February 2025, the largest and longest Phase 3 trial of a GLP-1 in Parkinson's disease — Exenatide-PD3 — reported no benefit. The results were published in The Lancet and substantially changed the field's assumptions about GLP-1 neuroprotection.

Here's what the Parkinson's research actually shows, what the negative Exenatide-PD3 result means, and what the broader neurodegenerative picture looks like in 2026.

The Biology That Prompted the Research

Several lines of evidence suggested GLP-1 medications might have neuroprotective effects:

Brain GLP-1 Receptors

GLP-1 receptors are widely distributed in the brain, including in the substantia nigra — the dopamine-producing region that degenerates in Parkinson's disease. Activating these receptors in animal models produces neurotrophic effects, reduced neuroinflammation, and protection of dopaminergic neurons against various toxic insults.

Animal Model Success

Multiple laboratories demonstrated that GLP-1 agonists — exenatide, liraglutide, semaglutide, lixisenatide, and others — protected dopaminergic neurons in mouse and rat Parkinson's models. The findings were consistent across MPTP-induced, rotenone-induced, and 6-OHDA-induced disease models.

Epidemiologic Signals

Two large epidemiologic studies of type 2 diabetes patients showed reduced subsequent rates of Parkinson's disease diagnosis among those treated with GLP-1 medications compared to other diabetes therapies. Effect sizes ranged from 25–50% reductions.

Early Clinical Trials

A 2017 Phase 2 trial of exenatide in 60 patients with Parkinson's (Athauda et al., Lancet) showed improvement in off-medication motor scores that persisted after exenatide was discontinued — suggesting disease modification rather than just symptomatic benefit. This was an especially promising signal because symptomatic effects of Parkinson's medications typically disappear when the medication is stopped.

Lixisenatide Phase 2 Signal

The 2024 LixiPark trial (NEJM) tested lixisenatide in early Parkinson's and showed that patients on lixisenatide held steady on their motor scores while placebo-group patients progressed. The effect was modest but statistically significant.

Exenatide-PD3 — What Happened

Exenatide-PD3 was designed to definitively test whether exenatide could slow Parkinson's progression. The trial design:

• Phase 3, randomized, double-blind, placebo-controlled
• 6 UK hospitals
• 194 patients with Parkinson's disease (Hoehn and Yahr stage ≤2.5)
• Exenatide 2 mg once weekly or placebo for 96 weeks
• Primary outcome: MDS-UPDRS part III motor score off dopaminergic medication at 96 weeks

Results published in The Lancet (February 4, 2025):

• MDS-UPDRS III motor scores increased (worsened) by 5.7 points in the exenatide group and 4.5 points in the placebo group — essentially no difference
• No benefit on secondary endpoints including non-motor symptoms
• No signal on DaT-SPECT imaging of dopamine neuron function
• No subgroup analysis identified a population that responded
• Safety was consistent with previous exenatide experience

The trial was well-designed, adequately powered, and the result was clear: exenatide did not slow Parkinson's progression over 96 weeks.

Why It Didn't Work

Several explanations have been proposed for why Exenatide-PD3 was negative despite promising earlier data:

1. Brain Penetrance Limitations

Cerebrospinal fluid analysis from trial participants suggested exenatide reached the brain in only relatively small amounts. The drug may be getting to its peripheral target (GLP-1 receptors on pancreatic beta cells, etc.) but not achieving sufficient brain exposure to produce neuroprotective effects. This is a drug-penetration problem, not necessarily a pharmacologic-principle failure.

2. Earlier-Trial Effects May Have Been Symptomatic, Not Disease-Modifying

The Phase 2 trial (2017) showed benefits that persisted 12 weeks after medication was stopped — interpreted at the time as evidence of disease modification. With hindsight, these effects may have represented symptomatic effects that took longer to wash out than expected rather than true disease modification. Larger trial may simply have been detecting the absence of a disease-modifying effect that the smaller trial was underpowered to rule out.

3. Placebo Effect Differences

Parkinson's disease trials are notoriously susceptible to large placebo effects. Differences between Phase 2 and Phase 3 placebo performance could have affected the apparent treatment-placebo difference.

4. Population Differences

Different inclusion criteria, geographic differences, and baseline disease severity between Phase 2 and Phase 3 could have affected results.

5. Wrong Drug for the Job

Perhaps exenatide specifically — with its limited brain penetrance — was not the right GLP-1 for neurologic targets. Other GLP-1 medications, particularly those with better blood-brain barrier penetration, might still work. This is an open question.

What's Still Ongoing

Despite the Exenatide-PD3 result, several GLP-1 neurologic trials continue:

evoke and evoke+ (Alzheimer's Disease)

These two pivotal Phase 3 trials of oral semaglutide in early Alzheimer's disease enrolled ~3,700 combined patients and are expected to report in 2026. Oral semaglutide has better brain penetrance than exenatide (smaller molecular structure, and pharmacology optimized for oral absorption). Results could be positive, negative, or mixed.

LixiPark Extension

The LixiPark trial of lixisenatide in early Parkinson's is being extended for longer follow-up. Lixisenatide showed a signal in the original Phase 2 trial; whether this holds over longer duration is being tested.

GLP-1 in Progressive Supranuclear Palsy (PSP) and Related

Exploratory trials in other neurodegenerative conditions continue.

Semaglutide in Parkinson's (Different From Exenatide)

A trial specifically testing semaglutide in Parkinson's is in earlier stages. Given semaglutide's different pharmacokinetics and brain penetrance, the result is not prejudged by the exenatide result.

What Patients and Families Should Take Away

If you have Parkinson's disease and are considering GLP-1 therapy for its potential neuroprotective effects:

• The evidence no longer supports this use. The best-designed, largest trial was negative. The Phase 2 signals that motivated the research didn't replicate.
• Standard Parkinson's treatments remain the foundation. Carbidopa-levodopa, dopamine agonists, MAO-B inhibitors, and other established therapies have much stronger evidence.
• Lifestyle approaches have stronger evidence than GLP-1s for Parkinson's. Exercise (particularly intensive aerobic exercise like fast walking, cycling, or boxing programs), adequate sleep, and cognitive engagement have reasonable evidence for modest benefit.
• If you have a separate indication for GLP-1 therapy (diabetes, cardiovascular disease, CKD, obesity), the standard indication still applies regardless of Parkinson's status. You don't avoid GLP-1 therapy because of Parkinson's — you just don't take it for Parkinson's.

The Alzheimer's Question Is Still Open

The Parkinson's result doesn't automatically translate to Alzheimer's. Several reasons why Alzheimer's might still respond where Parkinson's didn't:

• Different disease mechanism. Parkinson's is primarily dopaminergic neuron loss in the substantia nigra. Alzheimer's involves amyloid and tau pathology across cortex and hippocampus. GLP-1 might affect one pathology and not the other.
• Different drug being tested. evoke/evoke+ are testing oral semaglutide, which has different pharmacokinetics than exenatide.
• Stronger metabolic-disease connection in Alzheimer's. The "type 3 diabetes" framing of Alzheimer's as brain insulin resistance is well-developed. Parkinson's has less clear metabolic dysfunction in its pathophysiology.
• Different outcome measures. Cognitive decline in Alzheimer's may be more responsive to broad metabolic improvement than motor progression in Parkinson's.

The 2026 evoke readouts will significantly clarify where GLP-1 neuroprotection stands. Positive results would revive the field; negative results would substantially narrow it to specific indications only.

Other GLP-1 Brain Effects

Beyond neurodegeneration, GLP-1 medications have documented effects on:

• Appetite regulation (the primary weight-loss mechanism)
• Food reward signaling (the "food noise" effect)
• Alcohol use behaviors (small trials suggest reduced alcohol craving)
• Possibly nicotine use (early observational signals)
• Possibly cognitive function in obese/diabetic patients (subjective improvements often reported, mechanism unclear)

These effects are established to varying degrees and represent the real non-metabolic brain effects of GLP-1 therapy. Whether they translate to meaningful effects in neurodegenerative disease remains the open question the ongoing trials are testing.

Questions Worth Asking

For patients or families considering GLP-1 therapy in the context of neurologic conditions:

• If I have Parkinson's disease, what standard treatments have we optimized before considering anything experimental?
• If I have a separate GLP-1 indication, does my Parkinson's change the treatment calculus?
• If I'm interested in participating in GLP-1 neurologic trials, are there any actively recruiting relevant to my condition?
• What evidence-based interventions for Parkinson's or dementia progression should we prioritize in my treatment plan?