GLP-1 Microdosing and Parkinson's Disease: What the Exenatide and Lixisenatide Trials Show About Movement Symptoms
A diabetes drug that might protect dopamine neurons made headlines in 2017. Since then two trials have looked hopeful and two have come back flat. Here is what each one measured, and what none of them tested.
3.08
POINTS OF MOTOR SCORE SEPARATING LIXISENATIDE FROM PLACEBO AT 12 MONTHS IN LIXIPARK
0.47
P VALUE IN THE 96-WEEK PHASE 3 EXENATIDE TRIAL, MEANING NO DIFFERENCE FROM PLACEBO
0.81
ADJUSTED HAZARD RATIO FOR A NEW PARKINSON'S DIAGNOSIS IN SCANDINAVIAN GLP-1 USERS
Interest in GLP-1 microdosing and Parkinson's disease comes from a real scientific lead. GLP-1 receptors are found in the brain, GLP-1 drugs protected neurons in animal models of Parkinson's disease, and people who take these drugs for diabetes seem to be diagnosed with Parkinson's disease less often. Researchers have now run several randomized trials to find out whether that lead holds up in patients.
This article walks through those trials one by one, then the population studies, then two practical issues for anyone with Parkinson's disease who is thinking about a GLP-1: weight loss and levodopa absorption. One limit applies throughout. Every trial used a standard prescribed dose. No study has tested a microdose in Parkinson's disease, and nothing here replaces a conversation with a neurologist.
Key takeaways
- •In the 2017 Exenatide-PD trial of 62 patients, off-medication motor scores were 3.5 points better on exenatide 2 mg weekly than on placebo at 60 weeks.
- •In the LIXIPARK trial of 156 people with early Parkinson's disease, motor scores changed by -0.04 points on daily lixisenatide and worsened by 3.04 points on placebo over 12 months.
- •The phase 3 Exenatide-PD3 trial of 194 participants found no benefit after 96 weeks: scores worsened by 5.7 points on exenatide and 4.5 points on placebo.
- •A Scandinavian cohort of 158,961 new GLP-1 users found fewer new Parkinson's diagnoses than among sulfonylurea users (adjusted hazard ratio 0.81), an association that does not prove cause.
- •No trial has tested a GLP-1 microdose in Parkinson's disease, and low body weight at diagnosis is linked to higher mortality in people with the condition.
Why are GLP-1 drugs being studied for Parkinson's disease?
GLP-1 drugs are being studied for Parkinson's disease because of how insulin signalling works in the brain. The Cochrane review by Mulvaney and colleagues (2020) explains that GLP-1 receptors have been found in the brain, that insulin signalling there supports the metabolism and repair of neurons, and that this signalling is desensitised in people with Parkinson's disease. GLP-1 receptor agonists, which are licensed for type 2 diabetes, might restore some of it.
The first human test was small and unblinded to participants. Aviles-Olmos and colleagues (Journal of Clinical Investigation, 2013) followed 45 patients with moderate Parkinson's disease who either added exenatide injections for 12 months or acted as controls. Raters who did not know the group assignments scored videos of the patients. The exenatide group improved by 2.7 points on the MDS-UPDRS and the control group declined by 2.2 points.
That design could not rule out a placebo effect, because patients knew whether they were injecting. The authors presented the study as an inexpensive proof of concept to justify proper double-blind trials. Those trials are the ones that matter.
What did the exenatide and lixisenatide trials find?
Two mid-sized placebo-controlled trials found that people on a GLP-1 drug held their motor scores steady while people on placebo got worse. Both used the MDS-UPDRS part III, a clinician-rated motor exam scored from 0 to 132, where higher scores mean greater disability.
Athauda and colleagues (Lancet, 2017) randomized 62 patients with moderate Parkinson's disease at a single centre to exenatide 2 mg or placebo once weekly for 48 weeks, on top of their usual medication. Scores were taken at 60 weeks, after a 12-week washout, with patients off their dopaminergic medication. The exenatide group had improved by 1.0 points and the placebo group had worsened by 2.1 points. The adjusted difference was 3.5 points in favour of exenatide (95% CI -6.7 to -0.3).
Meissner and colleagues (New England Journal of Medicine, 2024) ran LIXIPARK, a phase 2 trial in 156 people diagnosed less than 3 years earlier. Participants injected lixisenatide or placebo daily for 12 months. Baseline motor scores were about 15 in both groups. At 12 months, scores had changed by -0.04 points with lixisenatide and by 3.04 points with placebo, a difference of 3.08 points (95% CI 0.86 to 5.30).
Change in motor score (MDS-UPDRS part III, points). Above zero means worse.
Athauda D et al., Lancet 2017 (60 weeks, off medication); Meissner WG et al., N Engl J Med 2024 (12 months, on medication); Vijiaratnam N et al., Lancet 2025 (96 weeks, off medication). Trials differ in length and design and are not directly comparable.
- GLP-1 drug
- Placebo
Neither team claimed the disease had been slowed. The 2017 authors wrote that they could not tell whether exenatide affects the underlying disease or simply produces long-lasting symptom relief. In LIXIPARK, the other secondary measures did not differ much between groups. The Cochrane reviewers rated the motor evidence available in 2020 as low certainty and found little or no effect on health-related quality of life.
Did the phase 3 exenatide trial confirm a benefit?
No. The longest trial, and the largest trial of exenatide, found no benefit over placebo. Vijiaratnam and colleagues (Lancet, 2025) randomized 194 participants at six UK research hospitals to extended-release exenatide 2 mg once a week or placebo for 96 weeks. That is twice the treatment period of the 2017 trial and about three times the number of patients.
At 96 weeks, off-medication motor scores had worsened by a mean of 5.7 points in the exenatide group and 4.5 points in the placebo group. The adjusted effect of exenatide was 0.92 points in the wrong direction, with a confidence interval from -1.56 to 3.39 and a p value of 0.47. Exenatide was safe and well tolerated: 9% of the exenatide group and 11% of the placebo group had a serious adverse event.
The authors' conclusion was blunt. They found "no evidence to support exenatide as a disease-modifying treatment" for people with Parkinson's disease. They suggested that future studies use agents with better target engagement or focus on specific subgroups of patients.
A second negative result came from NLY01, a longer-lasting pegylated form of exenatide designed to reach the brain. McGarry and colleagues (Lancet Neurology, 2024) randomized 255 people with early, untreated Parkinson's disease at 58 US clinics to 2.5 mg, 5.0 mg or placebo for 36 weeks. Neither dose differed from placebo on the combined MDS-UPDRS parts II and III score. A subgroup analysis hinted at a motor benefit in younger participants, which the authors treated as exploratory.
Do people taking GLP-1 medicines develop Parkinson's disease less often?
In three large observational studies of people with diabetes, GLP-1 users were diagnosed with Parkinson's disease less often than users of comparison drugs. Brauer and colleagues (Brain, 2020) studied 100,288 patients with diabetes in UK primary care records, of whom 329 were diagnosed with Parkinson's disease over a median follow-up of 3.33 years.
New Parkinson's diagnoses per 10,000 person-years, by diabetes drug
Brauer R et al., Brain 2020. UK primary care records, patients with diabetes. Unadjusted incidence rates.
After adjustment, the incidence rate ratio for GLP-1 drugs was 0.38 (95% CI 0.17 to 0.60) compared with other oral diabetes drugs. Two later studies with stricter designs found smaller effects. Tang and colleagues (Movement Disorders, 2024) compared 30,091 new GLP-1 users with 58,983 new DPP4 inhibitor users aged 66 or older in US Medicare and reported a hazard ratio of 0.77 (95% CI 0.63 to 0.95). Engström and colleagues (Diabetes, Obesity and Metabolism, 2026) compared 158,961 new GLP-1 users with 188,065 new sulfonylurea users in Denmark, Norway and Sweden. Incidence was 5.2 versus 8.0 per 10,000 person-years, an adjusted hazard ratio of 0.81 (95% CI 0.68 to 0.96).
Adjusted ratio for a new Parkinson's diagnosis, GLP-1 users vs comparison drug
Brauer R et al., Brain 2020 (vs other oral drugs); Tang H et al., Mov Disord 2024 (vs DPP4 inhibitors); Engström A et al., Diabetes Obes Metab 2026 (vs sulfonylureas). A value of 1 means no difference.
Three cautions apply. First, these are associations in people with type 2 diabetes, and people prescribed a GLP-1 may differ from those prescribed older drugs in ways the statistics cannot fully correct. Second, the drugs were mostly older ones. In the Scandinavian study liraglutide made up 72.9% of GLP-1 follow-up time and semaglutide 13.4%. Third, not every cohort agrees. Lin and colleagues (JAMA Network Open, 2025) matched 60,860 US adults with type 2 diabetes and obesity and found a lower risk of dementia with semaglutide or tirzepatide (hazard ratio 0.63) but no significant difference in Parkinson's disease.
Preventing a diagnosis and slowing an established disease are also different questions. The cohort studies speak only to the first. Our article on GLP-1 microdosing and Alzheimer's covers the same gap between cohort data and trial results in dementia.
What side effects showed up in the Parkinson's trials?
Nausea was the most common side effect, and it rose with dose. In LIXIPARK, nausea occurred in 46% of participants on lixisenatide and vomiting in 13%. In the NLY01 trial, nausea was reported by 39% of participants on 2.5 mg and 58% on 5.0 mg, against 19% on placebo.
Participants reporting nausea (%)
McGarry A et al., Lancet Neurol 2024 (placebo and NLY01 arms, 36 weeks); Meissner WG et al., N Engl J Med 2024 (lixisenatide arm, 12 months). Separate trials, shown together for scale.
The dose pattern in NLY01 is the one piece of this evidence that touches on microdosing. A lower dose produced less nausea. But the same trial found no motor benefit at either dose, so there is no sign that a gentler dose keeps a benefit while shedding the side effects. Our guide to managing nausea on a GLP-1 microdose covers the general dose and nausea relationship.
Is weight loss from a GLP-1 a problem for someone with Parkinson's disease?
It can be, because low body weight is tied to worse survival in Parkinson's disease. Yoon and colleagues (Journal of the American Medical Directors Association, 2022) followed 2,703 Korean patients with newly diagnosed Parkinson's disease for 11 years. Mortality fell steadily as BMI at diagnosis rose, and a 10% change in BMI was also associated with mortality.
Hazard ratio for death by BMI at Parkinson's diagnosis
Yoon SY et al., J Am Med Dir Assoc 2022. Korean national cohort, each group compared with the study's reference BMI group. A value of 1 means no difference.
The authors caution that few participants in the cohort had obesity, so the lowest bar on the chart rests on a small group. The overall direction still matters. Weight loss is a frequent symptom of Parkinson's disease on its own, and in the 2013 exenatide study the investigators noted that weight loss was common among treated patients.
A person with Parkinson's disease who also has obesity or type 2 diabetes may still have good reasons to use a GLP-1. The point is that appetite suppression is not automatically a bonus in this condition. Our article on GLP-1 microdosing after 65 goes deeper on muscle, falls and why the scale can mislead.
Can a GLP-1 affect how levodopa is absorbed?
It may, and the evidence is thin. Levodopa tablets depend on the stomach passing them along. A systematic review by Heetun and Quigley (Parkinsonism and Related Disorders, 2012) reports that stomach emptying is frequently delayed in Parkinson's disease, and that this can reduce levodopa absorption and affect control of motor symptoms. The same review says the true prevalence of the problem is not established.
GLP-1 medicines slow stomach emptying too, which is part of how they reduce appetite. Whether the two effects add up in practice has not been measured in a dedicated study. The 2013 exenatide trial did report levodopa dose failures in a single patient. Our article on GLP-1 microdosing and gastroparesis explains the stomach-emptying effect in detail.
Questions worth bringing to a neurologist before starting any GLP-1 dose:
- •Has my weight been stable since diagnosis, and what is the lowest weight you would be comfortable with?
- •Do I already have signs of slow stomach emptying, such as early fullness or doses that seem not to kick in?
- •If my levodopa starts wearing off sooner or working less reliably, who should I call?
- •Is there a clinical trial of a GLP-1 drug in Parkinson's disease that I could join instead?
The idea that a GLP-1 drug could protect the brain in Parkinson's disease is serious science with mixed results: two encouraging mid-sized trials, then a larger phase 3 trial and a second study that found nothing. None used a microdose. For now, a GLP-1 in Parkinson's disease makes sense for weight or diabetes reasons agreed with a neurologist, not as a treatment for the disease itself.
Frequently asked questions
Can a GLP-1 microdose slow Parkinson's disease?
No trial has tested a GLP-1 microdose in Parkinson's disease, so there is no evidence that one slows it. The longest trial, a 96-week phase 3 study of exenatide 2 mg once weekly in 194 people, found no difference from placebo in motor scores. A smaller dose has even less data behind it.
Did exenatide work for Parkinson's disease in clinical trials?
The results conflict. A 2017 single-centre trial of 62 patients found motor scores 3.5 points better on exenatide than placebo at 60 weeks. The 2025 phase 3 trial of 194 patients across six UK hospitals found no benefit at 96 weeks, and its authors found no evidence that exenatide modifies the disease.
What did the lixisenatide trial find in early Parkinson's disease?
In the LIXIPARK phase 2 trial of 156 people, motor scores were essentially unchanged after 12 months of daily lixisenatide (a change of -0.04 points) and worsened by 3.04 points on placebo. Nausea affected 46% of the lixisenatide group and vomiting 13%. The authors said longer and larger trials are needed.
Do people who take GLP-1 medicines for diabetes get Parkinson's disease less often?
In three observational cohorts, yes. A Scandinavian register study found 5.2 versus 8.0 new diagnoses per 10,000 person-years for GLP-1 users versus sulfonylurea users, an adjusted hazard ratio of 0.81. Cohort studies cannot prove the drug caused the difference, and one US cohort of semaglutide and tirzepatide users found no significant difference.
Have semaglutide or tirzepatide been tested in Parkinson's disease?
None of the randomized trials reviewed in this article tested semaglutide or tirzepatide. They tested exenatide, lixisenatide and NLY01. A US health-record study of 60,860 adults with type 2 diabetes and obesity found a lower risk of dementia with semaglutide or tirzepatide but no significant difference in Parkinson's disease.
Is weight loss a concern for someone with Parkinson's disease?
It can be. In a Korean cohort of 2,703 people with newly diagnosed Parkinson's disease, a BMI under 18.5 at diagnosis carried a hazard ratio for death of 1.872, and higher BMI groups had lower mortality. Weight loss was also common in the first exenatide trial. Anyone with Parkinson's disease should discuss appetite-lowering medicines with their neurologist first.
Can a GLP-1 medicine interfere with levodopa?
It is a reasonable question to raise with a neurologist. A systematic review notes that slow stomach emptying in Parkinson's disease can reduce levodopa absorption and affect control of movement symptoms. In the 2013 exenatide trial, levodopa dose failures were reported in a single patient. Evidence on how often this happens is limited.
What side effects did people with Parkinson's disease report on GLP-1 drugs?
Stomach symptoms led the list. Nausea was reported by 46% of people on lixisenatide in LIXIPARK, and by 39% and 58% on the two NLY01 doses compared with 19% on placebo. In the phase 3 exenatide trial, serious adverse events were about as common on exenatide (9%) as on placebo (11%).
Clinical references
- Athauda D, Maclagan K, Skene SS, et al. Exenatide once weekly versus placebo in Parkinson's disease: a randomised, double-blind, placebo-controlled trial. Lancet. 2017. doi:10.1016/S0140-6736(17)31585-4
- Meissner WG, Remy P, Giordana C, et al. Trial of Lixisenatide in Early Parkinson's Disease. New England Journal of Medicine. 2024. doi:10.1056/NEJMoa2312323
- Vijiaratnam N, Girges C, Auld G, et al. Exenatide once a week versus placebo as a potential disease-modifying treatment for people with Parkinson's disease in the UK: a phase 3, multicentre, double-blind, parallel-group, randomised, placebo-controlled trial. Lancet. 2025. doi:10.1016/S0140-6736(24)02808-3
- McGarry A, Rosanbalm S, Leinonen M, et al. Safety, tolerability, and efficacy of NLY01 in early untreated Parkinson's disease: a randomised, double-blind, placebo-controlled trial. Lancet Neurology. 2024. doi:10.1016/S1474-4422(23)00378-2
- Aviles-Olmos I, Dickson J, Kefalopoulou Z, et al. Exenatide and the treatment of patients with Parkinson's disease. Journal of Clinical Investigation. 2013. doi:10.1172/JCI68295
- Mulvaney CA, Duarte GS, Handley J, et al. GLP-1 receptor agonists for Parkinson's disease. Cochrane Database of Systematic Reviews. 2020. doi:10.1002/14651858.CD012990.pub2
- Brauer R, Wei L, Ma T, et al. Diabetes medications and risk of Parkinson's disease: a cohort study of patients with diabetes. Brain. 2020. doi:10.1093/brain/awaa262
- Tang H, Lu Y, Okun MS, et al. Glucagon-Like Peptide-1 Receptor Agonists and Risk of Parkinson's Disease in Patients with Type 2 Diabetes: A Population-Based Cohort Study. Movement Disorders. 2024. doi:10.1002/mds.29992
- Engström A, Svanström H, Hviid A, et al. Use of Glucagon-Like Peptide-1 Receptor Agonists and Risk of Parkinson's Disease: Scandinavian Cohort Study. Diabetes, Obesity and Metabolism. 2026. doi:10.1111/dom.70760
- Lin HT, Tsai YF, Liao PL, et al. Neurodegeneration and Stroke After Semaglutide and Tirzepatide in Patients With Diabetes and Obesity. JAMA Network Open. 2025. doi:10.1001/jamanetworkopen.2025.21016
- Yoon SY, Heo SJ, Lee HJ, et al. Initial BMI and Weight Loss over Time Predict Mortality in Parkinson Disease. Journal of the American Medical Directors Association. 2022. doi:10.1016/j.jamda.2022.07.015
- Heetun ZS, Quigley EM. Gastroparesis and Parkinson's disease: a systematic review. Parkinsonism and Related Disorders. 2012. doi:10.1016/j.parkreldis.2011.12.004
Medical disclaimer. This article is for general educational and informational purposes only. It is not medical advice and not a substitute for a neurologist who knows your history. The trials described here used standard prescribed doses in a research setting and did not examine sub-label "microdose" regimens. The GLP-1 drugs discussed were studied as experimental treatments in Parkinson's disease, and the trial results are mixed. Do not start, change or stop any Parkinson's or GLP-1 medicine without talking to your prescriber. Charts show values reported in the cited sources.