Parkinson's disease: treatment and what GLP-1 trials show

Complete guide: GLP-1 drugs: a complete guide to semaglutide and beyond

What is Parkinson's disease and how many people have it?

Parkinson's disease is a chronic neurodegenerative disorder that mainly affects movement and currently has no cure. The World Health Organization estimates that more than 8.5 million people were living with the disease in 2019; between 2000 and 2019 associated deaths rose by more than 100%, to 329,000, and disability-adjusted life years rose 81%, to 5.8 million. The WHO notes that prevalence has doubled in the past 25 years¹.

The Global Burden of Disease Study had estimated 6.1 million people with Parkinson's disease in 2016 (95% uncertainty interval 5.0 to 7.3) and concluded that the global burden more than doubled over a generation, driven by population ageing, with possible contributions from longer disease duration and environmental factors². A Lancet review describes it as a fast-growing condition whose rising prevalence resembles a pandemic, except for an infectious cause³.

In most populations, 3% to 5% of cases are explained by known monogenic Parkinson's genes, while 90 genetic risk variants collectively explain 16% to 36% of the heritable risk of non-monogenic disease. Having a relative with Parkinson's disease or tremor, constipation, and being a non-smoker each at least double the risk³.

Symptoms and diagnosis: how is Parkinson's disease diagnosed?

The WHO lists slowed movement, tremor, rigidity, walking difficulty and imbalance as motor symptoms; non-motor symptoms include cognitive impairment, mental health disorders, dementia, sleep disorders, pain and sensory disturbances¹. The clinical picture is multifaceted and may be preceded by a long prodromal period; for now, recognizing prodromal symptoms has no clinical implication other than suppressing symptoms³.

Diagnosis is clinical. Ancillary testing is reserved for atypical presentations³. The Movement Disorder Society (MDS) clinical diagnostic criteria define motor parkinsonism as bradykinesia plus rest tremor or rigidity. Once parkinsonism is documented, attributing it to Parkinson's disease relies on three categories of features: absolute exclusion criteria, which rule out the disease; red flags, which must be counterbalanced by additional supportive criteria; and supportive criteria, which increase confidence. The criteria delineate two levels of certainty, and their benchmark is expert clinical diagnosis⁴.

Current treatment of Parkinson's disease: what guidelines and evidence say

Treatment goals vary from person to person, and optimal management starts at diagnosis with a multidisciplinary team that includes non-pharmacological interventions. There is no reason to postpone symptomatic treatment once disability develops, and levodopa is the most common first-line medication³. The WHO describes levodopa/carbidopa as the most effective medicine for improving symptoms and notes it is not accessible, available or affordable everywhere, particularly in low- and middle-income countries¹.

The large, open-label, pragmatic PD MED trial compared dopamine agonists and MAO-B inhibitors with levodopa as initial treatment over the long term⁵. A later PD MED trial compared, in patients uncontrolled on levodopa, adjuvant COMT or MAO-B inhibitors with dopamine agonists and assessed long-term effectiveness⁶. These choices, and those for advanced disease, belong to each treating team and are not summarized here.

No available therapy can slow down or arrest the progression of Parkinson's disease. That statement from the Lancet review is the starting point for interest in peptides: disease-modifying strategies are being sought, informed by genetics and by mechanisms of neuronal death³.

What has peptide research shown in Parkinson's disease?

Glucagon-like peptide-1 (GLP-1) receptor agonists were developed for type 2 diabetes and showed neuroprotective properties in preclinical studies¹²¹³. This section separates each agent by the quality of its evidence in people.

Exenatide (human data, conflicting results). In a randomized, double-blind trial of 62 patients with moderate Parkinson's disease (48 weeks of treatment plus 12 weeks of washout), the off-medication MDS-UPDRS III motor score improved by 1.0 point with exenatide and worsened by 2.1 with placebo at 60 weeks; the adjusted difference was −3.5 points (p = 0.0318). The authors noted they could not distinguish an effect on the disease from a long-lasting symptomatic effect⁷. The phase 3 trial that followed, with 194 patients and 96 weeks at six UK hospitals, did not replicate it: off-medication MDS-UPDRS III worsened by 5.7 points with exenatide and 4.5 with placebo (coefficient 0.92; 95% CI −1.56 to 3.39; p = 0.47), and the authors concluded there is no evidence to support exenatide as a disease-modifying treatment for Parkinson's disease⁹. In June 2026 The Lancet published an expression of concern about that paper¹⁰.

Lixisenatide (human data, positive phase 2 trial). In LIXIPARK, 156 patients diagnosed with Parkinson's disease less than three years earlier received lixisenatide or placebo for 12 months. MDS-UPDRS III changed by −0.04 with lixisenatide and +3.04 with placebo, a difference of 3.08 points (95% CI 0.86 to 5.30; P = 0.007). Other secondary end points did not differ substantially; nausea occurred in 46% and vomiting in 13% of those given the drug⁸. It is an isolated signal, awaiting longer and larger trials.

NLY01 (human data, negative). NLY01 is a pegylated, longer-acting, brain-penetrant version of exenatide. In a 36-week randomized trial at 58 US clinics, with 255 patients with early untreated Parkinson's disease, the sum of MDS-UPDRS parts II and III did not differ from placebo: −0.39 (95% CI −2.96 to 2.18; p = 0.77) with 2.5 mg and 0.36 (95% CI −2.28 to 3.00; p = 0.79) with 5.0 mg. A subgroup analysis raised the possibility of motor benefit in younger participants, which the authors consider exploratory¹¹.

PT320 (animal only). PT320 is a sustained-release formulation of exenatide. In the MitoPark mouse, a progressive Parkinson's model, it preserved mitochondrial function and morphology and delayed dopamine degeneration¹⁵. These are animal-model data and do not replace a clinical result.

What the syntheses say. A meta-analysis of four placebo-controlled randomized trials found an overall improvement in on-medication MDS-UPDRS III (mean difference −0.75; 95% CI −1.50 to −0.0009; P = 0.04), but no significant improvement off medication (−1.39; P = 0.16) or in quality of life, and concluded that current evidence does not support routine use of GLP-1 drugs for motor improvement¹². A network meta-analysis of five trials (n = 708) found no significant overall improvement (mean difference −2.00; 95% CI −5.46 to 1.46; I² = 80.5%); its per-drug estimates favored exenatide and lixisenatide, but the authors conclude that evidence for broader clinical improvement is limited¹³.

Observational studies in type 2 diabetes. A meta-analysis of six cohort and case-control studies with 452,763 participants associated GLP-1 exposure with lower Parkinson's risk (HR 0.68; 95% CI 0.54 to 0.85), including versus DPP-4 inhibitors (HR 0.60) and metformin (HR 0.83). Its authors regard it as preliminary and hypothesis-generating because of residual confounding¹⁴.

The wider context, including the negative results of oral semaglutide in Alzheimer's disease (EVOKE and EVOKE+, 3,808 patients)¹⁷, is in GLP-1s, dementia and Alzheimer's: what EVOKE showed. A 2026 Science review adds that brain effects depend on each receptor's pharmacology, brain exposure and disease stage¹⁶.

Safety: adverse events observed in the trials

The most consistent adverse effect is gastrointestinal. In the meta-analysis of placebo-controlled trials, GLP-1 drugs were associated with more loss of appetite, nausea, vomiting, dyspepsia and reflux, and with more weight loss¹². In LIXIPARK nausea occurred in 46% and vomiting in 13%⁸; with NLY01 nausea was the most common event (39% at 2.5 mg and 58% at 5.0 mg, versus 19% with placebo) and there were no deaths¹¹. Weight loss is relevant in a disease in which weight can also be a problem and should be weighed by the treating team; the trials alone do not allow safety to be extrapolated to a new indication¹².

Summary by agent

Frequently asked questions

Is there an approved peptide for Parkinson's disease? No. No available therapy can slow the progression of the disease³, and the largest GLP-1 trials showed no benefit⁹¹¹.

Do GLP-1 drugs improve the motor symptoms of Parkinson's disease? Results are mixed: a phase 2 lixisenatide trial showed a 3.08-point difference⁸, while phase 3 exenatide and NLY01 did not⁹¹¹. The meta-analyses conclude the evidence does not support routine use¹².

Do GLP-1 drugs lower the risk of developing Parkinson's disease? In people with type 2 diabetes they were associated with fewer cases (HR 0.68), but these are observational studies and the authors call for trials to confirm a causal relationship¹⁴.

What is the most commonly used initial medication in Parkinson's disease? Levodopa is the most common first-line medication³; the WHO notes it is not available or affordable everywhere¹.

Why did the exenatide trial fail if the first one was positive? The first was small (62 patients) and its authors could not distinguish an effect on the disease from a long-lasting symptomatic effect⁷; the phase 3 trial was larger and longer and did not replicate the finding⁹.