Parkinson's Cure Journey

Landmark 101 · Trial in people

Exenatide pilot

A small study in 44 people, with no placebo injections. It seemed to help, but the authors said it was not proof. The larger trial later found no benefit.

In short

A 44-patient trial with no placebo injections suggested the diabetes drug exenatide helped movement and thinking in Parkinson's, which launched the controlled trials in landmark 20; the large Phase 3 later found no benefit.

What it is
Trial in people
Outcome
Replaced by a newer approach
When
2013
Approach
Look beyond the brain
Research lane
Metabolic drug repurposing
Trial registry
NCT01174810
Last checked
19 Oct 2026

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Iciar Aviles-Olmos, Tom Foltynie and colleagues, including Patricia Limousin and Andrew Lees, at the National Hospital for Neurology and Neurosurgery and University College London.

What happened

Forty-five people with moderate Parkinson's were randomised (21 to exenatide, 24 to controls); 44 were analysed. 20 had twice-daily exenatide injections (10 micrograms) for 12 months on top of usual medicines, and 24 had usual medicines alone. Patients knew their treatment; only the raters were blinded. Registered as NCT01174810 (UCL, funded by the Cure Parkinson's Trust). Exenatide is a GLP-1 drug (a gut-hormone mimic used for type 2 diabetes) that had protected dopamine neurons in animal models. Raters scored blinded video recordings of the patients. At 12 months, the movement score (MDS-UPDRS part 3, off medication) improved 2.7 points on exenatide and worsened 2.2 in controls (difference 4.9, P = 0.037). A Mattis cognitive scale also favoured exenatide at 14 months (difference 6.3, P = 0.001), two months after the drug stopped. The Mattis difference had a 95% CI of 2.7–9.9. The exenatide group lost more weight (3.2 kg versus 0.8 kg), and 3 of them withdrew (dose failures of levodopa, taste change with worsening symptoms, excessive weight loss); one control also withdrew. A follow-up, 12 months after exenatide stopped, reported advantages of 5.6 motor points (95% CI 2.2–9.0, P = 0.002) and 5.3 points on the Mattis scale (95% CI 1.4–9.3, P = 0.006) (J Parkinsons Dis 2014).

Why it moved us toward a cure

People with type 2 diabetes have a higher risk of Parkinson's (landmark 216). It was the first human signal that a repurposed drug might act on the disease rather than just the symptoms, and it justified the double-blind Phase 2 (landmark 20).

Outcome

Superseded. The authors themselves wrote that the data should not be read as evidence of symptom benefit or neuroprotection: they could not exclude placebo effects, controls were not given placebo injections, and a two-month washout cannot rule out a symptom effect. The larger placebo-controlled Phase 3 (194 people, 96 weeks) showed no difference: the off-medication motor score worsened by 5.7 points on exenatide and 4.5 on placebo (adjusted difference 0.92 in placebo's favour, 95% CI −1.56 to 3.39, P = 0.47; Lancet 2025;405:627–636).

Sources

How it connects

Built on

Led to

See it on the research map

Report an error on this page

Back to top