Do GLP-1 medications help with Parkinson's disease?
GLP-1 agents are showing preliminary positive signals on motor decline in Parkinson's disease, but the studies are still too small and too short-term for a clear conclusion. If you have Parkinson's disease or diabetes and want to know whether this is relevant for you, discuss it with your neurologist.
Animal and laboratory research shows that GLP-1 agents can inhibit several harmful processes in the brain that play a role in Parkinson's disease, such as inflammation, oxidative stress and the accumulation of the protein alpha-synuclein. GLP-1 receptors are present in the brain and the agents can cross the blood-brain barrier. That is the biological reason why researchers believe a disease-modifying effect may be possible, but animal results do not translate to humans nearly as often as one might hope.
In humans, two notable results have now emerged. Exenatide (weekly injection) improved motor scores by an average of 3.5 points more than placebo after 60 weeks, and that effect remained visible after people had stopped for twelve weeks. Lixisenatide (daily injection) showed almost no decline in people with early Parkinson's disease (-0.04 points), while the placebo group worsened by 3.04 points. Here too, scores remained more favourable after two months of stopping. Both outcomes are statistically significant and clinically striking, but the studies are relatively small.
When the available randomised studies are taken together, the overall picture is more nuanced. In the on-medication state, there is a small but statistically significant benefit on motor scores. Off medication, the effect is not consistent across all time points, except at 60 weeks. For outcomes that patients notice directly, such as quality of life and how much Parkinson's medication they need, no significant benefits have been found. The researchers conclude that the evidence is insufficient for routine use.
In a large Scandinavian study of patients with diabetes, users of GLP-1 agents had an incidence of 5.2 new Parkinson's cases per 10,000 person-years, compared with 8.0 per 10,000 person-years among users of another diabetes medication. This is an interesting association, but because people were not randomly assigned here, cause and effect cannot be proven.
An important point of concern is safety. Nausea occurred in 46% of lixisenatide users (versus a lower percentage with placebo), and vomiting in 13%. All GLP-1 agents cause gastrointestinal problems and weight loss significantly more often. In people with Parkinson's disease, who already have an increased risk of malnutrition, this is especially relevant to monitor. A Japanese study into an oral form (semaglutide) is still ongoing, but results are not yet available.
This answer is based on two randomised studies in Parkinson's patients (62 and 156 participants respectively), a pooled summary of four such studies, a large Scandinavian observational study in patients with diabetes, and animal and laboratory studies. The randomised studies are too small for firm conclusions about disease modification; the four studies combined show inconsistent effects depending on the time point and the patient's condition. The observational study cannot establish cause and effect.