Caffeine May Improve Motor Symptoms in Parkinson's

Susan Jeffrey

August 02, 2012

August 2, 2012 — A randomized trial shows improvement in objective motor measures with caffeine treatment over placebo in patients with Parkinson's disease (PD).

These improvements were significant, although treatment led to only an "equivocal borderline improvement" in the study's primary endpoint of change in excessive daytime somnolence in these patients, investigators say.

"What was most interesting, I think, is we found that there was a motor benefit on the Parkinson's disease, so that this is a potential treatment for Parkinson's," first author Ron B. Postuma, MD, from Montreal General Hospital, McGill University, in Montreal, Quebec, told Medscape Medical News.

"It was a modest improvement, but it was certainly there," he added, amounting to an approximately 5-point improvement on the Unified Parkinson's Disease Rating Scale (UPDRS).

However, the study was short term. "What we need to do is see if we can see this effect in the longer term, because if we do, we have a new agent for Parkinson's disease, essentially," Dr. Postuma said.

These findings were published online August 1 in Neurology. They were presented in June at the Movement Disorder Society's 16th International Congress of Parkinson's Disease and Movement Disorders, and were reported by Medscape Medical News at that time.

Epidemiologic Evidence

Previous evidence from epidemiologic studies has consistently shown that caffeine, which acts as an adenosine 2A (A2A) antagonist, is associated with lower risk of Parkinson's disease, the authors write.

"We've been very intrigued by the fact that there's a relation between non-use of caffeine and Parkinson's disease," Dr. Postuma said. It's possible that caffeine is protective against the disease, but it's also possible alternatively, that in those with Parkinson's, caffeine is not tolerated, it doesn't work for somnolence, or perhaps caffeine is actually treating symptoms of the disease.

In this study, the primary outcome of interest was change in daytime somnolence, as measured on the Epworth Sleepiness Scale, but researchers also looked at a variety of secondary endpoints, including motor severity, sleep markers, fatigue, depression, and quality of life.

A total of 61 patients with PD with excessive daytime somnolence, defined as an Epworth Scale score greater than 10, were randomly assigned to receive 100 mg twice daily for 3 weeks, then 200 mg twice daily for an additional 3 weeks, of caffeine or placebo. Effects were analyzed by Bayesian hierarchical models, with adjustment for study site, baseline scores, age, and sex.

"There was some clear improvement in their somnolence, it just didn't reach statistical significance," Dr. Postuma said. "Basically, there were a few protocol violations which prevented a positive effect, but at any rate, the somnolence effect was modest."

"What was most interesting, I think, is we found that there was a motor benefit on the Parkinson's disease," he added.

The primary endpoint at 6 weeks showed a nonsignificant reduction in the Epworth Scale score with caffeine, but there was improvement on the Clinical Global Impression–Change, and a significant reduction in somnolence in the per-protocol analysis.

Treatment with caffeine was associated with a significant change on the UPDRS and on the objective UPDRS motor component.

Table. Change in Primary and Secondary Endpoints With Caffeine vs Placebo in PD

Endpoint Change, points 95% Confidence Interval
Change in Epworth (ITT) — primary endpoint -1.71 -3.57 to 0.13
Somnolence on Clinical Global Impression–Change (ITT) +0.64 0.16 to 1.13
Change in Epworth (per protocol) -1.97 -3.87 to -0.05
Change in UPDRS -4.69 -7.7 to -1.6
Change in UPDRS (motor component) -3.15 -5.50 to -0.83

ITT = intention-to-treat; PD = Parkinson's disease; UPDRS = Unified Parkinson's Disease Rating Scale.

Analysis of UPDRS subcomponents showed significant effects of caffeine on bradykinesia and rigidity, but no increase in action tremor.

Other than some "modest" improvement in global health measures, there was no change in quality of life, measures of depression, or sleep quality, the authors note. Adverse event frequency was similar between groups.

Caffeine is often associated with some irritability, anxiety, or difficulty sleeping, Dr. Postuma noted, "but we really didn't find that was the case in our study. We found it was very, very well tolerated." However, they did give caffeine doses in the morning and at lunch. "Obviously, if you take it in the evening, it's going to impact upon your sleep."

The potential motor benefits suggest that a larger, long-term trial of caffeine is warranted, the authors conclude.

Currently, several other A2A antagonists are in clinical trials as treatment for Parkinson's disease, Dr. Postuma noted, "but these are going to cost thousands of dollars a year of course, while caffeine is going to cost pennies."

Cruise Control

In an editorial accompanying the publication, Michael A. Schwarzschild, MD, PhD, from the Department of Neurology at Massachusetts General Hospital, in Boston, points out that the ability of caffeine — "the world's most widely used psychomotor stimulant" — to potentiate the antiparkinsonian effects of levodopa was observed 40 years ago.

Clinical trials based on that observation failed to show any motor effect of caffeine beyond producing dyskinesia. However, these early studies used very high doses, around 1100 mg/day, Dr. Schwarzschild notes. Subsequent studies at lower doses showed that caffeine improved freezing of gait, he writes, "though tolerance to caffeine seemed to limit benefit."

This new study, although failing to show a significant benefit on excessive daytime sleepiness and, in fact, "providing Class I evidence against such an indication in PD," he notes, did show benefit on parkinsonian dysfunction, including the objective motor component and subscores for bradykinesia and rigidity. Similar findings were seen at Week 3 with the lower dose.

Limitations of the study include the "exploratory nature" of the findings because motor scores were not the primary endpoint, the possibility of incomplete blinding, and the short duration of the study, "leaving open the question of tolerance to caffeine," he writes. "Nevertheless, the findings are noteworthy, the first to suggest antiparkinsonian effects of caffeine in a randomized clinical trial."

This potential of caffeine as an antiparkinsonian agent supports progress made with several other, more specific A2A antagonists, including istradefylline, preladenant, and tozadenant, now in phase 2 and phase 3 trials, but calls into question whether the huge investment in agents with greater specificity will produce clinically meaningful advantages over caffeine.

"Such benefits should be substantial to offset the unmatchable advantages of caffeine's long-term safety experience and cost," Dr. Schwarzschild writes. "Moreover, as the authors note, their preliminary findings that caffeine improved total UPDRS score by 4–5 points, if substantiated, may be comparable to UPDRS improvements achieved to date with specific A2A antagonists."

Still, the more specific agents provide theoretical advantages, including the possibility that they will not produce tolerance, as has been shown in preclinical studies.

"Ultimately, head-to-head comparisons may be required to distinguish the utility of A2A-specific and mixed adenosine receptor antagonists for treating the motor symptoms or other features of PD," he writes. "For the time being, the results of Postuma et al. should encourage further investigation of a potential antiparkinsonian ('cruise control') benefit of caffeine without entirely discouraging pursuit of its putative alerting ('snooze patrol') action in PD.

"Although current data do not warrant a recommendation of caffeine as a therapeutic intervention in PD, they can reasonably be taken into consideration when discussing dietary caffeine use," he concludes.

This study was funded by grants from the Canadian Institute of Health Research and the Webster Foundation. The authors and the editorialist have disclosed no relevant financial relationships.

Neurology. 2012;79:651-658, 616-618. Published online August 1, 2012. Abstract Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: