Background: Preladenant is an adenosine 2A (A₂(A)) receptor antagonist. In animal models of Parkinson's disease, preladenant monotherapy improves motor function without causing dyskinesia and, as an adjunct to levodopa, it improves motor function without worsening dyskinesia. We aimed to assess the efficacy and safety of preladenant in patients with Parkinson's disease and motor fluctuations who were receiving levodopa and other antiparkinsonian drugs.

Methods: In this phase 2, dose-finding trial, patients with Parkinson's disease who were receiving levodopa were enrolled and treated at 44 sites in 15 countries between December, 2006, and November, 2008. Assignment to treatment was done centrally with an interactive voice response system, according to a block randomisation schedule that was computer generated by the sponsor. Patients were assigned to receive 1, 2, 5, or 10 mg oral preladenant twice daily, or matching placebo for 12 weeks. Patients, study staff, investigators, and all sponsor personnel were masked to treatment assignment. The primary outcome was change in mean daily off time from baseline to week 12, as assessed by home diaries. Efficacy analysis included all patients who received at least one dose of study drug and had data for assessments after baseline. This trial is registered with ClinicalTrials.gov, number NCT00406029.

Findings: 253 patients were randomised to receive preladenant (1 mg [n=49], 2 mg [n=49], 5 mg [n=49], 10 mg [n=57]) or placebo (n=49), of whom 234 on preladenant (1 mg [n=47], 2 mg [n=48], 5 mg [n=45], 10 mg [n=49]) and placebo (n=45) were eligible for the efficacy analysis. Mean daily off time from baseline to week 12 was reduced versus placebo in patients on 5 mg preladenant (difference -1·0 h, 95% CI -2·1 to 0·0; p=0·0486) and 10 mg preladenant (-1·2 h, -2·2 to -0·2; p=0·019). Changes in mean daily off time versus placebo were not significant for 1 mg preladenant (0·2 h, -0·9 to 1·2; p=0·753) or 2 mg preladenant (-0·7 h, -1·7 to 0·3; p=0·162). The most common adverse events in the combined preladenant group versus placebo were worsening of Parkinson's disease (22 [11%] vs 4 [9%]), somnolence (20 [10%] vs 3 [6%]), dyskinesia (18 [9%] vs 6 [13%]), nausea (17 [9%] vs 5 [11%]), constipation (15 [8%] vs 1 [2%]), and insomnia (15 [8%] vs 4 [9%]).

Interpretation: 5 and 10 mg preladenant twice daily might be clinically useful to reduce off time in patients with Parkinson's disease and motor fluctuations.

Funding: Schering-Plough, a subsidiary of Merck.