FAIRPARK-II: deferiprone

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

David Devos (Lille), Olivier Rascol and the FAIRPARK-II study group; sponsored by University Hospital, Lille, with 25 sites in eight European countries listed in the registry (NCT02655315). New England Journal of Medicine, 1 December 2022.

What happened

Iron builds up in the brain area where dopamine neurons die, and excess iron can drive cell damage. Deferiprone binds iron so the body can excrete it. Two earlier small trials (40 people on stable dopamine drugs, 22 with early Parkinson's) suggested it lowered iron in some brain regions and might help. FAIRPARK-II, a Phase 2 trial that ran from February 2016 to September 2020, randomised 372 newly diagnosed people who had never taken levodopa to deferiprone (15 mg/kg twice daily, 186 people) or placebo (186) for 36 weeks. The main measure was the total MDS-UPDRS score at 36 weeks. It worsened by 15.6 points on deferiprone and 6.3 on placebo, a 9.3-point difference (95% CI 6.3 to 12.2; P < .001) in the wrong direction. Dopamine medication was started in 22.0% on deferiprone and 2.7% on placebo. MRI confirmed iron fell more in the drug group. Serious blood-cell side effects included two cases of agranulocytosis (a dangerous drop in white cells) and three of neutropenia.

SKY (Phase 2, NCT02728843). A separate Phase 2 tested deferiprone in people with early Parkinson's (under 3 years from diagnosis) who were already on stable dopamine medication. 140 people were randomised, 28 per group, to deferiprone 300, 600, 900 or 1,200 mg twice a day, or placebo, for 9 months. The main measure was change in MDS-UPDRS Part III (motor). Most doses worsened about as much as placebo. Only 600 mg twice daily looked better (2.8 points better than placebo; registry: -0.8 versus +2.1), and that was not statistically significant. A companion study, EMBARK (36 people, 27 on dopamine drugs and 9 not), found no significant change in those on dopamine drugs (-1.6) and significant worsening in the 9 who were not (+8.3 points). The authors concluded deferiprone gave no significant motor benefit.

Why it moved us toward a cure

It tested the long-standing finding of excess iron in the Parkinson's nigra (landmark 128). It showed that a drug can reach and change its target and still harm. The Barcelona co-investigators proposed that iron is needed by tyrosine hydroxylase, the enzyme that makes dopamine, so removing it early may reduce dopamine; they say this is a hypothesis, untested here. It also warns that results in people on levodopa may not carry over to people who are not.

Outcome

Dead end for deferiprone in early, untreated Parkinson's. The Cure Parkinson's trust is funding other iron-targeting agents and asking when in the disease iron removal might help.

Sources

• Devos et al., N Engl J Med 2022;387(22):2045–2055, DOI 10.1056/NEJMoa2209254
• Cure Parkinson's, Dec 2022
• Clinic Barcelona on the tyrosine hydroxylase hypothesis
• "Therapeutic modalities of deferiprone in Parkinson's disease: SKY and EMBARK studies," J Parkinsons Dis 2025, DOI 10.1177/1877718X241300295, PMID 39973479 · ClinicalTrials.gov NCT02728843 (SKY) (140 enrolled; completed 4 Sep 2019; results posted)
• Earlier pilots: Devos et al., Antioxid Redox Signal 2014;21:195-210, PMID 24251381 (40 people, 37 completed); Martin-Bastida et al., Sci Rep 2017;7:1398, PMID 28469157 (22 people, no significant motor benefit)
• FAIRPARK-II, NCT02655315 (372 enrolled; Phase 2; 9 Feb 2016 to 22 Sep 2020)
• Not yet verified: the Cure Parkinson's statement on funding other iron-targeting agents (not rechecked).