Idebenone
Idebenone, short-chain synthetic benzoquinone analogue of coenzyme Q10
Written by Aaron CuhaReviewed Sep 2026
Also known as: Raxone, Catena, Sovrima, CV-2619
Authorised in the European Union for one rare mitochondrial optic neuropathy, granted under exceptional circumstances because the normal evidence standard was not met. The randomised trial that supported it missed its primary endpoint, and the trials in Alzheimer's disease and Friedreich ataxia were negative.
Overview
Idebenone is three separate products wearing one name, and they should not be read as supporting each other.
The medicine. Raxone received an EU marketing authorisation on 8 September 2015 for visual impairment in adolescent and adult patients with Leber hereditary optic neuropathy, and it was granted under exceptional circumstances, a category meaning the applicant could not provide comprehensive efficacy and safety data and was not expected to be able to. The supporting randomised trial, RHODOS, enrolled 85 patients on 900 mg a day for 24 weeks and did not reach statistical significance on its primary endpoint, best recovery in visual acuity, in the intention to treat population. A post hoc analysis found the benefit concentrated in patients with discordant visual acuities between eyes at baseline. A later non-randomised controlled trial, LEROS, published in 2024, is the other main efficacy dataset. It is not FDA approved for anything.
The nootropic. Idebenone was tested properly in Alzheimer's disease: 536 patients, one year, three doses against placebo, multicentre, double blind. There were no significant differences on either primary outcome in the prespecified four-group design, and the paper's title is the conclusion: idebenone treatment fails to slow cognitive decline in Alzheimer's disease. It was also taken through a 70-patient phase 3 in Friedreich ataxia, where neurological function did not differ significantly from placebo at 24 weeks.
The cosmetic. Topical idebenone at 0.5 and 1.0 percent was tested in 41 women with photodamaged skin for six weeks, with improvements in roughness, hydration and fine lines. The study had no vehicle control, meaning there was no way to separate the active ingredient from the lotion base it was carried in.
Mechanism of action
A short-chain benzoquinone that can accept electrons from cytosolic NAD(P)H quinone dehydrogenase 1 and deliver them to complex III, bypassing a defective complex I. That is a coherent rationale for a disease caused by complex I mutations, which is what Leber hereditary optic neuropathy is, and it is why the drug found its indication there rather than in general antioxidant use. The reduction step depends on NQO1: in NQO1-deficient cells idebenone has been reported to cause cell death rather than protect, so the mechanism is conditional on the tissue's enzyme status rather than being a generic antioxidant effect.
Human evidence
Unusually for this site, idebenone has been through several proper randomised trials. Most of them were negative, and the one approval it holds was granted on a trial that missed its primary endpoint.
- RHODOS, 85 patients with Leber hereditary optic neuropathy, 900 mg a day, 24 weeks: primary endpoint not significant in the intention to treat population; benefit found post hoc in the subgroup with discordant baseline acuities.
- LEROS, published 2024: a nonrandomised controlled trial, the other dataset supporting the EU indication.
- Alzheimer's disease, 536 patients, one year, three doses: no significant effect on either primary outcome.
- Friedreich ataxia, 70 paediatric patients, 24 weeks: 2.5 versus 1.3 point improvement on the ataxia rating scale, not statistically significant.
- Topical photodamaged skin, 41 women, six weeks: improvements in graded roughness, wrinkles and hydration, with no vehicle control.
What this does not tell you: The approved indication rests on a randomised trial that missed its primary endpoint plus a non-randomised follow-up, which is exactly why the authorisation carries the exceptional circumstances qualifier. Nothing in this record supports idebenone for cognition in healthy people: the one adequately powered cognitive trial, in Alzheimer's disease, failed. The skin study is six weeks, small, industry-formulated and uncontrolled for vehicle, which is the weakest common design in cosmetic dermatology. And no trial has tested idebenone in healthy adults for energy, ageing or general mitochondrial support, which is what the supplement market sells it for.
Reading the research record
Idebenone is the most useful case study in this batch, because it shows what happens when a mitochondrial antioxidant is actually put through the regulatory process.
It was tested in three serious indications. In Alzheimer's disease, in a 536-patient one-year trial, it failed. In Friedreich ataxia, in a phase 3, it failed on neurological function. In Leber hereditary optic neuropathy, a disease caused directly by complex I mutations, where the drug's electron-bypass mechanism is specifically applicable, it missed its primary endpoint and was approved anyway in Europe, under a category that exists precisely because the evidence was known to be incomplete. A rare, blinding, otherwise untreatable disease is the right place to accept that trade. A healthy adult buying a nootropic is not in that situation.
The mechanism also argues against generic use. Idebenone's reduction depends on NQO1, and in NQO1-deficient cells it has been reported to cause cell death rather than protect. That is not the profile of a broad-spectrum antioxidant, it is a conditional electron carrier whose benefit depends on the tissue biochemistry it lands in.
The skincare claim should be evaluated on its own terms and not on the drug approval. One six-week study in 41 women, with no vehicle control, is not a strong basis for a premium price, and the authors' own framing is that no clinical study had previously been done on topical idebenone.
The evidence, charted
Fig. 1 · evidence composition
5of 6 citations (83%) are in people
Counted from the citation list on this page. The Human count is the same number shown in the badge at the top. Both understate any literature larger than the sources we cite.
Fig. 2 · evidence over time
Evidence spans 6 distinct years, 2003 to 2024, counted from the citation list on this page.
Fig. 3 · legal status at a glance
US
Not approved
UK
Prescription
AU
Not approved
CA
Not approved
Approved in 0 of 4, prescription route in 1, not approved in 3. A jurisdiction's classification is a regulatory fact, not a verdict on the science; see Legal status below for the full text and any notes.
Key studies & citations
- Human2011
A randomized placebo-controlled trial of idebenone in Leber's hereditary optic neuropathy
RHODOS. 85 patients with m.3460G>A, m.11778G>A or m.14484T>C mutations, idebenone 900 mg a day for 24 weeks, multicentre, double blind, placebo controlled. The primary endpoint, best recovery in visual acuity, did not reach statistical significance in the intention to treat population. A post hoc interaction analysis found significant secondary endpoint differences in patients with discordant visual acuities at baseline. Idebenone was safe and well tolerated.
Brain - Human2024
Therapeutic benefit of idebenone in patients with Leber hereditary optic neuropathy: The LEROS nonrandomized controlled trial
The other main efficacy dataset for the approved indication. Note the word nonrandomized in the title: it is a controlled trial against an external natural history comparison, not a randomised one.
Cell Reports Medicine - Human2003
Idebenone treatment fails to slow cognitive decline in Alzheimer's disease
536 patients with probable Alzheimer's disease, one year, multicentre, double blind, randomised, idebenone 120, 240 or 360 mg three times daily versus placebo. No significant differences between groups on either primary outcome, ADAS-Cog and Clinical Global Impression of Change, in the prespecified four-group design, and no differences on any secondary outcome. This is the trial that tests the nootropic claim, and it is negative.
Neurology - Human2010
A phase 3, double-blind, placebo-controlled trial of idebenone in friedreich ataxia
70 ambulatory patients aged 8 to 18, randomised to lower dose idebenone, higher dose idebenone or placebo for 24 weeks. Idebenone improved the ataxia rating scale by 2.5 points against 1.3 on placebo, a difference that was not statistically significant, and the secondary scale showed the same pattern. Idebenone did not significantly alter neurological function.
Archives of Neurology - Human2005
Clinical efficacy assessment in photodamaged skin of 0.5% and 1.0% idebenone
41 women aged 30 to 65 with moderate photodamage used 0.5 or 1.0 percent idebenone lotion twice daily for six weeks. With the 1.0 percent formula, graders scored a 26 percent reduction in roughness and dryness, 29 percent reduction in fine lines and wrinkles, 37 percent increase in hydration and 33 percent global improvement. The paper describes itself as a nonvehicle control study, so the lotion base itself was never controlled for.
Journal of Cosmetic Dermatology - Review2015
Raxone (idebenone): EU marketing authorisation under exceptional circumstances
Authorised 8 September 2015 for visual impairment in adolescent and adult patients with Leber hereditary optic neuropathy, under exceptional circumstances, a category used when comprehensive efficacy and safety data cannot be provided under normal conditions of use.
European Medicines Agency medicine page
Frequently asked questions
Is idebenone FDA approved?
No. It has no FDA approval for any indication. It holds a European authorisation as Raxone for Leber hereditary optic neuropathy, granted under exceptional circumstances, which is a specific category meaning comprehensive efficacy and safety data could not be provided.
Does idebenone work as a nootropic?
The only adequately powered cognitive trial, in 536 patients with Alzheimer's disease over a year, found no significant effect on either primary outcome. There is no trial of idebenone for cognition in healthy adults at all.
Is idebenone better than CoQ10?
It is a shorter-chain synthetic analogue that is absorbed differently and works by a different route, accepting electrons via NQO1 to bypass complex I. For the one disease where that mechanism applies it reached a European authorisation, which CoQ10 has not. That is not the same as being a better general supplement, and no head-to-head trial in healthy people exists.
Does idebenone cream reduce wrinkles?
One six-week study in 41 women reported graded improvements in fine lines, roughness and hydration at 0.5 and 1.0 percent. It had no vehicle control, so the lotion base was never separated from the ingredient. Treat it as suggestive, not established.
Is idebenone safe?
In trials at 900 mg a day it was described as safe and well tolerated, and doses well above that have been used in Friedreich ataxia. The relevant caution is mechanistic rather than a documented clinical harm: its activation depends on NQO1, and in NQO1-deficient cells it has been reported to cause cell death.