Elamipretide

Does Elamipretide work?

Partly: some human trials, with gaps. The approval rests on a trial of 12 people.

The Barth syndrome approval is based on TAZPOWER, a 12-patient randomized crossover trial that missed its primary endpoints, followed by a 168-week open-label extension in which walking distance and strength improved against the natural history of the disease. ¹² The largest trial of elamipretide, MMPOWER-3 in 218 people with primary mitochondrial myopathy, failed both primary endpoints, so for the broader mitochondrial-health claims found online the evidence is negative. ⁴

TAZPOWER enrolled 12 patients aged 12 and older with genetically confirmed Barth syndrome. It was a double-blind crossover design. Patients took elamipretide for 12 weeks and placebo for 12 weeks, with a washout between. The blinded phase showed no significant differences on its main measures. These were the six-minute walk test and a fatigue questionnaire. Ten patients then entered a 168-week open-label extension. Eight completed it. They increased their six-minute walk distance by an average of 96 meters from baseline. Knee extensor strength rose over the same period. ¹

The FDA first declined the application in May 2025 because the controlled phase was negative. Stealth refiled under the accelerated approval pathway. It used knee extensor strength as a surrogate measure reasonably likely to predict benefit. The agency approved on that basis in September 2025. It required a post-approval randomized trial to confirm clinical benefit. ²

Read the full Elamipretide page

Reference information only, not medical advice. Drawn from the Elamipretide article, fact-checked against the cited sources. No dosing anywhere on this site.