The approval's evidentiary base is more layered than 'a trial succeeded.' The core study, TAZPOWER, randomised 12 genetically confirmed Barth syndrome patients (aged 12 and older, weighing more than 30 kg) to elamipretide 40 mg subcutaneously once daily or placebo in a crossover design, each period lasting 12 weeks. On the trial's primary endpoints — distance walked on the six-minute walk test and the Total Fatigue Score on the Barth Syndrome Symptom Assessment — elamipretide was not superior to placebo.[5]
Knee extensor muscle strength was only a secondary endpoint in that blinded comparison, and it did not separate from placebo there either. Ten of the twelve patients then entered a single-arm, open-label extension, continuing the same 40 mg daily dose; eight reached the week-168 visit. It was during this uncontrolled extension period that strength measurements rose: median muscle strength increased by 34 newtons at week 12 of the extension and by 63 newtons at week 168, against a pre-dose baseline median of 124 newtons — the open-label data behind the accelerated approval.[5]
A separate analysis strengthens that open-label signal with an external comparison. Researchers built a propensity-matched natural-history control group of 19 untreated Barth syndrome patients and compared them with the 8 patients who stayed on elamipretide through the extension. Relative to the untreated group, elamipretide was associated with a 79.7-metre greater six-minute-walk distance at week 64 and 91.0 metres at week 76 (both p<.001), alongside a 40.8-to-56.7-newton greater gain in muscle strength over the same period. A 2024 publication of the full 168-week extension also reported a cumulative 96.1-metre improvement in walk distance from the extension's own baseline (p=.003), and reduced fatigue scores at every extension time point.[8][7]
Put plainly: the randomised, blinded portion of the pivotal trial did not beat placebo on any endpoint. The approval rests on what happened afterward, without a placebo group, reinforced by a retrospective comparison against patients who were never treated — real evidence, but a lower tier than a positive randomised trial, which is why FDA required a confirmatory trial as a condition of continued approval. That confirmatory trial, a Phase 3b/4 randomised, double-blind, placebo-controlled study named 4TAZPower (72 weeks), began recruiting in July 2026.[5][15]