The MMPOWER-3 trial did not show that elamipretide, also called SS-31, improved walking distance or fatigue compared with placebo after 24 weeks. Placebo is an inactive comparison treatment. Both groups changed by almost the same amount. A later genetic subgroup finding gave researchers a reason to run another trial, but it did not turn this one into a success.
That distinction matters if SS-31 has been described as an energy treatment. This study tested people with inherited muscle disease, not ordinary tiredness or healthy aging.
What question did the MMPOWER-3 trial ask?
Could targeting mitochondria, the parts of cells that make usable energy, help people walk farther and feel less fatigued?
The condition was primary mitochondrial myopathy: inherited muscle disease caused by faults in cellular energy production. Elamipretide is a peptide, a short chain of protein building blocks, studied for binding cardiolipin, a fat in the mitochondrial membrane.
A plausible mechanism gives researchers something to test. It does not answer whether a person feels or functions better. The wider SS-31 benefits evidence puts this trial beside studies in other conditions.
Who took part, and for how long?
The study enrolled 218 participants, with 109 assigned to elamipretide and 109 to placebo. Table 1 reports ages 16–75. Entry required a confirmed genetic cause and symptoms or examination findings showing that muscle disease was the main problem.
People walking less than 100 meters or more than 450 meters during screening or baseline testing were excluded. Other exclusions included pregnancy, breastfeeding and severely reduced kidney function. This was a selected disease population, not a sample of everyone with fatigue.
Participants received 40 mg daily under the skin or placebo for 24 weeks. That amount describes the study; the prescriber sets an individual prescription. Neither participants nor the study team knew which treatment each person received. This is what double-blind means.
The MMPOWER-3 registry record, NCT03323749, records the study plan and posted results. Stealth BioTherapeutics funded the trial.
What was measured, and what happened?
The two primary endpoints, the main outcomes chosen before the results were known, were:
- Six-minute walk distance: how far someone walked during a timed six-minute test.
- Primary Mitochondrial Myopathy Symptom Assessment Total Fatigue Score: four daily questions about tiredness and muscle weakness, at rest and during activity.
The fatigue total runs from 4 to 16. A lower score means fewer symptoms. Researchers compared changes from the starting measurements to week 24.
Walking distance improved by 14.1 meters with elamipretide and 17.3 meters with placebo. The adjusted difference, calculated using the study’s statistical model, was −3.2 meters. Its 95% confidence interval, a range describing uncertainty around the estimate, was −18.7 to 12.3 meters.
Fatigue also improved similarly in both groups. The full Results section reports changes of −1.13 points with elamipretide and −1.05 with placebo. The adjusted difference was −0.07 points, with a confidence interval of −0.69 to 0.54.
Neither result established a benefit. These figures come from Karaa and colleagues' Neurology report, published in 2023. Looking only at improvement from the starting point would miss the fact that placebo improved too.
What the side-effect table showed
An adverse event is any medical problem recorded during a trial, whether or not treatment caused it. Almost everyone receiving elamipretide reported one. Reactions where the injection went in were especially common.
Selected rows from Table 4 of the same paper follow. Each group contained 109 people.
| Recorded event | Elamipretide | Placebo |
|---|---|---|
| Any adverse event | 107 (98.2%) | 83 (76.1%) |
| Injection-site redness | 94 (86.2%) | 31 (28.4%) |
| Injection-site itching | 82 (75.2%) | 10 (9.2%) |
| Injection-site pain | 43 (39.4%) | 20 (18.3%) |
| Injection-site swelling | 42 (38.5%) | 7 (6.4%) |
| Serious adverse event | 5 (4.6%) | 3 (2.8%) |
| Event leading to treatment discontinuation | 8 (7.3%) | 2 (1.8%) |
Most events were mild or moderate. None of the serious events was judged treatment-related, and no deaths were reported. Treatment-related events leading to discontinuation involved injection sites. Calling treatment well tolerated should not hide how often those reactions happened.
Why the later genetic finding needs another trial
A genotype is a person's genetic makeup. Researchers revisited the data to ask whether particular genetic causes predicted different responses. The broad genetic comparison had been planned; the narrower 2024 analysis was post hoc, meaning done after the trial results were available.
One small subgroup had faults in genes that maintain mitochondrial DNA and progressive weakness of the muscles that move the eyes. Its 18 elamipretide participants improved walking distance by 37.3 meters; its 14 placebo participants changed by −8.0 meters.
The 2024 Orphanet Journal of Rare Diseases paper presents that signal as the basis for NuPOWER. Searching smaller groups after a negative result increases the chance of finding an apparent win by accident. A new trial must test the idea in people selected beforehand.
As of October 1, 2026, the day this article was written, the NuPOWER registry record lists 102 participants, completion on December 4, 2024, and no posted results. Its last posted update is October 24, 2025. Completion alone cannot confirm the subgroup finding.
What the earlier crossover trial contributed
MMPOWER-2, published in the Journal of Cachexia, Sarcopenia and Muscle in 2020, enrolled 30 people. Each received both treatments in separate four-week periods, with four weeks between them. That is a crossover design.
Its walking difference was 19.8 meters favoring elamipretide, but the confidence interval included no difference. Exploratory fatigue findings looked more encouraging. The MMPOWER-2 paper helped justify the larger study; MMPOWER-3 did not confirm a fatigue benefit across its full population.
What Forzinity changes for this result
As of October 1, 2026, the day this article was written, Forzinity has a separate label for improving muscle strength in people with Barth syndrome, a rare inherited mitochondrial disorder, weighing at least 30 kilograms. The FDA label reads “Revised: 9/2025.” The approval anniversary explains that decision, and the prescribing-information guide explains the label.
As of October 1, 2026, the day this article was written, a separate Barth syndrome follow-up study is underway. On July 8, 2026, Mighty Therapeutics announced that the first participant had been dosed. That development does not change MMPOWER-3's negative result.
SS-31 is coming soon to Promise — add me to the list. It will be dispensed as a compounded medication, which is different from an FDA-approved product: the formulation Promise will offer is not FDA-approved. The trial did not test that preparation.
MOTS-c also appears in mitochondrial discussions, but it is a different peptide studied as a signal through AMPK, the cell's fuel sensor. Its original Cell Metabolism study tested treatment in cells and mice. Those findings cannot supply the missing human benefit in MMPOWER-3.
What belongs in a prescribing conversation
The useful questions are whether the evidence matches the person's condition, what outcome would be measured, and how unwanted effects would be followed. General fatigue, healthy aging and inherited muscle disease are different clinical questions.
A licensed provider may still prescribe a compounded formulation; that decision is between the patient and the doctor. At Promise, a licensed provider reviews every SS-31 request and may prescribe or decline on medical eligibility; not everyone qualifies.
