SS-31 dosage: four trial protocols, zero guidance for you

Elamipretide, the drug name for SS-31, just crossed a line almost nobody asking about its dosage seems to have noticed. In September 2025 the FDA granted it accelerated approval, its first anywhere, for Barth syndrome, a rare genetic disease most people searching “SS-31 dosage” have never heard of^[1]. The dose behind that approval is the same 40 mg subcutaneous injection researchers have used since the drug’s earliest trials. Nothing about the approval tells you anything new about using SS-31 for mitochondrial myopathy, heart attack recovery, macular degeneration, or general-purpose anti-aging use, because none of those are what got approved. If you are asking about SS-31 dosage for any reason other than Barth syndrome, the honest answer is that no dose has been shown to work for you, approved or not.

What four controlled trials tested, at four different doses

SS-31, known in trials as elamipretide, has been tested against placebo in randomized controlled trials across several conditions, most substantially primary mitochondrial myopathy, Barth syndrome, and age-related macular degeneration^[2]. Every completed human dataset on elamipretide comes from a randomized, placebo-controlled trial, not from forum reports or compounding-pharmacy protocols. That is unusually clean for a peptide with this much online chatter. It is also why the dosing picture is so narrow: four trials, four fairly different jobs for the drug, and one dose that keeps reappearing.

MMPOWER-2, a phase 2 crossover trial in 30 adults with genetically confirmed primary mitochondrial myopathy, gave 40 mg of subcutaneous elamipretide once daily for four weeks, then placebo for four weeks (or the reverse order), with a four-week washout between^[3]. The six-minute walk test improved by 19.8 meters over placebo, but the result missed statistical significance (95% CI -2.8 to 42.5, p = 0.08)^[3].

MMPOWER-3, the phase 3 follow-up, scaled that same 40 mg daily subcutaneous dose up to 218 adults for 24 weeks^[4]. It did worse on the primary endpoint, not better: the six-minute walk test difference came in at -3.2 meters (95% CI -18.7 to 12.3, p = 0.69), and the fatigue score difference was likewise non-significant^[4]. A bigger, longer, better-powered trial of the identical dose produced a flatter result. That alone should temper any assumption that “more data” automatically means “clearer benefit.”

TAZPOWER, the Barth syndrome trial, used the same 40 mg daily subcutaneous dose in just 12 patients, crossed over between drug and placebo for 12 weeks each^[5]. The blinded comparison met neither of its primary endpoints. Only in the subsequent open-label extension, with everyone now knowingly on the drug, did the six-minute walk test improve by 95.9 meters at 36 weeks (p = 0.024)^[5].

EMBRACE STEMI is the outlier on route entirely. Patients having a heart attack got intravenous MTP-131 (elamipretide’s earlier name), infused at 0.05 mg per kilogram per hour for one hour around the time of their emergency angioplasty^[6]. Infarct size, measured by CK-MB enzyme release over 72 hours, did not differ from placebo^[6].

Four elamipretide dosing protocols

All data from randomized, placebo-controlled trials.

40 mg the subcutaneous dose used in every chronic elamipretide trial, from the first mitochondrial myopathy study through the drug now approved for Barth syndrome PMID 32096613

Nobody titrated. Nobody self-adjusted based on response. Every arm in every trial got a fixed dose assigned by protocol, not by body weight, genotype, or how the participant felt partway through. That is worth sitting with before anyone extrapolates a dose from these numbers to their own situation. The only variable that changed across trials was the route and the disease being targeted. The amount stayed fixed at 40 mg.

An interventional record, with one retrospective exception

Most of what we know about elamipretide comes from exactly the design that should tell you about cause and effect: a randomized trial compares a group that got the drug to a group that, by chance alone, did not. That lets researchers say a dose caused an outcome, at least within that trial’s dose and endpoint.

There is one real exception, and it matters more than it sounds like it should. A separate natural history comparison study, explicitly described as retrospective and non-interventional, took 8 Barth syndrome patients who had continued on open-label elamipretide through the TAZPOWER extension and compared them against 19 untreated patients, using propensity scoring rather than randomization^[7]. That comparison found a six-minute walk test difference of 79.7 meters in favor of the treated group at week 64 (p = 0.0004)^[7].

The approval leans on the part of the trial that was not blinded. TAZPOWER’s actual randomized, placebo-controlled comparison met neither primary endpoint. The number regulators and patients are pointing to, a near-80-meter walking improvement, comes from comparing openly treated patients against an untreated historical group, not from the part of the study designed to rule out placebo effect and rater bias.

That is not a condemnation of the drug. Barth syndrome is rare enough, and progressive enough, that a natural history comparison is a defensible substitute when enrolling a large enough blinded cohort isn’t realistic. But it is a different kind of evidence than the randomized result, and conflating the two overstates what has been shown.

The cardiolipin mechanism, demonstrated in rats and dishes, not people

SS-31’s rationale rests on one specific biochemical claim: it binds cardiolipin, a phospholipid unique to the inner mitochondrial membrane. Researchers demonstrated this directly using a fluorescent SS-31 analogue in isolated mitochondria^[8]. Cardiolipin binding is proposed to stabilize inner mitochondrial membrane cristae structure, a mechanism supported by the same study’s finding that pretreating rats with SS-31 protected mitochondrial cristae membranes and prevented mitochondrial swelling during induced kidney ischemia, rather than in isolated human mitochondria^[8]. That is isolated-organelle and rat evidence. Nobody has shown the same cristae-protection effect directly in human mitochondria.

The logic of cardiolipin binding stabilizing cristae structure is sound mechanistically, and it is the same logic that explains why Barth syndrome, a disease caused by defective cardiolipin remodeling, became the drug’s first approved use. But “sound mechanistically” and “demonstrated in the tissue that matters” are different claims, and the gap between them is exactly where a rat kidney study sits.

Side effects at these doses

80% of MMPOWER-2 participants on elamipretide reported an injection site reaction, versus a lower rate on placebo PMID 32096613

Across the subcutaneous trials, one adverse event shows up consistently:

• Injection site reactions were the most commonly reported adverse event, including pruritus, pain, bruising, and erythema, reported in 86% of elamipretide recipients in a separate trial testing the same drug and dose for macular degeneration, versus 71% on placebo^[9].
• No serious drug-related adverse events forced discontinuation in meaningful numbers across the completed trials cited above.
• Nothing beyond each trial’s own treatment window has been systematically tracked, so continuous dosing beyond a few months to a couple of years remains uncharacterized.

In a separate look at KLOW’s injection pain, that discomfort traced mostly to the compounding vehicle, not the peptides themselves. Elamipretide’s injection site reactions look like a different problem: the same pattern shows up across independent trials, in different diseases, run by different research groups, using the same drug and a similar dose each time. That consistency is itself a kind of evidence, just not the kind that tells you anything about long-term safety.

Where SS-31 sits next to other mitochondrial peptides

MOTS-c gets sold alongside SS-31 in the same longevity-and-metabolism conversation, but the two peptides do not share a mechanism and do not share an evidence base. MOTS-c acts mainly through the Folate-AICAR-AMPK signaling pathway, not cardiolipin binding^[10]. Published human MOTS-c research consists of observational and correlational studies of circulating peptide levels in conditions like diabetes and kidney disease, not completed randomized trials of administering the peptide to people^[10].

That is a thinner human trial record than SS-31’s. The mechanism itself may well be sound; it is simply far less tested in people. If you are comparing the two because they both get marketed as “mitochondrial peptides,” the comparison breaks down as soon as you ask for a randomized human dose-response trial, where SS-31’s four outnumber MOTS-c’s zero in this search.

The FDA approval changes less than it sounds like

An approval is a meaningful signal, but it answers a narrower question than most readers assume. Elamipretide’s accelerated approval covers Barth syndrome specifically, at the 40 mg subcutaneous daily dose used throughout its trial program, with a confirmatory trial still required as a condition of keeping that approval^[1].

Everything else is still exactly where it was before September 2025:

• Primary mitochondrial myopathy: unapproved. MMPOWER-3, the larger and more recent trial, missed its primary endpoint at the same dose that worked in Barth syndrome’s open-label data.
• Reperfusion injury after a heart attack: unapproved. EMBRACE STEMI tested a completely different route and found no benefit on infarct size.
• Geographic atrophy and dry macular degeneration: unapproved, with injection site reactions as the most visible finding at the same 40 mg dose.
• General healthy-aging or longevity use: untested. No trial here enrolled healthy adults for that purpose at any dose.

None of these conditions inherited anything from the Barth syndrome approval except the same number, 40 mg, applied to a population and an indication that number was never tested against.

Unlike the BPC-157 and TB-500 combination, whose stacking dosage has never been tested against a placebo in any trial, every dose in this article has been. That is the strange part of the SS-31 story: an unusually well-controlled trial record for a peptide, attached to a regulatory approval that covers one specific disease, sitting next to a sprawling off-label conversation that the approval does nothing to justify. The dose that got approved is not a recommendation. It is the answer to a question almost nobody asking about SS-31 online was asking.

This article is for research and informational purposes only and is not intended to diagnose, treat, cure, or prevent any disease. The peptides discussed here are sold for research use only and are not for human consumption. Nothing in this article constitutes medical advice. Consult a qualified clinician before making changes to a health, training, or supplementation protocol.