SS-31 vs. MOTS-c.
Same organelle. Different evidence universe.

Elamipretide has crossed a threshold MOTS-c has not. Researchers have administered it by intravenous and subcutaneous routes in primary mitochondrial myopathy, heart failure and Barth syndrome. The largest randomized trial enrolled 218 people. In September 2025, FDA granted accelerated approval under the name Forzinity for improving muscle strength in adults and children with Barth syndrome who weigh at least 30 kg.^[1][2][6]

That is meaningful validation of a targeted mitochondrial strategy—but not a universal mitochondrial-performance approval. The pivotal Barth evidence is unusually narrow, the randomized primary outcomes were negative, and continued approval depends on confirmatory evidence. Trials in broader mitochondrial myopathy and heart failure have also produced important null results.^[2][6][8]

MOTS-c sits earlier on the staircase. In cells it moves to the nucleus during metabolic stress and interacts with stress-response pathways. In mice, administered peptide has improved metabolic and physical outcomes. Human studies show that endogenous MOTS-c varies with age, tissue and exercise. None of those observations demonstrate that injecting native MOTS-c improves energy, glucose control, body composition or performance in people.^{[11][12][13][14][15]}

SS-31 is a four-amino-acid peptide also known as elamipretide. It concentrates at the inner mitochondrial membrane and binds cardiolipin, a phospholipid that helps organize respiratory-chain proteins and membrane architecture. FDA’s label describes improved inner-membrane structure and mitochondrial function after cardiolipin binding. That makes elamipretide conceptually closest to a membrane-restoration drug.^[2]

MOTS-c is a 16-amino-acid peptide encoded within mitochondrial 12S ribosomal RNA. The foundational experiments linked it to metabolic homeostasis in cells and mice. Later work found that metabolic stress can drive MOTS-c into the nucleus in an AMPK-dependent process, where it interacts with NRF2-related antioxidant-response elements. That makes MOTS-c look less like a membrane patch and more like a retrograde signal: a message from the mitochondrion that changes nuclear gene expression.^[11][12]

Because the proposed targets differ, “both work on mitochondria” cannot prove that one substitutes for the other—or that combining them is additive. Mitochondria are not one switch. Membranes, electron transport, redox signaling, substrate use, quality control and nuclear communication can move together or in opposite directions.

The MMPOWER program is a useful lesson in why development moves from small trials to large ones. In the first randomized dose-escalation study, 36 adults with primary mitochondrial myopathy received five days of intravenous elamipretide or placebo. At the highest exposure, the unadjusted six-minute-walk change was 64.5 metres versus 20.4 metres with placebo, narrowly missing conventional significance (P=.053). A covariate-adjusted analysis was positive, but it was not the cleanest primary comparison.^[4]

MMPOWER-2 then tested 40 mg subcutaneously each day for four weeks in a 30-person crossover trial. Mean six-minute-walk distance was 398.3 metres with elamipretide and 378.5 metres with placebo: a 19.8-metre difference (95% CI −2.8 to 42.5; P=.0833). Patient-reported fatigue favored elamipretide, but the functional primary outcome remained uncertain.^[5]

MMPOWER-3 provided the larger answer. The 24-week phase 3 trial randomized 218 participants to 40 mg/day subcutaneous elamipretide or placebo. The between-group difference in six-minute-walk change was −3.2 metres (95% CI −18.7 to 12.3; P=.69), and the fatigue co-primary endpoint was also negative. Exploratory post-hoc genotype findings may justify an enriched follow-up trial; they do not reverse the prespecified overall result.^[6]

Heart-failure studies add the same discipline. A small single-infusion experiment showed dose-related changes worth pursuing. But PROGRESS-HF found that four weeks of daily elamipretide did not significantly improve left-ventricular end-systolic volume. A mitochondrial mechanism can be real while the chosen clinical endpoint, population or exposure still fails.^[7][8]

Barth syndrome is caused by variants in TAZ that disrupt cardiolipin remodeling. The mechanistic match with a cardiolipin-binding drug is unusually direct. TAZPOWER enrolled 12 males aged 12 to 35 in two 12-week randomized crossover periods with a four-week washout, using 40 mg subcutaneous elamipretide daily, then continued into an open-label extension.^[2][9][10]

During randomized treatment, the prespecified six-minute-walk and fatigue endpoints were not superior to placebo. FDA’s accelerated approval instead relied on knee-extensor strength as an intermediate endpoint during the open-label extension. Median strength was 124 newtons at baseline; descriptive median changes were +34 N at week 12, +68 N at week 24 and +35 N at week 72, with +63 N among eight remaining participants at week 168. Without a concurrent control, those changes cannot separate treatment from training, maturation, attrition or natural variation as cleanly as a randomized comparison.^{[2][3][9][10]}

Injection-site reactions were also common: 100% of elamipretide-treated participants experienced them versus 66.7% with placebo during the randomized study. Serious hypersensitivity has been reported. The approved 40-mg once-daily exposure applies to the labeled Barth population weighing at least 30 kg; it is not a general wellness protocol.^[2][3]

This is neither a hollow approval nor a settled endpoint. It is a conditional regulatory judgment for a devastating ultra-rare disease with no approved alternatives, built on an intermediate strength measure considered reasonably likely to predict clinical benefit. Continued approval requires verification by March 31, 2030.^[1][2]

The original 2015 experiments showed that administered MOTS-c improved insulin sensitivity and protected mice against high-fat-diet obesity while altering cellular folate and purine metabolism. Later studies reported improved physical capacity in old mice. These are unusually interesting translational findings because they connect a mitochondrially encoded peptide to whole-body metabolism and performance.^[11][13]

The human portion of the 2021 exercise paper is often described too aggressively. Ten sedentary healthy young men performed cycling exercise, and investigators measured MOTS-c in skeletal muscle and plasma. The body’s own peptide increased. Researchers did not inject MOTS-c into those men, randomize them to peptide or placebo, or test whether administration improved VO₂max, strength or recovery.^[13]

Cross-sectional work across young, middle-aged and older men is similarly observational. Circulating concentrations tended to fall with age while skeletal-muscle expression followed a different pattern. That is evidence that MOTS-c biology changes with age, not that low blood MOTS-c is a deficiency state or that replacement restores youthful function.^[14]

FDA’s July 2026 compounding review found no human exposure, pharmacokinetic or safety data for MOTS-c and raised product-quality concerns including impurities, aggregation, endotoxin and immunogenicity for injectable compounded products. A registered phase 2a study, NCT07505745, now aims to randomize adults with prediabetes and overweight or obesity to native MOTS-c or placebo for 12 weeks. It was recruiting with no posted results when this article was checked.^[15][16]

CB4211 sometimes appears in the same conversation. It is a MOTS-c analog, not native MOTS-c. A phase 1a/1b study enrolled healthy volunteers and people with fatty liver disease, but a registry entry without a peer-reviewed efficacy report cannot stand in for a proven clinical outcome—and analog data should not be silently reassigned to the parent peptide.^[17]

MOTS-c has the ingredients of a compelling “exercise mimetic” story: metabolic stress, AMPK signaling, nuclear translocation, improved mouse performance and an endogenous rise after human exercise. The honest bullish case is that biology this coherent deserves fast, well-designed human testing.

The missing step is pharmacology. A peptide released or expressed in a particular tissue, concentration and temporal pattern during exercise may not behave like an injected preparation. Exercise also changes mechanical loading, blood flow, sympathetic signaling, calcium handling, temperature, substrate flux and hundreds of secreted factors. Reproducing one signal is not reproducing training.

Elamipretide demonstrates the same principle from the other direction. It can improve a mitochondrial membrane target and still fail to improve walking in a broad phase 3 population. Biomarker or mechanism success earns the right to ask a functional question; it does not answer it.

A membrane-restoration drug plus a stress-response peptide creates an intuitively attractive stack: stabilize cardiolipin, then amplify adaptive signaling. Mechanistically, the targets are different enough that additive effects are possible. They are also different enough that antagonism, redundancy or a new adverse-effect profile is possible.

No controlled human study has tested elamipretide plus MOTS-c. No reliable interaction data establish that one improves the other’s exposure, target engagement or functional effect. And because native MOTS-c itself lacks published administered-human outcome data, a combination trial would be trying to validate two unknowns at once.

A 2×2 factorial trial could randomize participants to placebo, elamipretide, MOTS-c or both. That design would estimate each main effect and test whether the combination adds more than either single agent. Until then, “different mechanisms” is a reason to run the experiment—not a result from it.

Elamipretide is the more clinically mature story. Its cardiolipin mechanism has survived into multiple randomized human programs and a real, narrow accelerated approval. The same program also produced negative phase 3 and heart-failure results, proving that mitochondrial targeting is not a universal performance lever.

MOTS-c may be the more conceptually provocative story. A mitochondrially encoded peptide that enters the nucleus during stress and changes metabolic gene programs is exactly the kind of biology that could open a new therapeutic class. But human endogenous observations do not yet tell us what an injected product does, how much reaches the relevant tissues, or whether a person performs or feels better.

The bullish, evidence-based position is not that these peptides have been disproven. It is that their mechanisms are interesting enough to demand better trials. For SS-31, the next questions are population selection, confirmatory benefit and whether disease-specific restoration generalizes at all. For MOTS-c, the next milestone is more basic: controlled human exposure, pharmacokinetics, safety and a prespecified functional outcome. That is where the excitement becomes evidence.