Content

SS-31 Peptide: A Patient's Guide to Elamipretide

Creator: Julian Reyes, Health Writer
Published: 2026-08-13

SS-31 and elamipretide are two names for the same four-amino-acid peptide, and its branded injectable form received accelerated approval from the FDA on September 19, 2025 for a single ultra-rare genetic disease.

Julian Reyes, Health Writer August 13, 2026 13 min read

A person walking a small dog along a quiet tree-lined sidewalk at sunrise

The short version

SS-31 and elamipretide name the same molecule — a Szeto-Schiller tetrapeptide, D-Arg-Dmt-Lys-Phe-NH₂, that binds cardiolipin in the inner mitochondrial membrane.
It reaches its target by charge, not by voltage: a dissociation constant of about 2.9 µM in cardiolipin-containing model bilayers and roughly 1,000- to 5,000-fold enrichment at the inner-membrane surface. Triphenylphosphonium-based mitochondrial antioxidants need an intact membrane potential; this one does not.
On September 19, 2025 the FDA granted elamipretide accelerated approval — as the branded product Forzinity, not as any compounded form — marketed by Stealth BioTherapeutics, to improve muscle strength in adult and pediatric patients with Barth syndrome weighing at least 30 kg. That approved product is a specific manufactured drug, different from the compounded SS-31 offered here, which is not FDA-approved.
Five elamipretide trials missed their primary endpoints: MMPOWER-3 (n=218, graded Class I evidence of no benefit), MMPOWER-2, TAZPOWER part 1, ReCLAIM-2 and PROGRESS-HF. The supportive results come largely from open-label extensions, post hoc subgroups and studies with fewer than 25 participants.
In rats, elamipretide did not prevent cardiolipin loss after ischemia-reperfusion, yet cristae networks and respiratory-complex activity still improved — evidence that it acts on the organization of cardiolipin-containing membranes, not their cardiolipin content.
Injection-site reactions were the most common adverse event in every trial above, mostly mild, with the manufactured drug at 40 mg daily under study conditions.

The SS-31 peptide is a synthetic chain four amino acids long — D-arginine, 2′,6′-dimethyltyrosine, lysine and phenylalanine — built by Hazel Szeto and Peter Schiller to concentrate at one lipid inside mitochondria: cardiolipin. In clinical development the same molecule is called elamipretide; older papers label it MTP-131 or Bendavia. Nearly everything about this peptide follows from that pairing.

Cardiolipin is a dimeric phospholipid carrying four acyl chains instead of the usual two, made only in the inner mitochondrial membrane, where it accounts for roughly 15 to 20 percent of phospholipid mass. Its cone-like geometry helps build the tight folds — cristae — that hold the energy-producing machinery, and its negative charge creates the surface SS-31 binds.

SS-31 dispensed through a telehealth service is prepared as a compounded medication, which is different from an FDA-approved product: the compounded formulation offered here is not FDA-approved.

What the SS-31 peptide is, chemically

SS-31 is a mitochondria-targeting tetrapeptide with the sequence D-Arg-Dmt-Lys-Phe-NH₂. Two chemistry choices do specific work: the D-configuration arginine resists breakdown by proteases, which ordinary L-peptides do not, and the 2′,6′-dimethyl substitution on the tyrosine ring deepens insertion into the fatty core of a membrane.

The design principle is an alternating aromatic-cationic motif: the charged residues (D-Arg, Lys) are drawn to the negatively charged cardiolipin headgroup while the aromatic ones (Dmt, Phe) slot into the lipid interior. The peptide sits in the membrane's interfacial region — it does not cross into the mitochondrial matrix.

Cardiolipin and why mitochondria depend on it

Cardiolipin does two structural jobs: it stabilizes the curvature of cristae junctions, and it clusters into microdomains that act as platforms for the respiratory chain — complexes I, III and IV, plus the F1F0-ATP synthase, all of which need cardiolipin for optimal activity.

Those microdomains also glue individual complexes into higher-order assemblies called supercomplexes, or respirasomes. Grouping them shortens the distance electrons travel, which channels electron flow more efficiently and limits leakage at complexes I and III. The dependence is dramatic: in yeast engineered to lack cardiolipin synthase, about 90 percent of complexes III and IV existed as individual homodimers rather than in supercomplexes (Zhang, Mileykovskaya and Dowhan, 2002, in Saccharomyces cerevisiae). Peroxidation, enzymatic degradation and genetic defects in cardiolipin remodeling all degrade the same architecture.

How SS-31 binds the membrane

Binding is driven by cardiolipin density, not by membrane potential. In model bilayers the dissociation constant is approximately 2.9 µM, and affinity scales directly with surface charge density; the result is a roughly 1,000- to 5,000-fold enrichment of SS-31 at the inner-membrane surface relative to the surrounding solution.

That distinction matters. Triphenylphosphonium-conjugated antioxidants are lipophilic cations pulled into the mitochondrial matrix by the membrane potential, so they concentrate poorly in mitochondria whose potential has collapsed; SS-31 does not need the potential to find its target.

Once bound, the peptide neutralizes a fraction of the negative surface charge cardiolipin contributes, changing how ions and loosely docked peripheral proteins distribute at the membrane interface (Mitchell et al., 2020, model bilayers and isolated mitochondria). That work also found binding did not destabilize the bilayer even at maximal binding.

NMR and molecular dynamics show SS-31 adopting an extended conformation when membrane-bound, while related analogs such as SS-20 and SPN10 fold into compact reverse turns. The extended shape lets one peptide engage several cardiolipin headgroups at once — the structural basis for the clustering effect. In isolated rat cardiac mitochondria imaged by serial block-face scanning electron microscopy, treatment was associated with tighter cristae curvature.

The cytochrome c switch

Cytochrome c normally sits tethered to the outer face of the inner membrane by an electrostatic interaction with cardiolipin, shuttling electrons between complexes III and IV. When cardiolipin acyl chains are peroxidized, that tether is disrupted and cytochrome c changes shape, converting from an electron carrier into a peroxidase — a gain of function that generates lipid hydroperoxides, which oxidize still more cardiolipin.

SS-31 interrupts that feedforward cycle by stabilizing the native cardiolipin-cytochrome c interaction, so cytochrome c keeps carrying electrons and does not flip (Szeto, 2014, isolated mitochondria and in vitro systems). A specific percentage reduction in mitochondrial reactive oxygen species is often quoted alongside this mechanism online; that figure does not appear in the paper it is credited to, so this guide reports the direction and leaves the number out.

Organization, not quantity

The sharpest test of what the peptide changes came from a rat cardiac ischemia-reperfusion study. Mass spectrometry showed that elamipretide did not protect against the drop in cardiolipin concentration — and yet the structure and function measures still improved (Allen et al., 2020). Three-dimensional electron microscopy showed less fragmentation of cristae networks and greater cristae connectivity versus untreated controls, and respirometry in permeabilized ventricular fibres showed that ischemia-reperfusion-induced decrements in complex I, II and IV activity were alleviated.

The evidence therefore points to SS-31 acting on how cardiolipin-containing membranes are organized and protected, not on how much cardiolipin a cell holds — a narrower claim than "restores cardiolipin," which is how the mechanism is often summarized.

What isolated mitochondria, animals and one human tissue study show

Cross-linking mass spectrometry mapped the peptide's interactions in isolated mitochondria: two clusters, one of oxidative-phosphorylation subunits (complexes I, III, IV and the ATP synthase) and one of 2-oxoglutarate metabolism enzymes. Every protein identified was already a known cardiolipin binder, and respirometry in the same preparations showed improved oxygen consumption and reduced hydrogen peroxide (Chavez et al., 2020).

In aged mouse muscle mitochondria, elamipretide improved sensitivity to ADP by increasing its uptake through the adenine nucleotide translocator, a cardiolipin-dependent carrier it binds directly along with the ATP synthase; treatment also rescued muscle force and heart systolic function — in mice, in vivo (Pharaoh et al., 2023). In a mouse model of Barth syndrome created by knocking down TAFAZZIN, SS-31 restored cardiac mitochondrial morphology and corrected defective mitophagy (Russo et al., 2024).

The one human datum in this mechanistic literature deserves a precise label: freshly explanted failing human heart tissue, from children and adults, treated with elamipretide in the laboratory. In that ex vivo tissue, complex I and complex IV activities and supercomplex-associated complex IV activity improved (Chatfield et al., 2019). Those were donated tissue samples, not treated patients, and explanted tissue does not predict what happens in a living person.

What the elamipretide trials found

Elamipretide has been tested in randomized human trials across four conditions, and the pattern is mixed. Five trials — in primary mitochondrial myopathy, Barth syndrome, dry age-related macular degeneration and heart failure — missed their primary endpoints. The supportive findings come mostly from uncontrolled extensions, secondary endpoints and post hoc subgroups.

Trial	Population	N	Primary endpoint result
MMPOWER-2 (2020, crossover)	Adults with genetically confirmed primary mitochondrial myopathy	30	Not met. Six-minute walk difference 19.8 m (95% CI −2.8 to 42.5; P=0.0833)
MMPOWER-3 (2023, phase 3, 24 weeks)	Adults with primary mitochondrial myopathy, mean age 45.6	218	Both not met. Six-minute walk −3.2 m (95% CI −18.7 to 12.3; p=0.69); fatigue score p=0.37. Class I evidence of no benefit on either
TAZPOWER part 1 (2021, crossover, 12 weeks per arm)	Barth syndrome	12	Neither primary endpoint met
ReCLAIM-2 (2025, phase 2, 48 weeks)	Adults ≥55 with dry AMD and geographic atrophy	176	Both not met (low-luminance visual acuity; geographic atrophy area)
PROGRESS-HF (2020, phase 2, 28 days)	Heart failure with reduced ejection fraction, mean age 65	71	Not met. Left ventricular end-systolic volume: 40 mg vs placebo 2.3 mL (95% CI −1.9 to 6.5; P=0.28)

The results pointing the other way need their designs stated. The Barth evidence base is as small and homogeneous as rare-disease medicine gets — TAZPOWER's 12 participants were all white males with a mean age of 19.5 — and an international registry of 376 primary-mitochondrial-myopathy patients (RePOWER) records that before Forzinity no approved treatment existed. In the TAZPOWER open-label extension — no placebo group, no blinding, 10 patients entering and 8 reaching week 168 — six-minute walk distance improved by a cumulative 96.1 m (P=.003), and the monolysocardiolipin-to-cardiolipin ratio improved in step with clinical measures. A retrospective comparison of 8 treated Barth patients against 19 untreated natural-history controls reported a 79.7 m walking difference at week 64 (P=0.0004), growing to 91.0 m by week 76 (P=0.0005), alongside handheld-dynamometry strength differences of 40.8 newtons at week 64 (P=0.0002) and 56.7 newtons at week 76 (P=0.0005) — the muscle-strength evidence behind the eventual label language. The two ReCLAIM phase 1 studies (n=21 and n=19) were open-label safety studies whose visual-function gains were exploratory: best-corrected visual acuity improved 3.6 letters (P=0.014) and low-luminance acuity 5.6 letters (P=0.004) in the intermediate-AMD study, while drusen volume, fundus autofluorescence, microperimetry and dark adaptation did not change significantly — function moved, anatomy did not. A post hoc MMPOWER-3 analysis found participants with chronic progressive external ophthalmoplegia walked 37.3 ± 9.5 m further at week 24 versus −8.0 ± 10.7 m on placebo (p=0.0024) — a hypothesis for the planned NuPOWER trial, not a result to rely on.

What the September 2025 approval covers

On September 19, 2025 the FDA granted accelerated approval to elamipretide, marketed as Forzinity by Stealth BioTherapeutics, to improve muscle strength in adult and pediatric patients with Barth syndrome weighing at least 30 kg. It is the first disease-specific therapy approved for that disorder, and a confirmatory trial is required as a condition of accelerated approval. Elamipretide remains in phase III development for dry age-related macular degeneration and for mitochondrial myopathies.

That is a narrow indication for an ultra-rare genetic disease, reached through a specific manufactured product. An approved elamipretide product exists for Barth syndrome, which is different from what a compounding pharmacy prepares: the compounded SS-31 offered here is not FDA-approved, for that indication or any other. A licensed provider may still prescribe — that decision is between you and your doctor.

Safety signals reported in the trials

The consistent finding across every elamipretide trial is injection-site reactions, mostly mild. In MMPOWER-2 they were the most common adverse event, affecting 80 percent of 30 adults, with no serious adverse events and no deaths. In ReCLAIM-2, adverse events occurred in 86 percent on elamipretide versus 71 percent on placebo, most commonly injection-site itching, pain, bruising and redness.

The two ReCLAIM phase 1 studies showed the same pattern: in the 21-person drusen study every participant had at least one adverse event, all mild (57 percent) or moderate (43 percent), mostly at the injection site, with no serious systemic events; in the 19-person geographic atrophy study one participant exited for an intolerable injection-site reaction. The 168-week Barth extension also named injection-site reactions the most common adverse event.

Two caveats belong with those numbers. They describe the manufactured drug at 40 mg subcutaneously per day in monitored trials, not a compounded product used outside one. And a tolerability record in 12 to 218 people with a specific diagnosis says little about a different population — exactly the gap a prescribing conversation exists to close.

Where SS-31 sits among the mitochondrial compounds

SS-31 is one of several compounds studied for mitochondrial function, and their evidence bases do not transfer to one another. MOTS-c is a peptide encoded by mitochondrial DNA itself, with a rodent-heavy literature and one ongoing human trial. NAD+ is a coenzyme, not a peptide, whose human data comes mostly from oral precursors. SS-31 has the most human trial data of the three, including the negative results above.

Promise groups all three under Energy & Focus. That grouping is a navigation convenience, not a claim that they are interchangeable: each has a distinct mechanism, study population and regulatory status.

References

Peer-reviewed sources cited in this guide, with PubMed listings.

Zhang M, Mileykovskaya E, Dowhan W. (2002). Gluing the respiratory chain together: cardiolipin is required for supercomplex formation in the inner mitochondrial membrane. Journal of Biological Chemistry, 277(46), 43553–43556. PubMed
Szeto HH. (2014). First-in-class cardiolipin-protective compound as a therapeutic agent to restore mitochondrial bioenergetics. British Journal of Pharmacology, 171(8), 2029–2050. PubMed
Chatfield KC, et al. (2019). Elamipretide improves mitochondrial function in the failing human heart. JACC: Basic to Translational Science, 4(2), 147–157. PubMed
Butler J, et al. (2020). Effects of elamipretide on left ventricular function in patients with heart failure with reduced ejection fraction: the PROGRESS-HF phase 2 trial. Journal of Cardiac Failure, 26(5), 429–437. PubMed
Karaa A, et al. (2020). A randomized crossover trial of elamipretide in adults with primary mitochondrial myopathy. Journal of Cachexia, Sarcopenia and Muscle, 11(4), 909–918. PubMed
Mitchell W, et al. (2020). The mitochondria-targeted peptide SS-31 binds lipid bilayers and modulates surface electrostatics as a key component of its mechanism of action. Journal of Biological Chemistry, 295(21), 7452–7469. PubMed
Chavez JD, et al. (2020). Mitochondrial protein interaction landscape of SS-31. Proceedings of the National Academy of Sciences, 117(26), 15363–15373. PubMed
Allen ME, et al. (2020). The cardiolipin-binding peptide elamipretide mitigates fragmentation of cristae networks following cardiac ischemia reperfusion in rats. Communications Biology, 3(1), 389. PubMed
Reid Thompson W, et al. (2021). A phase 2/3 randomized clinical trial followed by an open-label extension to evaluate the effectiveness of elamipretide in Barth syndrome. Genetics in Medicine, 23(3), 471–478. PubMed
Allingham MJ, Mettu PS, Cousins SW. (2022). Phase 1 clinical trial of elamipretide in intermediate age-related macular degeneration and high-risk drusen. Ophthalmology Science, 2(1), 100095. PubMed
Mettu PS, Allingham MJ, Cousins SW. (2022). Phase 1 clinical trial of elamipretide in dry age-related macular degeneration and noncentral geographic atrophy. Ophthalmology Science, 2(1), 100086. PubMed
Mitchell W, et al. (2022). Structure-activity relationships of mitochondria-targeted tetrapeptide pharmacological compounds. eLife, 11, e75531. PubMed
Hornby B, et al. (2022). Natural history comparison study to assess the efficacy of elamipretide in patients with Barth syndrome. Orphanet Journal of Rare Diseases, 17(1), 336. PubMed
Karaa A, et al. (2023). Efficacy and safety of elamipretide in individuals with primary mitochondrial myopathy: the MMPOWER-3 randomized clinical trial. Neurology, 101(3), e238–e252. PubMed
Pharaoh G, et al. (2023). The mitochondrially targeted peptide elamipretide (SS-31) improves ADP sensitivity in aged mitochondria by increasing uptake through the adenine nucleotide translocator. GeroScience, 45(6), 3529–3548. PubMed
Thompson WR, et al. (2024). Long-term efficacy and safety of elamipretide in patients with Barth syndrome: 168-week open-label extension results of TAZPOWER. Genetics in Medicine, 26(7), 101138. PubMed
Russo S, et al. (2024). SS-31 treatment ameliorates cardiac mitochondrial morphology and defective mitophagy in a murine model of Barth syndrome. Scientific Reports, 14, 13655. PubMed
Karaa A, et al. (2024). Genotype-specific effects of elamipretide in patients with primary mitochondrial myopathy: a post hoc analysis of the MMPOWER-3 trial. Orphanet Journal of Rare Diseases, 19(1), 431. PubMed
Ehlers JP, et al. (2025). ReCLAIM-2: a randomized phase II clinical trial evaluating elamipretide in age-related macular degeneration, geographic atrophy growth, visual function, and ellipsoid zone preservation. Ophthalmology Science, 5(1), 100628. PubMed
Sabbah HN, et al. (2025). Contemporary insights into elamipretide's mitochondrial mechanism of action and therapeutic effects. Biomedicine & Pharmacotherapy, 187, 118056. PubMed
Shirley M. (2026). Elamipretide: first approval. Drugs, 86(3), 377–383. PubMed
Zhao C, Zhuang X, Gao J. (2026). Elamipretide: the first cardiolipin-directed mitochondrial therapeutic for Barth syndrome approved under accelerated approval. Drug Discovery & Therapeutics, 19(6), 435–436. PubMed

Where mitochondrial medicine goes from here

The field's understanding of this peptide has moved. Early descriptions framed SS-31 as a mitochondria-targeted antioxidant that mopped up reactive oxygen species. The current model, laid out in a 2025 review by Sabbah and colleagues, centers instead on cardiolipin-dependent modulation of membrane electrostatics and the assembly of cardiolipin-dependent protein complexes, with lower reactive-oxygen-species output as a downstream consequence of better-organized respiratory machinery. One relative of the peptide, SPN10, outperformed SS-31 in cell-culture stress assays, so the series has not finished producing candidates.

What comes next depends on trials that have not reported: the confirmatory study required by the accelerated approval, the NuPOWER trial in mitochondrial DNA maintenance disorders, and the phase III ophthalmology program built on ellipsoid-zone preservation as an endpoint. That endpoint choice has data behind it: within ReCLAIM-2's missed-primary result, prespecified secondary measures showed a 43% reduction in progression of total ellipsoid-zone attenuation (nominal P=0.0034), a 47% reduction in partial EZ degradation (nominal P=0.0040), and ≥10-letter low-luminance acuity gains in 14.6% of treated patients versus 2.1% on placebo (nominal P=0.0404) — nominal findings, which is exactly why a phase III exists instead of a label.

SS-31 is available only by prescription. A licensed provider reviews every request and decides whether it is medically appropriate — not everyone qualifies, and a provider may decline. When it is prescribed through Promise, SS-31 is prepared by a licensed U.S. compounding pharmacy.

This article is for educational purposes only and is not medical advice. Talk with a licensed healthcare provider about your individual health questions and before starting or stopping any treatment.

Common questions

What is the difference between SS-31 and elamipretide?

There is none — they are two names for one molecule. SS-31 is the laboratory designation from the Szeto-Schiller peptide series; elamipretide is the nonproprietary name used in clinical development and on the approved product. Older papers call the same compound MTP-131 or Bendavia.

Is SS-31 FDA-approved?

No. An FDA-approved elamipretide product (Forzinity) exists for Barth syndrome in patients weighing at least 30 kg, but the compounded SS-31 offered here is not FDA-approved. Compounded medications are prepared by a pharmacy for an individual patient under a prescription rather than reviewed and manufactured as an approved drug product.

What does SS-31 do to cardiolipin?

It binds cardiolipin at the membrane surface and changes how it is organized. Measured effects in model membranes and isolated mitochondria include partial neutralization of cardiolipin's negative surface charge, tighter clustering into microdomains, reduced lateral lipid movement, and stabilization of the cardiolipin-cytochrome c interaction.

Does SS-31 increase how much cardiolipin a cell has?

No. In rats subjected to cardiac ischemia-reperfusion, elamipretide did not prevent the fall in cardiolipin concentration, even though cristae structure and respiratory-complex activity improved in the same animals. The evidence supports an effect on cardiolipin organization and protection, not on cardiolipin quantity.

How does SS-31 differ from mitochondria-targeted antioxidants?

Triphenylphosphonium-conjugated antioxidants are lipophilic cations that accumulate in the mitochondrial matrix, pulled there by the membrane potential. SS-31 instead binds cardiolipin at the inner-membrane surface, driven by lipid charge density, and its described mechanism centers on membrane electrostatics and protein-complex organization rather than on scavenging molecules one for one.

Has SS-31 been tested in people?

Yes — as elamipretide, in randomized trials across primary mitochondrial myopathy, Barth syndrome, dry age-related macular degeneration and heart failure. Five of those trials missed their primary endpoints, including a 218-person phase 3 trial graded as Class I evidence of no benefit on walking distance or fatigue at 24 weeks.

What side effects were reported in the elamipretide trials?

Injection-site reactions were the dominant adverse event in every trial: 80 percent of participants in a 30-person crossover trial, and the most common event in the 176-person ReCLAIM-2 trial, where adverse events overall occurred in 86 percent on elamipretide versus 71 percent on placebo. Most were mild. These describe a manufactured drug under study conditions.

What happens if a provider declines to prescribe SS-31?

A decline is a clinical judgment that this compound is not appropriate for you right now, based on your history, your medications, or a mismatch between the evidence and your goals. Ask what the provider suggests instead, and what would need to change for the answer to change.