Compound Physician required WADA prohibited

SS-31 (Elamipretide)

Mitochondria-targeted peptide with a narrow FDA-approved indication (Barth syndrome) — no longevity/aging human RCT yet.

Approved Verified 27 August 2026

The story

SS-31 did not begin as a mitochondrial drug. It began as a by-product of opioid peptide chemistry. Through the 1990s Peter Schiller, at the Institut de recherches cliniques de Montréal, was building small analgesic peptide analogues built around an unusual synthetic aromatic residue, 2',6'-dimethyltyrosine (Dmt). Hazel Szeto, a pharmacologist at Weill Cornell Medicine in New York, was studying how such peptides crossed biological membranes. What the two found, working together, was that a short peptide with alternating aromatic and basic residues did something the design had not intended: it crossed the plasma membrane without a transporter and then accumulated inside mitochondria, concentrating roughly a thousand-fold in the inner mitochondrial membrane. The series was named for its two inventors — the Szeto-Schiller, or SS, peptides — and its members were numbered. The thirty-first, H-D-Arg-Dmt-Lys-Phe-NH2, became SS-31 (Zhao et al., J Biol Chem 2004; Szeto, AAPS J 2006 — "Cell-permeable, mitochondrial-targeted, peptide antioxidants").

The 2004 paper framed the peptides as targeted antioxidants: they scavenged hydrogen peroxide and peroxynitrite, blocked mitochondrial swelling and permeability transition, and prevented cytochrome c release in isolated mitochondria and in neuronal cells (DOI 10.1074/jbc.M402999200). Early animal work followed that framing, in models from ALS to myocardial infarction (PMID 16895581; PMID 17429296).

The mechanistic account that survives is different and more specific, and it is what makes this compound worth understanding properly. In 2013, Birk and colleagues used a fluorescent analogue of SS-31 to show that the peptide binds with high affinity to cardiolipin — a four-tailed anionic phospholipid found essentially nowhere in the cell except the inner mitochondrial membrane, where it is required to fold that membrane into cristae and to organize the respiratory-chain complexes into supercomplexes. By occupying cardiolipin, SS-31 blocks the cardiolipin-cytochrome c interaction that turns cytochrome c from an electron carrier into a peroxidase, protecting cristae architecture and letting ATP synthesis recover faster after ischemia (DOI 10.1681/ASN.2012121216). Szeto set this out as a general therapeutic thesis the following year: cardiolipin peroxidation and depletion are common to many energy-deficient states, so a cardiolipin-protective compound is a platform rather than a single-disease drug (DOI 10.1111/bph.12461). This is the origin of the "it improves the mitochondria you already have" framing that later travelled into consumer marketing — and, importantly, it is a claim about structure and coupling efficiency, not about making new mitochondria.

Szeto founded Stealth Peptides, later Stealth BioTherapeutics, in 2006 to commercialize the compound, at a point when faculty company formation was uncommon at her institution (Weill Cornell Medicine, 2025-12). The clinical arc that followed was long, broad and mostly unsuccessful. Under the code MTP-131 and the trade name Bendavia, the compound was taken into reperfusion injury after STEMI (NCT01572909, n=300), renal artery stenosis, heart failure (NCT02914665, n=308; NCT02788747, n=71), skeletal-muscle function in older adults (NCT02245620, n=41), Leber's hereditary optic neuropathy, Fuchs' corneal endothelial dystrophy, dry age-related macular degeneration, primary mitochondrial myopathy, and Barth syndrome. Two of those programs defined the compound's public reputation in opposite directions. MMPOWER-3, a 218-participant phase 3 trial in primary mitochondrial myopathy, missed both of its primary endpoints in 2020 and the open-label extension was terminated with the registry entry reading "Registration trial did not meet the primary end points" (NCT03323749; DOI 10.1212/WNL.0000000000207402). TAZPOWER, a 12-patient trial in Barth syndrome, also missed its randomized primary endpoints — but its 168-week open-label extension produced the muscle-strength data on which FDA granted accelerated approval of FORZINITY (elamipretide HCl) on 2025-09-19, the first drug ever approved for Barth syndrome (FDA Drug Trials Snapshot; DOI 10.5582/ddt.2025.01111). That approval followed an unusually visible regulatory fight: an FDA advisory committee voted 10-6 in favour on 2024-10-10, FDA nonetheless issued a Complete Response Letter on 2025-05-29, and Stealth resubmitted on 2025-08-18 (Barth Syndrome Foundation regulatory timeline).

The route into longevity and biohacking culture ran in parallel and ahead of that approval, under the older laboratory name. Research-chemical vendors and wellness clinics have marketed vials labelled "SS-31" as a mitochondrial anti-aging compound — one clinic page describes it as offering "mitochondrial repair for energy and longevity" (Revolution Health & Wellness) — and the September 2025 approval was immediately absorbed into that channel as validation, with one widely-circulated substack titled "The biohacker mitochondrial enhancing peptide, SS-31, just received FDA approval" (Glorioso, Substack). The gap between those two sentences — a 12-patient ultra-rare-disease approval, and a general anti-aging claim — is the subject of.

  1. 2004 Szeto and Schiller report cell-permeable aromatic-cationic peptides that concentrate ~1000-fold in the inner mitochondrial membrane; SS-31 named
  2. 2006 Szeto founds Stealth Peptides (later Stealth BioTherapeutics) at Weill Cornell
  3. 2012–2015 EMBRACE-STEMI: MTP-131 ("Bendavia") tested for reperfusion injury after primary angioplasty, n=300
  4. 2013 Birk et al. identify cardiolipin as the binding target and explain cristae protection
  5. 2014 Szeto publishes the "first-in-class cardiolipin-protective compound" thesis
  6. 2018 MMPOWER (phase 1/2, n=36) reports a dose-dependent 6-minute-walk increase after 5 days of IV infusion
  7. 2020 MMPOWER-2 crossover (n=30) misses its 6MWT primary endpoint (p=0.083) but improves patient-reported fatigue
  8. 2020 MMPOWER-3 fails. Phase 3, n=218, misses both primary endpoints; open-label extension terminated
  9. 2021 First randomized human study in healthy older adults: a single 2-hour IV infusion raises in-vivo muscle ATPmax, with no effect at day 7 and no effect on fatigue resistance
  10. 2021 TAZPOWER (Barth syndrome, n=12) reported: randomized phase misses primary endpoints
  11. 2024-10-10 FDA Cardiovascular and Renal Drugs Advisory Committee votes 10–6 that the evidence supports effectiveness in Barth syndrome
  12. 2024 TAZPOWER 168-week open-label extension published: cumulative +96.1 m on 6MWT from OLE baseline, improved cardiac volumes and MLCL/CL ratio — all uncontrolled
  13. 2025-05-29 FDA issues a Complete Response Letter; Stealth resubmits 2025-08-18
  14. 2025-09-19 FDA accelerated approval of FORZINITY (elamipretide) to improve muscle strength in adults and children with Barth syndrome weighing ≥30 kg
  15. 2025 ReCLAIM-2 (dry AMD with geographic atrophy, n=176) misses both primary endpoints; secondary ellipsoid-zone findings are nominal only
  16. 2026-07 A 72-week randomized placebo-controlled trial in Barth syndrome, 4TAZPower, is registered with a start date of 2026-07-02 and status RECRUITING. The sponsor's own registry description calls it post-marketing and confirmatory; no FDA document naming it as the postmarketing requirement was located, so that identification is this dossier's inference. No results are posted

What it does

A mitochondria-targeted peptide with a narrow FDA-approved indication (Barth syndrome) and no longevity/aging RCT. Tracked, not recommended for general use.

  • mitochondrial function

What the evidence shows

The shape of the approval is the finding. FDA granted FORZINITY accelerated approval on 2025-09-19 on an intermediate endpoint — knee-extensor strength — drawn largely from the uncontrolled 168-week open-label extension of a 12-patient crossover trial whose randomized phase FDA itself states was not superior to placebo; the larger trials failed, with MMPOWER-3, the largest in the programme at 218 participants, missing both of its primary endpoints, the 176-participant ReCLAIM-2 missing its own in dry AMD, and the heart-failure trials before them returning negative results. So the strongest evidence here establishes one muscle-strength measure, in an ultra-rare genetic cardiolipin disorder affecting roughly 150 Americans, on a surrogate that continued approval is contingent on confirming — the yield of more than twenty registered trials since 2012, most of them randomized and placebo-controlled. The single biggest limitation is that nothing in that evidence base speaks to a healthy adult: the only randomized healthy-adult exposure is a single intravenous dose in 41 older adults pre-screened for poor mitochondrial function, which raised a biochemical marker and did not improve function, and no chronic-dosing trial in healthy people has ever been run. NuPOWER, the trial that would have prospectively tested the post-hoc subgroup analysis invoked to rescue MMPOWER-3, completed in 2024-12 and has reported nothing.

Overall evidence strength: Low certainty

  • Karaa A et al. (2023) — MMPOWER-3, primary mitochondrial myopathy (2023) Randomized controlled trial

    The largest and most rigorous test elamipretide has ever faced found that people on the drug did not walk further or feel less fatigued than people on placebo. This is the single most important efficacy result in the compound's record, and it is a negative one; it does not tell us the drug does nothing in every setting, but it does mean the flagship claim failed under the conditions best designed to detect it.

    Source PMID 37268435

  • Reid Thompson W et al. (2021) + Hornby B et al. (2024) — TAZPOWER, Barth syndrome (2021) Randomized controlled trial

    In the properly controlled part of the study, elamipretide was not better than placebo on either primary endpoint. In the uncontrolled part that followed, the same small group kept improving for more than three years — walking a cumulative 96.1 m further and gaining a median 63 newtons of knee-extensor strength — and that open-label improvement, in eight boys and young men, is the evidence base on which the drug was approved. What it establishes is that something changed over three years in eight patients with an untreated, progressive disease; what it cannot establish, without a control group, is how much of that was the drug.

    Source PMID 38602181

  • Karaa A et al. (2018) — MMPOWER, phase 1/2 dose escalation (2018) Randomized controlled trial

    After five days of intravenous infusion, the highest-dose group walked 64.5 m further versus 20.4 m on placebo — a difference that did not reach the conventional significance threshold (p=0.053) but showed a dose-dependent trend (p=0.014). This is the encouraging early signal that the much larger MMPOWER-3 later failed to reproduce, which is the ordinary and instructive fate of small phase 2 signals.

    Source PMID 29500292

  • Karaa A et al. (2020) — MMPOWER-2 crossover (2020) Randomized controlled trial

    The objective endpoint — how far people walked — did not separate from placebo (398.3 m vs 378.5 m; difference 19.8 m, p=0.083), while the questionnaires about how tired people felt did (p=0.0006). When a subjective endpoint moves and the objective one does not in an 80%-injection-site-reaction drug, unblinding is a live explanation the trial cannot exclude.

    Source PMID 32096613

  • Roshanravan B et al. (2021) — single-dose IV in healthy older adults (2021) Randomized controlled trial

    A single infusion raised the muscle's measured capacity to make ATP relative to placebo (%ΔATPmax p=0.045), the effect was gone by day 7 in line with the drug's short blood half-life, and there was no improvement in how long people could keep contracting the muscle. This is the closest thing to evidence in healthy people that exists — and what it shows is a transient, reversible biochemical change in one small hand muscle, with no accompanying functional benefit.

    Source PMID 34264994

  • Ehlers JP et al. (2025) — ReCLAIM-2, dry AMD with geographic atrophy (2025) Randomized controlled trial

    Neither vision nor the rate at which the retina's atrophic patch grew was significantly better on drug. The trial's positive-sounding results — a 43% reduction in progression of ellipsoid-zone loss, and more patients gaining ≥10 letters of low-luminance acuity — are all nominal p-values on secondary measures in a trial that missed its primaries, which is hypothesis-generating and not evidence of benefit. A phase 3 in the same indication is registered — ReNEW, NCT06373731, n=313, status `ACTIVE_NOT_RECRUITING` and `hasResults: false` as verified against the ClinicalTrials.gov API on 2026-08-29 (registry record — self-asserted, unverified). No result from it exists. Nothing in the record indicates what it will find or when, and no such expectation is asserted here.

    Source PMID 39605874

What studies used — not a recommendation.

  • Barth syndrome, adults and pediatric patients weighing ≥30 kg — the approved US indication; FORZINITY label regimen · 40 mg subcutaneously once daily (20 mg once daily in adults with eGFR <30 mL/min not on dialysis; no regimen established for dialysis patients). 80 mg/mL solution containing benzyl alcohol (a regulatory product presentation, not an exposure); not approved for intravenous use; not approved in neonates · Ongoing, per prescriber
  • Barth syndrome — 12 males with genetically confirmed disease, ages 12–35; the trial the approval rests on · 40 mg subcutaneously daily · 12 weeks drug + 4-week washout + 12 weeks placebo (crossover), then a 168-week open-label extension

Common misconceptions

“SS-31 is now FDA-approved, so the anti-aging use is validated.”

The approval is real, narrow, and about a different question. On 2025-09-19 FDA granted accelerated approval to FORZINITY (elamipretide) "to improve muscle strength in adult and pediatric patients with Barth syndrome weighing at least 30 kg," on the basis of an improvement in knee extensor muscle strength, an intermediate clinical endpoint, with continued approval "contingent upon verification and description of clinical benefit in a confirmatory trial" (FORZINITY label). Barth syndrome is an X-linked disorder of TAFAZZIN-dependent cardiolipin remodelling affecting roughly 150 people in the United States (BioPharma Dive, 2025-09). These are patients whose cardiolipin is structurally abnormal by genetic defect. A cardiolipin-binding drug in that population is a mechanistically targeted replacement for a specific broken step. Nothing about that logic transfers to a person whose TAFAZZIN gene works. The transfer also fails empirically. Elamipretide has been tested in three large, non-Barth populations with acquired or age-related mitochondrial dysfunction, and in each it missed its primary endpoints: 218 adults with primary mitochondrial myopathy (MMPOWER-3), 176 adults with dry AMD (ReCLAIM-2), and — in the only randomized study in people without a mitochondrial diagnosis — 41 healthy older adults, where a single infusion raised measured muscle ATPmax but produced no improvement in fatigue resistance and no detectable effect seven days later (DOI 10.1371/journal.pone.0253849). Honest answer: The approval establishes that elamipretide improves one muscle-strength measure in a genetic cardiolipin-remodelling disease, in a trial of twelve people, under a mechanism that requires the disease. It establishes nothing about healthy adults, and the best available healthy-adult data shows a transient biochemical change with no functional payoff. Consumer marketing that treats "FDA-approved" as validation of general mitochondrial rejuvenation — for example the widely-shared framing "the biohacker mitochondrial enhancing peptide, SS-31, just received FDA approval" (Glorioso, Substack) — is describing a real regulatory event and drawing a conclusion the regulatory event does not support.

“The mitochondrial myopathy trials showed it works.”

MMPOWER-3, the phase 3, n=218, missed both co-primary endpoints — 6-minute walk distance and total fatigue (DOI 10.1212/WNL.0000000000207402). The registry entry for the open-label extension records why it was terminated in plain language: "Registration trial did not meet the primary end points" (NCT02976038). The earlier and smaller MMPOWER-2 also missed its 6MWT primary (p=0.083) while improving self-reported fatigue (DOI 10.1002/jcsm.12559). What circulates instead is the subgroup story: a sponsor-authored post hoc analysis of the failed trial reporting improvement in the nuclear-DNA and mtDNA-replisome subgroups, and a significant result in the chronic progressive external ophthalmoplegia subset (DOI 10.1186/s13023-024-03421-5). The authors present these as hypothesis-generating for a future trial, which is the correct framing. The future trial was run, and it has reported nothing. NuPOWER / SPIMD-301 (NCT05162768, n=102, 48 weeks, adults with nuclear-DNA primary mitochondrial disease) completed on 2024-12-04; as verified against the ClinicalTrials.gov API on 2026-08-29 the record stands at `COMPLETED` with `hasResults: false` and no results-first-posted date, no peer-reviewed publication was located, and the sponsor has made no statement of completion or outcome (registry record — self-asserted, unverified — cited here only for the record's own status, which is the one thing it can establish). Twenty months after completion, the prospective test of the subgroup story has produced nothing on the public record. That does not mean the trial failed — nothing available says so either way — but it does mean the subgroup story remains exactly what its authors called it: a hypothesis, generated post hoc by the sponsor from its own failed trial, and still untested in public.

“It's a mitochondrial biogenesis compound — it makes new mitochondria.”

The opposite kind of claim. Elamipretide's documented mechanism is structural: it binds cardiolipin on the inner mitochondrial membrane, stabilizing cristae and the assembly of respiratory-chain complexes into supercomplexes, which improves electron-transport efficiency and reduces reactive oxygen species production from an existing mitochondrion (DOI 10.1681/ASN.2012121216; DOI 10.5582/ddt.2025.01111). Nothing in the primary mechanism literature describes it as increasing mitochondrial number. This matters for the marketed "mito stack" division of labour — "MOTS-c makes new mitochondria, SS-31 improves the ones you have" — where the second half is roughly right about the mechanism and the pairing itself is untested.

“A vial labelled SS-31 is the same thing as the approved drug.”

SS-31 and elamipretide are the same molecule — H-D-Arg-Dmt-Lys-Phe-NH2, also carrying the codes MTP-131 and the former trade name Bendavia (DOI 10.3390/ijms26030944). The molecule being the same does not make the products the same. FORZINITY is a specific aqueous formulation with benzyl alcohol as preservative — a regulatory product presentation, not an exposure; its strength is recorded in with the label regimen — manufactured to FDA standards, dispensed by prescription through a named specialty-pharmacy channel (Stealth BioTherapeutics / AnovoRx announcement). A research-chemical vial labelled "SS-31" is an unregulated product of unverified identity, purity and sterility. SS-31 appears by name on published lists of peptides circulating in the biohacking gray market (Observer Research Foundation), and ECRI and ISMP have reported that testing of gray-market peptide products found purity ranging from 5% to 75%, with arsenic and lead above toxicity thresholds — a finding about the market generally, not about any specific compound (ECRI/ISMP, 2026).

“An FDA approval means the evidence was strong.”

This approval is a case study in the opposite. The pivotal randomized phase of TAZPOWER — 12 patients, crossover — did not beat placebo on either primary endpoint; FDA's own snapshot states elamipretide "was not superior to placebo on these primary endpoints" (FDA Drug Trials Snapshot). The approval rests on the uncontrolled open-label extension, where median knee-extensor strength rose 63 newtons from a baseline median of 124 N by week 168 in the eight patients still enrolled. FDA's advisory committee voted 10–6 in favour in October 2024; FDA nonetheless issued a Complete Response Letter in May 2025 before approving on resubmission in September 2025 (Barth Syndrome Foundation). The approval is explicitly accelerated, on an intermediate endpoint, with continued approval contingent on verification of clinical benefit in a confirmatory trial (FORZINITY label). That verification does not yet exist. A 72-week randomized placebo-controlled trial in Barth syndrome is registered and recruiting (NCT07531251; status `RECRUITING`, `hasResults: false`, verified against the ClinicalTrials.gov API 2026-08-29 — registry record — self-asserted, unverified), and its sponsor-written registry description calls it post-marketing and confirmatory; no FDA document identifying it as the postmarketing requirement was located, so treating it as that requirement is this dossier's inference, not a cited fact. No date is asserted here for when a confirmatory result will appear. None of this makes the approval wrong — for a disease affecting ~150 Americans, a 12-patient trial may be the largest one possible — but "FDA-approved" here means "reasonably likely to predict benefit, pending proof," not "proven."

“It's well tolerated, so there's no real downside.”

Tolerability in the trials was dominated by one signal, and it was near-universal. FDA records that all FORZINITY-treated patients had skin irritation or other injection-site reactions: injection-site erythema 100%, pain 75%, induration and pruritus 67% each (FDA Drug Trials Snapshot); MMPOWER-2 reported injection-site reactions in 80% of participants (DOI 10.1002/jcsm.12559). The label carries warnings for hypersensitivity reactions, including serious allergic reactions requiring emergency medical intervention, and for benzyl alcohol toxicity, with the product not approved for use in neonates (FORZINITY label). A near-100% visible local reaction rate also has a methodological consequence: in a trial where the objective endpoint did not move and the questionnaire did, functional unblinding is a live alternative explanation. ---

Who it's for, who should skip

Approved (Forzinity/elamipretide) only for Barth syndrome — an ultra-rare indication (~150 US patients). There is no practical access channel for longevity use, and research-use-only product remains unlawful. Tracked pending an aging/longevity RCT.

Running it

Form
injectable

What to expect

  • Immediate (hours) — In healthy adults aged 60–85 with low baseline mitochondrial function, a single 2-hour IV infusion raised in-vivo muscle ATPmax versus placebo immediately after infusion (%ΔATPmax p=0.045). Resting mitochondrial coupling (P/O) did not change, and fatigue resistance did not change (Confidence: Trial-documented (randomized, double-blind, n=39 analysed) — DOI 10.1371/journal.pone.0253849)
  • Days 1–7 — The same trial found no difference from placebo at day 7, consistent with the drug's short blood half-life. In mitochondrial myopathy, 5 days of IV infusion produced a dose-dependent 6MWT increase that did not reach significance (p=0.053) (Confidence: Trial-documented — DOI 10.1371/journal.pone.0253849; DOI 10.1212/WNL.0000000000005255)
  • Weeks 4 — 4 weeks of daily subcutaneous dosing in mitochondrial myopathy did not improve walking distance versus placebo (p=0.083); self-reported fatigue improved (p=0.0006). 4 weeks in heart failure did not improve left-ventricular end-systolic volume (Confidence: Trial-documented — DOI 10.1002/jcsm.12559; DOI 10.1016/j.cardfail.2020.02.001)
  • Weeks 12–28 (randomized) — In Barth syndrome, 12 weeks of drug was not superior to placebo on 6MWT or fatigue; knee-extensor strength changed +4 N on drug vs −5 N on placebo at week 12 (from a baseline median of 124 N) (Confidence: Trial-documented — FDA Drug Trials Snapshot)
  • Weeks 24–48 (randomized) — 24 weeks in mitochondrial myopathy (n=218) and 48 weeks in dry AMD (n=176) both failed their primary endpoints (Confidence: Trial-documented — DOI 10.1212/WNL.0000000000207402; DOI 10.1016/j.xops.2024.100628)
  • Long term (to 168 weeks) — In the Barth open-label extension, 6MWT improved at every timepoint (cumulative +96.1 m at week 168, p=0.003), median knee-extensor strength rose 63 N, 3D LV volumes improved, and the MLCL/CL ratio improved. No control group, no blinding, 8 patients at the final visit (Confidence: Trial-reported, uncontrolled — DOI 10.1016/j.gim.2024.101138)
  • Post-cessation — Not studied as a withdrawal endpoint in any trial. The one relevant datum is the healthy-older-adult study, where the measured effect on ATPmax was gone by day 7 after a single dose (Confidence: Trial-documented for a single dose; No data for chronic use)
  • Beyond 168 weeks — No human exposure data of any kind (Confidence: No data)
  • Controlled long-term evidence, any indication — None exists. No randomized trial of elamipretide has run beyond 48 weeks and reported. A 72-week randomized placebo-controlled trial in Barth syndrome is registered and recruiting (NCT07531251, status `RECRUITING`, `hasResults: false`, verified 2026-08-29) — a pointer, not a readout, and this dossier asserts no date for one (Confidence: No data. Registry pointer only — registry record — self-asserted, unverified)
  • Chronic exposure in healthy adults — None exists, at any duration. Every chronic-dosing trial enrolled a diagnosed disease population; the only healthy-adult exposure on record is a single infusion. A 4-week study in adults aged 65–80 is registered and recruiting (NCT07275424, status `RECRUITING`, `hasResults: false`, verified 2026-08-29), with safety and tolerability — not performance — as its primary objective. It has produced nothing (Confidence: No data. Registry pointer only — registry record — self-asserted, unverified)

Response signs

Likely working

  • Increased leg/knee-extensor strength on formal dynamometry (Barth syndrome) — Trial-documented at 12 weeks (small, +4 N vs −5 N placebo); Trial-reported, uncontrolled for the larger long-term gain · FDA Drug Trials Snapshot Trial-documented
  • Walking further on a formal 6-minute walk test over many months (Barth syndrome) — Trial-reported, uncontrolled — did not beat placebo when it was tested against placebo · DOI 10.1016/j.gim.2024.101138 Trial-reported
  • Reduced self-reported fatigue — Trial-documented but confounded — improved on questionnaires in MMPOWER-2 while the objective walk test did not, in a trial with 80% visible injection-site reactions · DOI 10.1002/jcsm.12559 Trial-documented
  • "More energy" within days in a healthy adult — Inference, not documented. The single-dose healthy-adult trial found a biochemical change with no change in fatigue resistance, gone by day 7 · DOI 10.1371/journal.pone.0253849 Inference

Adverse — seek review

  • Rash, papular lesions, eczematous dermatitis, or cough — the label's named hypersensitivity manifestations; reactions may begin anywhere from minutes to months after starting — Label-documented. Serious reactions require emergency treatment; patients who have had one should not be rechallenged · FORZINITY label
  • Injection-site reaction that is spreading, ulcerating, or not resolving between doses — Trial-documented that reactions are near-universal; escalation beyond mild local reaction is the part that warrants review · FDA Drug Trials Snapshot Trial-documented
  • Low blood pressure or presyncope — Label-documented as the expected, histamine-related presentation of overdose · FORZINITY label
  • Any systemic reaction — fever, malaise — following a dose of a product not obtained through a pharmacy — Inference, not documented for elamipretide. Flagged because gray-market peptide purity has been measured as low as 5%, with heavy-metal contamination above toxicity thresholds · ECRI/ISMP, 2026 Inference

Common / neutral

  • Mild injection-site redness, pain, induration or itching — in trials this was the rule, not the exception — Trial-documented: erythema 100%, pain 75%, induration and pruritus 67% · FDA Drug Trials Snapshot Trial-documented
  • No perceptible subjective effect — Consistent with the trial record. Objective endpoints failed to separate from placebo in the three largest randomized trials · WNL.0000000000207402; xops.2024.100628; cardfail.2020.02.001

Getting it right

  • Monitoring: Hypersensitivity monitoring is an explicit label instruction, not a general precaution. The label directs prescribers to monitor for signs and symptoms of hypersensitivity throughout treatment, notes that reactions have occurred anywhere from minutes to months after initiation, and states that a patient who has had a serious reaction should not be rechallenged (FORZINITY label). mitigate side effect
  • Monitoring: Renal function determines the dose. No adjustment for mild or moderate impairment; the dose is halved in adults with eGFR <30 mL/min not on dialysis; no regimen is established for dialysis patients (FORZINITY label). mitigate side effect
  • Monitoring: The efficacy endpoint that was approved is a measured one. Approval rests on knee-extensor strength by handheld dynamometry, not on how a patient reports feeling (FDA Drug Trials Snapshot). Whatever is tracked, tracking it from a documented pre-treatment baseline is what makes any later change interpretable — the randomized phase moved 9 N between arms over 12 weeks, which is not a difference anyone perceives without measuring. mitigate side effect
  • Monitoring: Accept the surrogate problem. The approval is on an intermediate endpoint "reasonably likely" to predict clinical benefit, with continued approval contingent on verification in a confirmatory trial (FORZINITY label). No such verification exists today. A 72-week randomized placebo-controlled trial is registered and recruiting (NCT07531251 — registry record — self-asserted, unverified, verified 2026-08-29); identifying it as the postmarketing requirement is this dossier's inference, and no date is asserted for a result. mitigate side effect
  • General safety: Injection-site reactions are the expected experience, not a rare event. The label's administration instructions address them directly: rotate the injection site daily between abdomen (at least 2 inches from the navel) and outer thigh, and do not inject where skin is tender, bruised, red or hard, or into scars or stretch marks (FORZINITY label). Pharmacokinetic exposure is comparable at thigh and abdomen, which is why rotation is possible. mitigate side effect
  • General safety: Benzyl alcohol. As a regulatory formulation record — an excipient concentration in the approved product presentation, not an administered quantity of elamipretide — the product contains 20 mg of benzyl alcohol per mL as preservative and is not approved for use in neonates; fatal reactions including gasping syndrome have been reported in low-birth-weight and preterm neonates given benzyl-alcohol-containing drugs (FORZINITY label). mitigate side effect
  • General safety: Route matters. The label states FORZINITY is not approved for intravenous use, even though the compound's early trials used IV infusion. mitigate side effect
  • General safety: Product handling is specified. Vials are stored refrigerated, must be visually inspected and not used if cloudy or containing particulates, and are discarded 8 days after first opening (FORZINITY label). mitigate side effect
  • General safety: Product identity is the dominant risk outside the pharmacy channel. Gray-market peptide testing has found purity from 5% to 75% and heavy-metal contamination above toxicity thresholds (ECRI/ISMP, 2026). None of the label protections above exist for a research-chemical vial. mitigate side effect
  • The physician conversation: Is there a genetically or biochemically confirmed mitochondrial diagnosis, and which one? (The one approved indication requires a confirmed TAFAZZIN variant; the largest trial in genetically confirmed mitochondrial myopathy was negative.) mitigate side effect
  • The physician conversation: What specific, measurable outcome would count as it working, measured how, and by when? (The approval endpoint is dynamometry, not symptoms.) mitigate side effect
  • The physician conversation: What is baseline renal function? (It determines the labelled dose.) mitigate side effect
  • The physician conversation: Is there any history of drug hypersensitivity or serious allergic reaction? mitigate side effect
  • The physician conversation: What is the plan when injection-site reactions occur, given that in trials essentially everyone had them? mitigate side effect
  • The physician conversation: Where would the product come from — a pharmacy dispensing an FDA-approved product, or anywhere else? (.) mitigate side effect
  • The physician conversation: Is there any competitive-sport or military testing obligation, and is the product's contents verifiable? (.) mitigate side effect
  • The physician conversation: What does insurance coverage look like, and what does the manufacturer support program actually cover? (.) mitigate side effect

Stacking & alternatives

  • Complements MOTS-c — The in-batch mitochondrial comparator and the compound SS-31 is most often paired with in marketing. Mechanistically distinct: elamipretide binds cardiolipin and stabilizes existing cristae (ASN.2012121216); MOTS-c is a mitochondrially-encoded peptide signalling through folate/purine metabolism to AMPK. The evidence contrast is stark: elamipretide has 20+ registered trials and an FDA approval; MOTS-c has no completed human trial of exogenous administration

Access & cost

A lawful channel exists, so this section reports real prices. FORZINITY is dispensed through a limited specialty-pharmacy arrangement with AnovoRx, alongside a manufacturer patient-support program ("Mito Assist") (Stealth BioTherapeutics). Published cash pricing is consistent with ultra-rare-disease economics. A price-guide listing puts FORZINITY at approximately $4,259.68 per mL, or $59,635.52 for 14 mL — one carton of four 3.5 mL vials (Drugs.com price guide). Reading that against the label makes the scale explicit: on the label's once-daily regimen — stated in and, under the batch rule on administered quantities, not restated here — a 14 mL carton is approximately a 28-day supply, implying roughly $770,000–$780,000 per year at cash list price (FORZINITY label). ⚠️ That annual figure is this dossier's own arithmetic — a derived number, not a published one. No manufacturer, payer, regulator or journalist has published it; it is the price-guide per-mL listing multiplied out against the label's once-daily regimen and the carton's days of supply, and it is repeated nowhere else. It is also a cash-payer figure that does not reflect negotiated, rebated, or patient-assistance pricing. The two sourced figures are the per-mL and per-carton listings; everything past them is derivation. Barth syndrome affects roughly 150 people in the United States (BioPharma Dive), and payer-side considerations for the indication have begun appearing in the managed-care literature (J Manag Care Spec Pharm 2026). > # ⚠️ None of the figures above is the price of anything sold as "SS-31." > > This is the third of the three banners this page carries in place of a split into two dossiers. The naming was ruled on 2026-08-29: one page, titled Elamipretide (SS-31), slug `ss-31` kept on purpose. So the boundary has to be said here too. Everything priced above is FORZINITY — an approved orphan drug, dispensed by prescription through a named specialty pharmacy for a genetic disease affecting roughly 150 Americans. A vial labelled "SS-31" from a research-chemical seller is not FORZINITY. It has no label, no verified identity, purity or sterility, and nothing in or applies to it: not the regimen, not the price, not the pharmacy channel, not the support program. A high pharmacy price for an orphan drug is not evidence that a cheap gray-market vial contains the same thing, and it is not evidence that either one does anything for a healthy adult. A separate gray market in vials labelled "SS-31" also exists and predates the approval; SS-31 appears by name on published inventories of peptides circulating through biohacking channels (Observer Research Foundation), and consumer-facing clinic and vendor content markets it for energy and longevity (Revolution Health & Wellness). > No INTERNAL subsection appears in this dossier. We found no journalist-reported or otherwise non-seller-sourced specific price figures for gray-market "SS-31." Every specific figure we located came from sellers or affiliate "buying guides," which this brief does not permit as sources. That is an absence, not an omission. ---

Questions

Does SS-31 actually work?
It depends entirely on who is being asked about. In Barth syndrome, FDA accepted that it improves knee-extensor muscle strength — an intermediate endpoint, in twelve patients, mostly from an uncontrolled extension, under accelerated approval pending confirmation (FORZINITY label). In every larger randomized trial in a non-Barth population — primary mitochondrial myopathy (n=218), dry AMD (n=176), heart failure (n=71) — it missed its primary endpoints (WNL.0000000000207402; xops.2024.100628; cardfail.2020.02.001). In healthy older adults, one infusion moved a biochemical measure and did not move a functional one (pone.0253849).
Is SS-31 the same thing as elamipretide?
Yes as a molecule, no as a product. SS-31 is the original laboratory designation from the Szeto–Schiller peptide series; MTP-131 and Bendavia are development codes; elamipretide is the international nonproprietary name; FORZINITY is the US brand (DOI 10.3390/ijms26030944). A vial sold under the "SS-31" research-chemical label is not the approved drug and has no verified identity, purity or sterility — SS-31 appears on published inventories of peptides circulating in the gray market (Observer Research Foundation), and gray-market peptide testing has found purity between 5% and 75% with arsenic and lead above toxicity thresholds (ECRI/ISMP, 2026).
Oral, nasal, or injection — which works?
Only subcutaneous and intravenous administration have ever been studied in humans, plus a topical ophthalmic solution in eye trials (NCT02693119; strength not reproduced — that programme is investigational and has no channel). The approved product is subcutaneous once daily and the label states explicitly that it is not approved for intravenous use (FORZINITY label). There is no human bioavailability or efficacy study of oral, nasal or transdermal elamipretide. Any claim that an oral or nasal "SS-31" product delivers what the injectable delivers is unsupported by any published human data in either direction.
How long until it works?
The only rapid, controlled human effect on record is biochemical and transient: a rise in muscle ATPmax measured immediately after a single 2-hour infusion, absent by day 7 (pone.0253849). Functional change, where it has been reported at all, is on the scale of a year or more and comes from an uncontrolled extension — the Barth 6MWT gains accumulated across 168 weeks (gim.2024.101138). Trials at 4, 12, 24 and 48 weeks in various populations mostly showed nothing versus placebo.
What are the side effects?
Injection-site reactions, at very high rates. FDA records that all FORZINITY-treated patients had injection-site skin reactions — erythema 100%, pain 75%, induration and pruritus 67% each (FDA Drug Trials Snapshot); MMPOWER-2 reported 80%, mostly mild (jcsm.12559). The label warns of hypersensitivity reactions including serious allergic reactions requiring emergency intervention, and of benzyl alcohol toxicity (the approved product carries benzyl alcohol as a preservative — a formulation record, quantified in — and is not approved in neonates) (FORZINITY label).
Will it help a healthy person's energy, endurance or aging?
No trial has answered this. The single relevant randomized study enrolled older adults specifically screened for poor mitochondrial function, gave one intravenous dose, and found improved ATP production with no improvement in fatigue resistance (pone.0253849). Chronic exposure in healthy adults has never been studied at any duration, so there is no evidence base here to summarise — that absence is the answer. A four-week study in adults aged 65–80 is registered and recruiting, with safety and tolerability rather than performance as its primary objective (NCT07275424; status `RECRUITING`, `hasResults: false`, verified against the ClinicalTrials.gov API 2026-08-29 — registry record — self-asserted, unverified). It has reported nothing, and nothing here predicts what it will report or when.
Is it banned in sport?
Not as of the 2026 List. This changed with the approval: before September 2025 an unapproved investigational compound was captured by S0, and elamipretide now has approval from a governmental regulatory health authority for human therapeutic use, which is the criterion S0 turns on (2026 Prohibited List).
Why did it fail so many trials if the mechanism is real?
The mechanism is well characterized — cardiolipin binding, cristae stabilization, supercomplex assembly (ASN.2012121216) — and that is exactly the point worth taking from this compound's record: a clean, replicated, molecular mechanism is not the same as a clinical benefit. Where the disease is a defect in cardiolipin remodelling itself (Barth syndrome), the drug reached approval. Where mitochondrial dysfunction is one contributor among many (mitochondrial myopathy of mixed genotype, AMD, heart failure), it repeatedly did not beat placebo. ---

Last reviewed 27 August 2026. Regulatory status verified 27 August 2026.