Body journey · 8 steps
Inside the body with Elamipretide (SS-31)
Elamipretide is a tiny four-piece protein that gets inside cells and sticks to the inner wall of mitochondria, the parts that make energy. In September 2025 the US FDA gave it a conditional (accelerated) approval for Barth syndrome, a rare genetic muscle and heart disease. Its other tried uses are not approved.
Scroll to follow it through the body
- 01Skin
Injected under the skin and absorbed
The medicine is a small protein, so it cannot be swallowed. It is injected into the fat just under the skin and passes into the blood, mostly within an hour.
- 02Bloodstream
Travels in the blood and is cleared quickly
Only a bit of it sticks to blood proteins. The body chops it into two smaller pieces that do nothing, and the kidneys remove it in urine. Half of it is gone in about three to four hours.
For experts +For experts −
Distributed through total body water (about 0.5 L/kg) with about 39% plasma protein binding. Metabolised by sequential C-terminal degradation to the M1 tripeptide and M2 dipeptide, both pharmacologically inactive; about 100% of a dose is recovered in urine within 48 h. Half-life is about 3-4 h. AUC increases 39%, 75% and 125% in mild, moderate and severe renal impairment, so the label reduces the dose in severe impairment.[src][src]
- 03Elsewhere
Slips into cells and settles in mitochondria
Elamipretide can pass into cells on its own. In lab work with cells and isolated mitochondria it piled up in the inner wall of mitochondria, the tiny power plants inside cells. This has been shown in the lab, not measured directly inside people.
For experts +For experts −
The Szeto-Schiller design (alternating basic and aromatic residues, dimethyltyrosine) gave peptides that were cell-permeable and concentrated about 1000-fold in the inner mitochondrial membrane in cell and isolated-mitochondria experiments. They reduced mitochondrial ROS production and permeability-transition swelling in isolated mitochondria and improved contractile force in an ex vivo (isolated) heart model of ischaemia-reperfusion; analogues without dimethyltyrosine did not. This was preclinical work; direct tissue mitochondrial concentrations in humans are not established.[src][src]
- 04Elsewhere
Sticks to cardiolipin on the mitochondrial membrane
Mitochondria have a special fat, cardiolipin, that holds the energy machinery in place. Elamipretide clings to it and to nearby proteins. Scientists think this helps keep the membrane in good shape. This is based on lab and animal studies.
For experts +For experts −
Elamipretide binds cardiolipin with high affinity (fluorescent-analogue work, rat kidney model), partitions into anionic bilayers and modulates surface electrostatics and calcium distribution, and cross-links to cardiolipin-binding proteins of oxidative phosphorylation and 2-oxoglutarate metabolism in mouse-heart mitochondria. In vitro, the elamipretide-cardiolipin complex inhibited cytochrome c peroxidase activity; in rat ischaemia-reperfusion models (kidney, heart) it preserved cristae membranes and ATP recovery, although in the rat heart it did not prevent the fall in cardiolipin content. These are cell, biophysical and animal findings; the exact molecular mechanism in humans is not settled.[src][src][src][src]
- 05Muscle
Barth syndrome: muscle strength in an open-label extension
In Barth syndrome the body cannot finish building cardiolipin, so mitochondria work poorly and muscles are weak. In a tiny study, elamipretide did not beat placebo at first. Later, when everyone knew they were on the drug, knee muscle strength rose. Without a comparison group, that result is weak proof.
For experts +For experts −
Barth syndrome results from TAZ variants that impair cardiolipin remodelling. In TAZPOWER (12 patients), neither primary endpoint (6-minute walk, fatigue score) was met in the randomised part. In the open-label extension, median knee-extensor strength change from the pre-dose baseline (median 124 N) was +63 N at week 168 in the 8 patients still enrolled. This intermediate endpoint underlies the accelerated approval; the FDA review recorded reviewer disagreement because of the uncontrolled extension design. The sponsor-authored extension paper reported improvement in the MLCL/CL ratio, but the FDA review concluded that elamipretide does not meaningfully change the elevated cardiolipin ratio and did not accept it as a surrogate endpoint.[src][src][src][src]
- 06Heart
Heart: promising in animals, no clear gain in human trials
In rats, the drug protected the inner structure of heart-cell mitochondria after a temporary blood-flow block. In people with heart attacks or long-term heart failure, the main goals of the studies were not reached.
For experts +For experts −
Rat cardiac ischaemia-reperfusion data show preserved cristae ultrastructure and respiratory complex activity (animal-only evidence). In humans, EMBRACE STEMI (phase 2a, intravenous, anterior STEMI) did not reduce infarct size by CK-MB AUC over 72 h, and PROGRESS-HF (phase 2, 71 patients with HFrEF, 28 days) did not change left-ventricular end-systolic volume versus placebo. Neither is an approved use.[src][src][src]
- 07Muscle
A short-lived energy boost in older muscle
In one small human study, a single dose may have briefly raised how fast a hand muscle in older adults could make energy. The result was borderline, it was gone a week later, and the muscle did not tire more slowly.
For experts +For experts −
In a randomised, double-blind, placebo-controlled trial in 39 healthy adults aged 60-85 selected for poor mitochondrial function, a single 2-hour intravenous infusion was associated with higher in vivo mitochondrial capacity (ATPmax) in the first dorsal interosseous (hand) muscle immediately afterwards, a borderline result (P=0.055 for the absolute change, P=0.045 for percent change), with no difference at day 7 and no effect on fatigue resistance. Use in ageing is investigational and unapproved.[src]
- 08Elsewhere
Eye disease: tested, not approved
Researchers also tried elamipretide against a common eye disease that damages the light-sensing cells (dry macular degeneration). The first mid-size trial missed its main targets. In extra measures that were not its main goals, one layer of those cells was lost more slowly. A larger trial is still running. It is not approved for the eye.
For experts +For experts −
ReCLAIM-2 (phase 2, 176 patients with geographic atrophy, 48 weeks) missed its primary endpoints (low-luminance BCVA and square-root GA area). It reported a 43% smaller progression of total ellipsoid-zone loss versus placebo (nominal P=0.0034) and more patients with a gain of 10 letters or more in low-luminance BCVA (14.6% vs 2.1%, nominal P=0.0404). These are secondary or predefined endpoints, not confirmatory. The phase 3 ReNEW trial (NCT06373731) has an estimated primary completion of August 2027.[src][src]
Schematic animation — real structure where marked
Under the skin
AI-generated illustration · not to scale
About these visuals
What is real and what is drawn
The scenes are schematic animations made for this page. Shapes, colours, speeds and sizes are chosen to explain, not to measure. Nothing is to scale.
Transitions between scales use short clips labelled “AI-generated illustration”. They are generated images, not recordings or simulations.
Evidence, legal status and all sources for Elamipretide (SS-31) →
Keep exploring
More body journeys
Related molecules