Elamipretide: Its Mechanism, Uses, and Clinical Status

Elamipretide is a synthetic four-amino-acid peptide engineered to reach mitochondria and stabilize their internal membranes. In September 2025, it became the first disease-specific treatment approved for Barth syndrome in the United States, granted under the FDA’s accelerated approval pathway. But behind that milestone sits a broader and more complicated clinical story, with trials across eye disease, heart failure, mitochondrial myopathy, and kidney injury producing results that range from encouraging to flatly negative.

How Elamipretide Reaches and Repairs Mitochondria

Mitochondria generate most of the energy your cells use, packaging it as ATP. They do this across a series of folded internal membranes called cristae. Those folds depend on a specialized fat molecule called cardiolipin to maintain their shape and function. When cardiolipin gets damaged or depleted, as happens in genetic disorders, aging, or after a heart attack, the cristae lose their structure, energy production drops, and harmful byproducts pile up.

Elamipretide’s small size and positive charge let it cross cell membranes and concentrate inside mitochondria. Once there, it binds directly to cardiolipin in the inner mitochondrial membrane, stabilizing the cristae structure, reducing oxidative stress, and improving ATP output.¹ The peptide works even on cardiolipin molecules that are already oxidized. Experiments in model membranes showed that elamipretide physically clusters cardiolipin molecules together, acting like a kind of adhesion factor that gathers whatever functional cardiolipin remains and reorganizes it into working domains.² Think of it less as replacing broken parts and more as rearranging the parts that are still there into a configuration that functions better.

This mechanism is what makes elamipretide interesting across seemingly unrelated diseases. Mitochondrial dysfunction shows up in inherited metabolic disorders, heart failure, retinal degeneration, kidney injury, and neurodegeneration. A drug that can shore up mitochondrial membranes regardless of what damaged them, at least in theory, has a wide potential reach. The clinical evidence, however, has not always matched that theoretical promise.

Barth Syndrome and the First FDA Approval

Barth syndrome is an ultra-rare genetic disorder, X-linked and recessive, meaning it almost exclusively affects boys and men. The underlying problem is a mutation in the gene responsible for remodeling cardiolipin, which means patients have abnormal cardiolipin from birth. The consequences are severe: weakened heart muscle, skeletal muscle weakness, fatigue, exercise intolerance, and immune problems. With no disease-specific treatments previously available, management was limited to addressing symptoms.

The pivotal trial for elamipretide in Barth syndrome was called TAZPOWER, and the long-term extension data told a compelling story. Over 168 weeks of treatment, patients showed a cumulative improvement of about 96 meters on the six-minute walk test, a standard measure of functional exercise capacity. Fatigue scores improved and stayed below baseline at every measured time point. Cardiac imaging showed improvements in left ventricular stroke volume and other measures of heart function. And the ratio of monolysocardiolipin to cardiolipin, a biochemical marker specific to Barth syndrome, improved in a way that correlated with clinical outcomes.³ Injection-site reactions were the most common side effect, and overall tolerability was good.

In September 2025, the FDA granted accelerated approval for elamipretide to improve muscle strength in adult and pediatric patients with Barth syndrome weighing at least 30 kilograms, making it the first treatment targeting the disease itself rather than its symptoms.⁴ Accelerated approval means the drug can be marketed based on a surrogate or intermediate endpoint that is reasonably likely to predict clinical benefit, with the requirement that confirmatory trials continue. For a community as small as the Barth syndrome population, estimated at a few hundred diagnosed individuals worldwide, that approval was a significant event.

Age-Related Macular Degeneration

Dry age-related macular degeneration, the form that involves gradual retinal deterioration rather than leaky blood vessels, has far fewer treatment options than its “wet” counterpart. A key feature of advanced dry AMD is geographic atrophy, where patches of retinal cells die and vision progressively worsens. The retina is one of the most metabolically active tissues in the body, and its light-sensing cells depend heavily on mitochondria. That made it a logical target for elamipretide.

A small Phase 1 trial called ReCLAIM enrolled patients with dry AMD and noncentral geographic atrophy. Among the 15 participants who completed the study, best-corrected visual acuity improved by an average of about 4.6 letters after 24 weeks, and low-luminance visual acuity, which reflects how well you see in dim light, improved by about 5.4 letters. Patients also reported better dim-light reading and general dim-light vision.⁵ For a disease where any visual improvement is unusual, those results generated real enthusiasm.

The follow-up Phase 2 trial, ReCLAIM-2, was larger and randomized against a placebo. Here the results were more nuanced. The study did not meet its primary endpoints for mean change in low-luminance visual acuity or geographic atrophy growth. But elamipretide did produce a roughly 43 percent reduction in the progression of complete loss of the ellipsoid zone, a layer of photoreceptor structures visible on retinal imaging, and a 47 percent reduction in partial degradation of that zone compared to placebo. More patients on elamipretide gained 10 or more letters in low-luminance visual acuity than those on placebo.⁶ So the drug appeared to slow structural damage to photoreceptors even though it didn’t clearly move the overall group averages on the headline measures. The ellipsoid zone findings have been enough to keep elamipretide in Phase 3 development for dry AMD.

Heart Failure Trials and Their Mixed Signals

Failing hearts are energy-starved hearts. Mitochondrial dysfunction is a well-established feature of heart failure with reduced ejection fraction, and animal studies of elamipretide in heart failure models were striking. In dogs with experimentally induced advanced heart failure, chronic elamipretide treatment restored mitochondrial respiration rates, membrane potential, ATP synthesis, and the ATP-to-ADP ratio. It also lowered inflammatory markers in the blood.⁷

Early human testing seemed to track with those preclinical results. In a first-in-patient trial, a four-hour intravenous infusion of the highest dose of elamipretide produced a significant decrease in left ventricular end-diastolic volume (about 18 mL) and end-systolic volume (about 14 mL) compared to placebo, and the changes correlated with how much drug reached the bloodstream.⁸ Smaller heart volumes in someone with heart failure generally mean the heart is pumping more efficiently.

Then came PROGRESS-HF, a Phase 2 trial testing four weeks of subcutaneous elamipretide injections at two different doses in patients with heart failure and reduced ejection fraction. Neither dose produced a significant change in left ventricular end-systolic volume or ejection fraction compared to placebo.⁹ The gap between the acute intravenous effect and the four-week subcutaneous result raised difficult questions: was the earlier positive result a transient hemodynamic effect rather than a true structural change? Was four weeks too short? Was the subcutaneous route delivering enough drug? The field has not fully resolved those questions, and heart failure is not currently listed among elamipretide’s active development programs.

Primary Mitochondrial Myopathy

Primary mitochondrial myopathies are a group of inherited disorders where mitochondrial dysfunction directly impairs skeletal muscle. Patients experience exercise intolerance, fatigue, and progressive weakness. If elamipretide’s mechanism works anywhere, it should logically work here, and there was strong rationale for testing it.

The MMPOWER-3 trial was a randomized, placebo-controlled study in patients with genetically confirmed primary mitochondrial myopathy. It did not meet its primary endpoints. The difference in six-minute walk distance between elamipretide and placebo at 24 weeks was about negative 3.2 meters, essentially no difference. The fatigue score showed the same story.¹⁰ This was a clear negative result, and a disappointing one given how biologically plausible the target seemed. Elamipretide remains in Phase 3 development for mitochondrial myopathies according to its regulatory profile, but the path forward after a failed pivotal trial is not straightforward.

Kidney Protection During Acute Injury

The kidneys are among the most metabolically demanding organs in the body, and they are exquisitely sensitive to interruptions in blood flow. When blood supply returns after a period of reduced flow, as happens during certain surgical procedures or after a hemorrhage, the resulting “reperfusion” paradoxically causes further damage through oxidative stress and mitochondrial disruption. This ischemia-reperfusion injury is a major clinical problem with no specific treatment.

In a Phase 2a trial involving patients undergoing stent placement for atherosclerotic renal artery stenosis, elamipretide infusion during the procedure attenuated the post-stenting rise in tissue hypoxia (a measure of oxygen deprivation in the kidney). In the placebo group, a marker of kidney tissue oxygen deprivation rose by 47 percent after the procedure. In the elamipretide group, it was essentially unchanged.¹¹ These were small numbers, but the direction was consistent with the biological rationale.

Related work in the acute-care setting has explored whether elamipretide can protect organs during hemorrhagic shock and resuscitation, though that research remains at a preclinical stage. The common thread is that whenever cells face energy crises from interrupted blood supply, stabilizing their mitochondrial membranes might limit the cascade of damage.

Preclinical Neuroprotection Research

Mitochondrial dysfunction is a recurring feature of neurodegenerative diseases. Neurons are energy-intensive cells, and their long projections make them especially vulnerable when mitochondria falter. A systematic review of elamipretide’s effects in preclinical models of neurodegeneration found that it boosted mitochondrial respiration, promoted the growth of new mitochondria, encouraged healthy mitochondrial fusion, and increased the clearance of damaged mitochondria. It also reduced markers of oxidative stress and inflammation across multiple models.¹²

In one mouse study, elamipretide protected against memory and learning impairments caused by systemic inflammation. The drug reversed changes in proteins important for building and maintaining the connections between neurons and supported signaling pathways involved in neural survival.¹³ These findings are interesting, but they remain entirely in animal and cell models. No clinical trials for neurodegenerative conditions have been completed, and the history of drugs that look promising in mouse brains but fail in human brains is long. Still, the breadth of preclinical evidence across different neurodegeneration models has been enough to sustain interest in this application.

A Window Into Energy and Aging Muscle

Age-related decline in muscle function tracks closely with declining mitochondrial performance. A randomized trial in older adults tested whether even a single two-hour intravenous infusion of elamipretide could measurably change mitochondrial energy production in skeletal muscle. Using specialized imaging to measure ATP generation in the leg, the study found that participants who received elamipretide had a mean 27 percent increase in maximal ATP production capacity from baseline to immediately after the infusion, compared to 12 percent in those who received a placebo.¹⁴ The finding just crossed the threshold for statistical significance, and it was from a single dose, so it says more about proof-of-concept than about practical treatment for age-related sarcopenia. But it does suggest that the drug’s mechanism translates from animal models into measurable human physiology, even in people whose mitochondrial decline is from aging rather than a genetic defect.

What Happens When Elamipretide Stops

One of the most telling pieces of evidence about elamipretide’s real-world impact comes not from randomized trials but from case reports of patients who received the drug through expanded-access programs and then had it discontinued. A case series of patients with rare mitochondrial disorders involving ophthalmic symptoms documented both sides of that experience. One patient’s walking capacity on the six-minute walk test more than doubled during treatment, from 120 meters to 259 meters, and his fatigue and quality-of-life scores improved. But within about four months of stopping elamipretide, he lost all those gains. His family reported worsened balance, mobility, weakness, lethargy, and confusion. He lost most of his independence and could no longer be left alone at home.¹⁵

These are individual cases, not controlled data, and they should be interpreted with that in mind. But they do point to two things worth noting. First, the improvements patients experience on elamipretide appear to be real and functionally meaningful, affecting daily independence rather than just lab numbers. Second, the drug does not repair the underlying genetic defect; it compensates for it. Remove the compensation, and the underlying dysfunction reasserts itself. For conditions like Barth syndrome or other genetic mitochondrial disorders, that means elamipretide is a chronic therapy rather than a cure.

Biomarkers and Tracking Mitochondrial Damage

Developing better ways to measure mitochondrial injury in real time could help clinicians decide when and how to use drugs like elamipretide. An interesting side finding from a study in patients with ST-elevation myocardial infarction, the most severe type of heart attack, involved a protein called HtrA2. This protein is released from mitochondria when cells undergo programmed death. Researchers found that HtrA2 was detectable in the blood of heart attack patients and that elamipretide treatment reduced its levels, suggesting less mitochondrial-driven cell death.¹⁶ Whether HtrA2 becomes a clinically useful biomarker remains to be seen, but the concept of being able to measure the specific type of damage elamipretide targets, rather than relying on general markers of organ injury, could eventually help match the right patients to the drug and confirm whether it is working in each individual.

Why Results Vary So Much Across Diseases

If elamipretide works by fixing mitochondrial membranes, why does it clearly help some patients and fail in formal trials for seemingly similar conditions? Several factors likely contribute. The genetic mitochondrial myopathy trial, for instance, enrolled patients with a range of different underlying mutations. Some of those mutations may directly affect cardiolipin biology, making elamipretide highly relevant, while others may impair mitochondrial function through pathways that have little to do with membrane structure. In Barth syndrome, where the genetic defect is specifically in cardiolipin remodeling, elamipretide’s mechanism is a much closer match to the problem.

Trial design matters as well. The six-minute walk test, used as a primary endpoint in both the Barth syndrome and mitochondrial myopathy trials, may capture changes better in one disease context than another. Patients with different baseline severities, different compensatory adaptations, and different day-to-day variability in symptoms will produce different signal-to-noise ratios in clinical measurement. A drug can be biologically active without producing statistically significant results on a particular test in a particular population over a particular time frame.

The route and duration of treatment also differ across trials. A four-hour intravenous infusion in a controlled setting delivers drug differently than daily subcutaneous injections at home. The heart failure experience, where acute infusion produced measurable cardiac changes but weeks of injections did not, illustrates how pharmacokinetic details can determine whether a biological effect translates into a clinical signal. These are not failures of the underlying science so much as reminders that drug development involves far more than having the right molecular target.