Mitochondrial transplantation into the eye: the first human case

Mitochondrial transplantation into the eye has now been performed in a human: mitochondria were isolated from leg muscle and injected into the vitreous of both eyes. In a 26-year-old woman with profound vision loss, no obvious acute toxicity was detected, and measurable pupillary responses to light temporarily reappeared after treatment. Her vision itself did not measurably recover.

Mitochondria supply much of the energy that cells need to function. Neurons are particularly dependent on them, which means severe loss of blood flow and oxygen can damage a neuron’s energy system before the cell itself is completely lost. That creates an unusual therapeutic possibility: perhaps injured but still viable cells could be supported by supplying them with additional functioning mitochondria.

The idea had already been explored in experimental models. Mitochondria delivered into the vitreous of animals can be taken up by retinal cells, and mitochondrial transplantation after optic nerve injury has been associated with improved survival of retinal ganglion cells. Those experiments, however, could not establish whether the same approach would be safe or useful in a human eye.

Moving mitochondria from muscle to the eye

The patient was a previously healthy 26-year-old woman who suffered a severe intracerebral hemorrhage after rupture of a cerebral arteriovenous malformation. Prolonged hypoperfusion and increased intracranial pressure were followed by profound damage to the visual pathways. Optic nerve atrophy was evident by eight weeks, and the severe visual deficit had remained largely fixed for about three months by the time the experimental procedure was performed.

The treatment was authorized in the United States under emergency expanded access for a single patient. For each procedure, doctors removed a small biopsy from the vastus medialis muscle and isolated fresh autologous mitochondria immediately before administration. The mitochondria were not cultured, expanded, genetically modified or frozen.

About 75 million mitochondria were injected into the right eye and about 25 million into the left. The procedures were performed 24 hours apart.

For this first use, the main question was not whether vision would return. It was whether fresh autologous mitochondria could be prepared and delivered into the human eye without obvious harm. The authors reported no serious adverse events, increase in intraocular pressure, intraocular inflammation, endophthalmitis or traction retinal detachment after either injection. Measures used for inflammatory and immune surveillance remained within reference ranges through four weeks.

A single patient cannot establish that the procedure is generally safe. It does show, however, that the technical process was feasible.

What happened to the response to light

The most intriguing signal came from automated pupillometry, which measures how the pupils respond to light. During the 71 days immediately before treatment, 45 measurements were obtained, and none reached the study’s predefined threshold for a normal response.

After treatment, each eye reached that threshold within days of its own injection: on day four in the right eye and day two in the left. Overall, 13 of 52 post-treatment measurements met the criterion.

The effect was transient. In the left eye, the last qualifying response occurred on day 11. In the right eye, occasional normal responses appeared later, including on day 39, but neither eye established a new sustained baseline.

Most importantly, measurable visual acuity did not improve. A pupillary response to light is an objective physiological sign that part of the visual pathway is responding, but it is not equivalent to useful vision. These results therefore cannot be described as restoration of sight.

Why one case cannot establish efficacy

Cell and animal studies have shown that cells can take up mitochondria from their surroundings. If an injured neuron is still alive but its own mitochondria are damaged and energy production is impaired, functioning mitochondria supplied from outside could in principle provide temporary metabolic support. They cannot, however, revive a neuron that has already died.

That mechanism explains the rationale for the treatment, but the human case did not demonstrate that it actually occurred. The investigators did not directly track the injected mitochondria inside the patient’s eyes, so it remains unknown which cells took them up, how long they persisted, or whether they caused the change in pupillary responses.

There are other important limitations. This was a single uncontrolled case. Pre-treatment pupillometry was intermittent rather than continuous and included a 52-day gap, while isolated normal readings had occurred during the first weeks after the original hemorrhage. The patient was also receiving other interventions, making it impossible to completely exclude spontaneous fluctuation or contributions from concurrent treatment.

The report is also currently a preprint and has not yet undergone peer review.

This is therefore not a new treatment for blindness. The result is narrower: fresh mitochondria isolated from a patient’s own muscle were successfully delivered into her eyes without detected short-term toxicity, followed by transient objective changes in the response to light. Whether the injections actually caused those changes will require controlled studies.