What exactly was in the vial?
An eye drop for dry eye, an injection into an arthritic knee and an infusion carrying a cancer drug can all be described as exosome treatments. The name alone tells you very little about the contents of the vial or what happened to people who received it.
Cells release tiny particles enclosed in a membrane. These extracellular vesicles, or EVs, can carry proteins, fats and RNA. An exosome is a type of EV that forms inside a cell in a compartment called an endosome. Scientists use the broader term EV when an experiment identifies the particles but cannot establish where they formed. Size alone cannot make that distinction. [1]
The source matters. In one laboratory experiment, researchers put small vesicles from six cell sources alongside rat retinal nerve cells. Vesicles from fat-derived and bone marrow stromal cells helped the nerve cells survive under the test conditions. Vesicles from several other sources showed no measurable benefit in that experiment. That finding helps researchers choose what to test next. It cannot tell a patient which preparation will preserve vision. [2]
What did people experience in the eye and knee trials?
Eight people with Sjögren’s dry eye used drops made from umbilical cord tissue MSCs in one randomly assigned eye and a salt solution in the other. MSCs are mesenchymal stromal cells. After two weeks of treatment, the researchers found less staining on the surface of the treated eyes, a sign of less surface damage. The difference was still detectable at 12 weeks. Each participant contributed both eyes, so the study gives us information from eight people. A larger trial could establish how much relief people feel and how long it lasts. [3]
A knee study used a similar design and found a different result. Thirty-one people with arthritis in both knees enrolled. One knee received vesicles made from placenta-derived MSCs and the other received saline. Among the 29 people included in the analysis, the vesicles showed no significant advantage for pain or function at two or six months, or for MRI findings at six months. Four participants reported moderate pain and warmth after injection that settled within two days. [4]
Did an infusion help people with respiratory failure?
A trial randomly assigned 102 adults with severe COVID-19 respiratory failure to one of two amounts of an intravenous bone marrow MSC-derived EV preparation called ExoFlo, or placebo. At 60 days the higher-dose group had fewer deaths than the placebo group. The difference fell short of statistical significance on the primary analysis. A survival finding in a subgroup identified after the fact deserves further study. Investigators reported no adverse events attributed to the treatment in this trial. [5]
The larger phase 3 EXTINGUISH ARDS trial is testing ExoFlo for acute respiratory distress syndrome. Its registry lists it as recruiting, with a target of 970 participants and no posted results as of November 9, 2026. A separate study of inhaled MSC-derived exosomes lists 18 actual participants and completed status. Its registry has no posted results either. For someone considering an infusion, the route and the actual outcomes both matter. [6] [7]
Why put a cancer drug inside a vesicle?
In a pancreatic cancer study, researchers loaded MSC-derived exosomes with a short piece of RNA designed to reduce the activity of a specific cancer mutation, KRAS G12D. They gave the preparation intravenously to people with advanced pancreatic cancer carrying that mutation. In the small phase 1 study, no dose-limiting toxicity was seen at the tested doses. Some participants had a period of stable disease. The study had no comparison group to establish a survival benefit. [8]
The trial registry also describes research combining the engineered vesicles with an immunotherapy drug. The published phase 1 human findings above concern the vesicles given alone. The paper’s combination-treatment experiments were in mice. [8] [9]
What does the brain research show?
In mice that had experienced prolonged seizures, vesicles from human bone marrow MSCs reached the hippocampus after delivery through the nose. Compared with mice given the vehicle alone, treated mice had less inflammation and performed better on memory tests. The disease model and route both matter when interpreting that result. [10]
A separate open-label study gave an intranasal preparation from fat-derived MSCs to nine people with Alzheimer’s disease. The group as a whole had no significant change on cognitive tests at 12 weeks. Three people at the middle dose had encouraging changes on some scores. With three people per dose and no placebo group, the study cannot establish that the spray slowed the disease. [11]
What should a provider be able to tell you?
Ask which cells produced the vesicles, how they were separated from other material and whether anything was added to them. A count of particles tells you how many the instrument detected. A functional test asks what the preparation does under specified conditions. Neither test tells you whether patients with your condition felt or functioned better. The international MISEV research guidelines explain how investigators document and test these preparations. [1]
If you see the term “Muse exosomes,” ask for a study of vesicles from that specific source. Clinical studies in which patients received living Muse cells evaluated a different intervention. The evidence to look for is a human study that matches the proposed product, route and reason for treatment.
In the United States, FDA’s December 2019 public safety notice reported serious adverse events associated with unapproved products marketed as exosomes. An August 2026 FDA letter also identified unapproved products sold for regenerative uses. Ask the clinical team to identify the relevant authorization, trial oversight and plan for reporting problems in the country where treatment would take place. [12] [13]
Product details
- Which cells produced the vesicles?
- How were the contents, activity and storage conditions checked?
Patient outcomes
- What did patients receive in the comparison group?
- Which symptoms or functions improved, and for how long?
Follow-up
- What side effects occurred with this route?
- Who will assess and report a suspected reaction?
Questions people ask
Does a higher particle count mean a better treatment?
A count measures particles under the conditions of a particular test. It cannot tell you what they contain, how active they are or whether a higher amount improves a patient outcome. Ask how the preparation was identified and what clinical comparison supports the proposed use.
Can results from exosome eye drops apply to an infusion?
The Sjögren’s study tested drops from a particular cell source applied directly to the eye. An infusion has a different route, exposure and safety question. It needs its own evidence for the condition being treated.
Is ExoFlo already proven to reduce deaths from lung injury?
The 102-person COVID-19 trial did not establish a significant survival benefit on its main comparison. A larger phase 3 trial is listed as recruiting. Its registry has no posted results as of November 9, 2026.
Are exosome treatments FDA approved?
FDA’s December 2019 public safety notification reported that no exosome products had received U.S. approval. Its August 2026 warning letter identified specific unapproved offerings. For a proposed U.S. treatment, ask which FDA approval or investigational pathway applies to that exact product and use.
The evidence
Sources checked November 9, 2026. Study types are listed with each source. Visit the Science library for more research.
This article provides general education. Any treatment decision should be made with a qualified physician. Individual results vary.