Exosomes

Summary

Exosomes are not a molecule: they are nanometer-scale extracellular vesicles, separated from the medium in which cells were cultured — with no sequence to cite and no purity to check against a certificate of analysis, in the same category of problem as Thymalin. The field's own consensus, MISEV2023, still lists nomenclature and the separation of non-vesicular particles among the open obstacles. The survey finds an asymmetry with no parallel on this site: 45,674 articles on PubMed and 83 randomized trials, less than 0.2% of the total. Of the 378 registrations on ClinicalTrials.gov, 142 are observational — they measure exosomes as a marker, they don't treat anyone with them — and only 9 have posted results. I opened the nine phase 3 registrations: one is a PET/CT study in prostate cancer that entered the search because it cites “exosome analysis”, seven of the other eight have 135 participants or fewer, and the only large one is EXTINGUISH ARDS, with 970 participants, still recruiting through 2027. The phase 2 of the same product, published in Chest, did not meet the primary endpoint: what came out significant was a post hoc subgroup. At the FDA, checked on September 12, 2026, there is no approved exosome — and the warning letters keep coming, the last one I verified in February 2026.

What it is — and the name is already the first problem

Exosomes are not a molecule. They are a nanometer-scale extracellular vesicle, released by cells — in the marketed preparations, almost always mesenchymal stem cells from umbilical cord, placenta or adipose tissue, or dermal papilla cells. What is bought is not a defined substance: it's the material separated from the medium in which those cells were cultured.

This puts exosomes in the same category of problem as Thymalin, which is bovine thymus extract: there is no sequence to cite, no purity to check against a certificate of analysis, and two vials with the same name may not be the same thing. The difference is that Thymalin doesn't pretend to be a molecule, and “exosome” sounds like it is one.

The difficulty is neither mine nor the sellers': it belongs to the field itself. The consensus document from the International Society for Extracellular Vesicles — MISEV2023, written with contributions from more than a thousand researchers — lists, among the field's still-open obstacles, nomenclature challenges and the separation of vesicles from non-vesicular extracellular particles. In plain English: the field's own scientific society says that naming and purifying remain unsolved problems.

This page did not come from the secondary source that most of this site is built on. It was built from scratch, using PubMed, ClinicalTrials.gov, the FDA's label database, and open data.

How much evidence really exists

The numbers below were compiled by me on September 12, 2026. Each row's query is next to the number, for anyone to repeat and contradict me.

Each query is written out in full, not as “the same one above plus a filter”: that way each row reproduces on its own, in a single paste, and the script that rechecks this site's counts can run all of them. Abbreviating the query is the flaw this reference already made once and fixed with a safeguard.

And at both agencies

The two regulatory databases are left out of the table above because they are not literature counts, and the script that rechecks this site's counts only knows how to repeat PubMed and ClinicalTrials.gov. I ran both by hand on September 12, 2026.

In the FDA label database, openfda.generic_name:"exosome" returns NOT_FOUND — zero labels. A free-text search for exosome in the same database returns 3 labels, and none of them is an approved product; what they are is covered in the FDA section further below.

In the open data on registered medicines, exossom and exosom return zero occurrences. What that absence means — and what it doesn't — is covered in the section about Brazil.

Where the numbers don't add up

• 45,674 articles and 83 randomized trials. The randomized ones are less than 0.2% of the literature; with the broader filter, of any type of clinical trial, it's 168 — less than 0.4%. No other compound on this site has an asymmetry of this magnitude.
• 142 of the 378 registrations are observational, not interventional. Most of them don't give exosomes to anyone: they measure exosomes in the blood as a disease marker. That's a different subject, and it's where much of the word's prestige comes from.
• 9 of the 378 registrations have posted results in the database itself. A registration without a result is not evidence: it's declared intent.
• The size of the literature measures the biology, not the product. Exosomes are a real cellular mechanism, studied long before any clinic existed selling vials of them — the same way cytochrome C has hundreds of biochemistry articles that say nothing about the injectable. A high count here is not a sign of a validated treatment.

The nine “phase 3” ones, one by one

“Nine phase 3 trials” is the kind of number that circulates in sales material. I opened all nine. Four are pure phase 3; five are phase 2/3, the combined label. And one of them isn't exosome therapy at all.

What reading the nine shows

• Without the PET/CT study, eight are left. One false positive in nine is the reason I treat the 378 registrations as a ceiling, not as a count of exosome therapies.
• Seven of the eight have 135 participants or fewer. Twenty, thirty, forty, sixty, ninety, one hundred and thirty-five. “Phase 3” here doesn't mean what it means in a registration program: it means the label the sponsor chose on the form.
• Three of the eight are marked UNKNOWN — the tag the registry applies when the sponsor stops updating. One of them should have finished in 2014.
• Half is cosmetic. Facial skin, alopecia, laser. That's where the market is, and where the trials are smaller.

The one large program, and what it found

ExoFlo, from Direct Biologics, is a bone-marrow mesenchymal cell extracellular vesicle. It's the only exosome in this survey with a serious clinical program, and the published result deserves to be read in full, not just by the headline.

The phase 2 came out in Chest in December 2023 (NCT04493242): 102 patients with moderate to severe COVID-19 ARDS, at five centers in the United States, randomized to placebo, 10 mL or 15 mL of ExoFlo on days 1 and 4. The primary endpoint was all-cause mortality at 60 days.

The primary endpoint was not met. The reduction with 15 mL versus placebo came out non-significant (χ², P = 0.1343). What did come out significant was a post hoc subgroup analysis: relative risk of 0.385 (95% CI 0.159–0.931; P = 0.0340; n = 50) and, among those aged 18 to 65, 0.423 (95% CI 0.173–1.032; P = 0.0588; n = 24). Ventilator-free days improved (P = 0.0455; n = 50). No treatment-related adverse event was reported.

Post hoc subgroup generates a hypothesis; it doesn't confirm effect — that's exactly what phase 3 exists for, and it's exactly what EXTINGUISH ARDS, with 970 participants, is running until the end of 2027. The article is signed by the medical director and by employees of the manufacturer, with the conflict declared, and it received an erratum in 2024.

The other human data with a comparative design that I found is EXO-CD24, an inhaled vesicle: 35 patients treated in 2020 against 105 matched controls, retrospective follow-up of 4.6 years, no deaths in the treated group against 14 (13.3%) in the controls. It's retrospective and matched, not randomized — the design most prone to selection among those that compare groups.

Where the market is: skin, hair, and knee

Almost everything sold as exosome in Brazil is aesthetics. I went to see what the literature supports in these three areas.

Hair. A 2026 systematic review gathered 39 studies published between 2019 and 2025 on exosomes in androgenetic alopecia. The early clinical studies report a gain of 8% to 20% in hair density — and the authors themselves list the limitations: small samples, retrospective design, absence of a control group, and outcomes measured differently in each study. The conclusion calls for large randomized trials, with a standardized outcome and good-manufacturing-practice production, before wide clinical adoption.

Knee. The best quantitative synthesis I found on exosomes in knee osteoarthritis is a 2026 network meta-analysis with 27 randomized trials — in 456 rats. The effect on cartilage histopathological damage is large (SMD −3.78; 95% CI −4.76 to −2.80), and it's still rats. The authors themselves say it's useful for preclinical optimization and clinical hypothesis generation.

Skin. The trials that exist are small and split-face, many comparing exosome added to laser or microneedling against the procedure alone — a design that measures the add-on effect, and in which the base procedure already produces an effect.

The commercial scope of this has been measured. A 2023 survey analyzed 978 companies advertising direct-to-consumer stem cells and exosomes in the United States; less than half made any identifiable claim about the safety and efficacy of what they sold. In the COVID-19 subset, 38 companies operating or brokering 60 clinics, priced from US$ 2,950 to US$ 25,000, with an advertised average of US$ 11,322.

What the FDA says, with a date

• December 6, 2019 — public safety notification. The agency reported multiple reports of serious adverse events in Nebraska patients treated with unapproved products sold as containing exosomes. The reports reached the FDA through the CDC, and the agencies worked with the state health department.
• The central sentence, verbatim: “There are currently no FDA-approved exosome products.” Rechecked by me on the agency's page on September 12, 2026.
• Exosomes used to treat disease are a drug and a biological product, subject to prior review and approval, under the Public Health Service Act and the Federal Food, Drug, and Cosmetic Act. The FDA notes that clinics claim these products escape drug rules, and responds that this “is simply untrue”.
• What the agency tells the patient to do: ask whether the FDA reviewed the treatment, and ask for the IND number — the research authorization — before accepting the treatment.
• The warning letters keep coming. I checked two: Platinum Biologics, of Orlando, on August 15, 2025, over the products “NanoEx” and “Nano Xsomes”, with the sentence “Your products are not the subject of an approved BLA”; and Dynamic Stem Cell Therapy, of Henderson, Nevada, on February 11, 2026, which refers back to the same 2019 notification. I did not count how many exist — I verified these two.
• The catch with the label database. A free-text search for exosome returns 3 labels: a placenta-exosome solution with an anti-aging claim, an ampoule against hair loss, and a wound-healing fluid. None has an approval number, and being listed in the label database is not the same as being approved — the database also accepts unapproved products. Searching by generic name returns nothing.

What exists in Brazil

I downloaded the open data on registered medicines on September 12, 2026 and searched for exossom and exosom: zero occurrences. There is no medicine with a registered exosome in that dataset — and that doesn't close the question, because that dataset doesn't cover Advanced Therapy Products, which have separate registration under RDC 505/2021 — the same trap described in Casgevy. If any exosome was filed as an advanced therapy, that's not where it would show up. I DID NOT VERIFY the list of advanced therapy products: the official page that's supposed to display it doesn't render a single product.

The Brazilian Society of Dermatology addressed the subject by name in a technical note dated September 13, 2024: “such products cannot be used in injectable therapies, regardless of technique, indication, the prescriber's technical competence, or product quality.”

Why there is no dose table on this page

Because there is no unit that cuts across products. An exosome dose is advertised sometimes in particle count, sometimes in micrograms of protein, sometimes in milliliters, sometimes in “vials” — and each laboratory measures by the method it chose, in a field whose own consensus treats separation and characterization as an open problem.

Building a dilution table here would give the appearance of a protocol to something that has no approved dose, no registered formulation, and no comparable batch. It's the same decision made on the Thymalin, bioregulators, and mazdutide pages: where there is no number that means the same thing twice, the table is invention dressed up as data.

What this survey did not do

• I read abstracts and study records, not full articles. The exceptions are the FDA and SBD pages, which I opened and read — and from which the quoted sentences are verbatim.
• I did not consult any Asian or European agency. The cosmetic exosome market is larger in South Korea, Japan and China, and I have nothing verified about MFDS, PMDA, NMPA or EMA. Absence here is not absence there.
• I did not find a peer-reviewed description of the Nebraska outbreak. The only source is the FDA notification, which does not name the clinic, the product, the number of patients, or the microorganism. What is known is what the agency wrote.
• I did not cover exosomes as a diagnostic. The 142 observational registrations are largely liquid biopsy and disease markers — a different subject, and outside this site's scope.
• The ClinicalTrials.gov count is a ceiling, not a measurement. Searching by intervention field captures registrations that merely mention exosomes, and the PET/CT case proves it. Anyone who wants the exact number of exosome therapies has to open all 378, one by one; I opened the nine phase 3 ones.
• There is no reconstitution table on this page, by decision, for the reason given in the section above.

References

1. FDA. Public Safety Notification on Exosome Products, December 6, 2019 — the origin of the sentence “There are currently no FDA-approved exosome products” and of the report of serious adverse events in Nebraska.
2. FDA. Public Safety Alert Due to Marketing of Unapproved Stem Cell and Exosome Products — the consumer-alert version, with guidance to ask for the IND number.
3. FDA. Warning letter to Platinum Biologics LLC, August 15, 2025 — products “NanoEx” and “Nano Xsomes”, no approved BLA.
4. FDA. Warning letter to Dynamic Stem Cell Therapy, February 11, 2026 — refers back to the 2019 notification on exosomes.
5. Lightner AL et al. Bone Marrow Mesenchymal Stem Cell-Derived Extracellular Vesicle Infusion for the Treatment of Respiratory Failure From COVID-19: A Randomized, Placebo-Controlled Dosing Clinical Trial. Chest. 2023;164(6):1444-1453 — ExoFlo, phase 2, NCT04493242, 102 patients, non-significant primary endpoint.
6. EXTINGUISH ARDS registration on ClinicalTrials.gov — phase 3 of ExoFlo, 970 participants, expected completion by the end of 2027.
7. Shapira S et al. Reduced Mortality in COVID-19 Patients Treated With Inhaled Extracellular Vesicles Expressing CD24. J Extracell Vesicles. 2026;15(3):e70253 — 35 treated against 105 matched controls, retrospective follow-up of 4.6 years.
8. Malih S et al. Exosome-driven treatments for hair regrowth in androgenetic alopecia: a systematic review of preclinical studies, clinical experiments, safety, and future prospects. Mol Biol Rep. 2026;53(1):1046 — 39 studies, reported gain of 8% to 20% in hair density, and the design limitations.
9. Zhao Y et al. Exosome therapy for knee osteoarthritis: a network meta-analysis based on rat models. Stem Cells Transl Med. 2026;15(9):szag069 — 27 randomized trials, 456 rats.
10. Turner L et al. Safety and efficacy claims made by US businesses marketing purported stem cell treatments and exosome therapies. Regen Med. 2023;18(10):781-793 — 978 companies analyzed.
11. Turner L et al. Businesses marketing purported stem cell treatments and exosome therapies for COVID-19: An analysis of direct-to-consumer online advertising claims. Stem Cell Reports. 2023;18(11):2010-2015 — 38 companies, 60 clinics, from US$ 2,950 to US$ 25,000.
12. Welsh JA et al. Minimal information for studies of extracellular vesicles (MISEV2023): From basic to advanced approaches. J Extracell Vesicles. 2024;13(2):e12404 — the ISEV consensus, which lists nomenclature and separation of non-vesicular particles among the open obstacles.
13. NCT03824275 registration on ClinicalTrials.gov — the PET/CT study in prostate cancer that turns up in the search for “exosome analysis” in the description, and is not exosome therapy.
14. Sociedade Brasileira de Dermatologia. Nota técnica sobre o uso dermatológico de exossomos, September 13, 2024.
15. FDA's public label API, used to confirm the absence of an approved product and to find the three listed labels.

← Back to all compounds