Exosomes that aren't exosomes: what FDA found in 2025–2026

On 14 August 2026, FDA sent a warning letter to R3 Medical Companies of Scottsdale, Arizona, parent company of a manufacturer (Bello Bio), a private-label distributor (Regen Suppliers) and, according to the letter, R3 Anti-Aging clinics in Scottsdale, Nashville and Beverly Hills¹. Two of its umbilical cord-derived products, ReBellaXO and ReBellaXOL, were sold as exosomes. The ReBellaXO page offered them for "procedures involving IV's, joints, soft tissue, hair, facial or sexual rejuvenation"¹. In a footnote, the agency wrote that the manufacturing process did not appear to isolate exosomes¹.

In Exosomes versus peptides we already covered what an extracellular vesicle is and what the meta-analyses show. This article is about something else: what inspectors found, what the latest warning letters say, and what has been published on complications.

The August 2026 letter: "does not appear" to isolate exosomes

FDA inspected the Bello and Regen facilities between 17 November and 16 December 2025¹. Bello made three umbilical cord products, ReBellaWJ (Wharton's jelly), ReBellaXO and ReBellaXOL, which Regen distributed under its label¹. Footnote 3 states that Regen markets ReBellaXO and ReBellaXOL as exosome products, but that they are made with the same starting material and the same manufacturing process, apart from an added lyophilization step for ReBellaXOL, a process that, according to the agency, does not appear to isolate exosomes¹. FDA made the same observation about two of the manufacturer's own lines, BelloXO and BelloXOL:

The observation matches the field's main technical problem. In its MISEV2023 guidelines, ISEV lists separating vesicles from non-vesicular extracellular particles, along with characterization, among the hurdles to bringing extracellular vesicles to the clinic⁷. A product that, according to FDA, shares material and process with a Wharton's jelly derivative and is still labeled "exosomes" shows that the label does not guarantee what the vial contains¹.

Sterility: findings any physician understands

The second half of the letter lists current good manufacturing practice (CGMP) violations for products that, according to FDA, "purport to be sterile and are expected to be sterile"¹. Among them:

• Unvalidated aseptic processes for the umbilical cord products¹.
• Unvalidated manufacturing process with respect to identity, strength, quality and purity¹.
• Samples frozen and thawed before sterility testing. FDA explains that freezing can destroy or compromise any microbial content present, so contamination present beforehand may not be reliably detected¹.
• Unvalidated cleaning and disinfection of the biological safety cabinet where the product is exposed during aseptic processing¹.
• A 24-month expiry assigned without supporting stability data¹.
• Quality-unit procedures not in writing, including those for approving or rejecting batches and handling complaints¹.

The letter adds two further concerns: the donor eligibility criteria submitted lacked qualifiers for transmissible spongiform encephalopathies such as Creutzfeldt-Jakob disease, and there did not appear to be evidence that testing of a process component had been done before lots were released¹. FDA notes that it had already warned the chief executive in a May 2019 letter, and it judged the company's responses inadequate, even though the company committed to temporarily suspend manufacturing, in part because they do not address distributed products still within expiry¹.

Not an isolated case: four letters in twelve months

August 2026 was not the first time. On 15 August 2025, after reviewing its website, FDA told Platinum Biologics (Orlando, Florida) that its exosome products NanoEx and Nano Xsomes are drugs and biological products subject to premarket review and approval⁴. On 11 February 2026, in its letter to Dynamic Stem Cell Therapy (Henderson, Nevada), the agency added that exosomes intended to treat disease "and/or are intended to affect the structure or function of the body" are subject to drug requirements². On 26 May 2026, the letter to Blue Horizon International (Hackensack, New Jersey) cited the claim that its Wharton's jelly mesenchymal stem cell exosomes "demonstrate strong regenerative and anti-inflammatory properties", and concluded that it had neither a biologics license (BLA) nor an investigational application (IND) in effect³.

For aesthetics, the Dynamic letter's wording is what matters: the test is not limited to claims of curing disease but includes affecting the structure or function of the body, which is what rejuvenation promises². Three of the four letters point to FDA's 2019 safety notice, issued "following multiple reports of serious adverse events"²³⁴. The underlying position has not changed: "There are currently no FDA-approved exosome products"⁵⁶.

What the case reports describe

In 2024, Thai dermatologists described a middle-aged man who received intradermal injections of lyophilized exosomes for enlarged pores. By day three he had painful, non-blanchable purpuric papules and nodules; leftover product injected into his arm produced similar lesions, and biopsy showed necrotic keratinocytes, leukocytoclastic vasculitis and eccrine necrosis⁸. He improved on 7 days of oral prednisolone and was left with post-inflammatory hyperpigmentation⁸. In 2025, a case from Riyadh described a 38-year-old man with ischemic necrosis on both cheeks three days after intradermal injections of Korean-manufactured exosomes for acne scars; he was left with atrophic scars and hyperpigmentation, and three laser sessions gave only modest improvement⁹.

A 2025 case series from Seoul gathered four women with persistent erythema, nodules, granulomatous inflammation and scarring after intradermal injection of a cosmetic exosome-containing formulation given outside medical supervision, in non-dermatology clinics or beauty salons; none resolved completely¹⁰. The author explains that cosmetics are formulated for topical use, where the stratum corneum acts as a barrier, and that once placed in the dermis their excipients, preservatives and active compounds can trigger reactions¹⁰. A scoping review of 17 clinical studies found adverse events were uncommon but included granulomas, necrosis and allergic reactions after injection¹¹.

None of these cases involves the products named in the FDA letters, and case reports cannot estimate frequency or prove causation⁸⁹¹⁰.

And microneedling?

It is the most common route in practice and the least studied. A systematic review published in April 2026 started from 256 references and ended with 8 studies, each with 3 to 60 participants and 171 patients in total, in androgenetic alopecia, skin aging, hyperpigmentation, scarring and enlarged pores¹². Among studies using exosomes from external suppliers, none disclosed the supplier's regulatory status or ethical governance, and the authors conclude that more evidence is needed before the safety and efficacy profile can be established¹².

The Korean origin of much of the market adds another layer. The Seoul series notes that these products are approved for topical use and that intradermal injection remains off-label and unregulated in many countries, including South Korea¹⁰. A 2026 regulatory review summarizes that exosomes are classified as cosmetics in Korea when applied topically and as biologics in the US when injected, and that Korean cosmetic safety rules prohibit live human cells, tissues or their inadequately purified derivatives as ingredients¹³. A topical cosmetic vial, therefore, does not come with an assessment for the intradermal route¹⁰¹³.

What to ask before opening the vial

• What exactly is it? Source (human, plant, culture medium), separation method and characterization against ISEV criteria⁷.
• How was sterility shown? Whether the aseptic process is validated and whether samples are frozen before testing¹.
• What supports the expiry date? Stability data, not just a number on the label¹.
• Which route is it registered for, and where? Topical cosmetic does not equal intradermal¹⁰¹³; the microneedling review found no such information in any study with an external supplier¹².
• If it is human-derived, how donor eligibility was assessed¹.