Last updated 2026-08-28in-progress
Disclosure Audit
What exosome vendors publish about their product, field by field, from public pages only — source, passage, isolation method, particle count and its measurement method, markers, sterility, traceability, storage, and COA availability.
Signal is audited first, on the same fields, from its own public pages.
Conflicts
spray.signal.clinic publishes a per-dose particle count
(5×10⁹ particles/dose), a size range (30–150 nm), a marker panel
(CD9, CD63, CD81), and a filtration step (0.22 µm), and publishes no
source cell type, no isolation method, no measurement method behind the
particle count, no sterility or endotoxin result, no lot or traceability
reference, no storage or shipping condition, and no COA offer. The
0.22 µm figure is a process step, not a sterility test result, and is
recorded in the sterility field as filtration only.
This repo's own cdmo-readiness.md (2026-08-27) records the same
parameters — cell source, isolation method, particle-count method,
release testing — as questions Signal cannot currently answer in full.
The public pages and the internal readiness record are consistent with
each other on those fields: both say the information does not exist
publicly. The published figures above are the exception, and
sgl-001-spec.md (2026-08-27) already records that the bottle
arithmetic behind them does not close. That unresolved point is not
re-litigated here; it is noted so this audit's Signal row is not read as
independent confirmation of those numbers.
The table
| Vendor | Source cell type / line | Passage no. | Isolation method | Particle count per dose (method) | Size distribution | Marker panel | Sterility / endotoxin | Lot no. / traceability | Storage / shipping | COA to buyers | Date checked |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Signal | not disclosed | not disclosed | not disclosed | 5×10⁹ particles/dose; 100 doses/bottle (method not disclosed) | 30–150 nm | CD9, CD63, CD81 | 0.22 µm sterile filtration stated; no sterility or endotoxin result disclosed | not disclosed | not disclosed | not disclosed | 2026-08-28 |
| Kimera Labs | "perinatal mesenchymal stem cell (MSC)-derived"; "single-donor pre-COVID MSC cell bank"; no line identifier | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | 2026-08-28 |
| Direct Biologics | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | 2026-08-28 |
| Vitti Labs | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | product-tracking form linked; no lot format or record disclosed | not disclosed | not disclosed | 2026-08-28 |
| EriVan Bio | "Umbilical Cord Derived"; no line identifier | not disclosed | not disclosed | 1×10⁹/mL (method not disclosed) | not disclosed | not disclosed | not disclosed | not disclosed | "shipped on dry ice"; storage temperature not disclosed | not disclosed | 2026-08-28 |
| Stem Nova Network | not disclosed for the exosome line | not disclosed | not disclosed | 12, 60 and 150 billion per vial, by SKU (NTA, stated per production lot) | not disclosed | not disclosed | sterility, mycoplasma, endotoxin and viral screening stated as covered on the lot COA; no values published | "lot-specific Certificate of Analysis" with every shipment | "overnight on dry ice with cold-chain logging" | yes — lot-specific, stated as in-house plus third-party | 2026-08-28 |
| BioRegenEx | "Umbilical cord-derived MSCs" from "Wharton's jelly"; no line identifier | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | not disclosed | stated as included in a physician-portal documentation package; no COA or sample published | 2026-08-28 |
| Exocel Bio | "placental stem cell–derived"; no line identifier | not disclosed | not disclosed | 5, 12, 25 billion and 5 trillion (skin SKUs); 7, 15, 30 billion (hair SKUs); 75, 100, 450 billion (research SKUs) — method not disclosed | not disclosed | not disclosed | donor "infectious disease panel screened" and "sterile C-section collection" stated; no product sterility or endotoxin result disclosed | not disclosed | not disclosed | not disclosed | 2026-08-28 |
Aggregate, this run
Counts over the eight rows above. These are arithmetic on the table, not a ranking.
| Field | Vendors disclosing | of |
|---|---|---|
| Source cell type | 4 | 8 |
| Cell line identifier | 0 | 8 |
| Passage number | 0 | 8 |
| Isolation method | 0 | 8 |
| Particle count per dose or per vial | 4 | 8 |
| Measurement method behind that count | 1 | 8 |
| Size distribution | 1 | 8 |
| Marker panel | 1 | 8 |
| Sterility and endotoxin (any statement) | 1 | 8 |
| Lot number / traceability | 1 | 8 |
| Storage and shipping | 2 | 8 |
| COA available to buyers | 2 | 8 |
Four vendors publish a number of particles. One publishes how that number was measured.
Sources
One entry per row. Pages fetched 2026-08-28; field values above are taken from these pages only.
- Signal — https://spray.signal.clinic/ and https://spray.signal.clinic/science. Homepage carries "5×10⁹ particles / dose", "100 µL", "10 mL · METERED SPRAY", "0.22 µm". Science page carries "5×10⁹ particles per dose, 100 doses per bottle", "30-150 nm vesicles", "tetraspanins CD9, CD63, CD81". Neither page carries a source cell type, isolation method, storage condition, or COA offer.
- Kimera Labs — https://kimeralabs.com/products/. Carries
"perinatal mesenchymal stem cell (MSC)-derived exosome products",
"fully characterized, single-donor pre-COVID MSC cell bank", and
"used for cosmetic and research purposes". No characterisation values.
(
kimeralabs.com/root returned a connection reset on first attempt; the products page fetched normally.) - Direct Biologics — https://www.directbiologics.com/pipeline
(
/exoflo301-redirects here). Page is pipeline-level: products "are being evaluated in clinical trials conducted under active Investigational New Drug applications" and "All Direct Biologics programs are investigational and have not been approved by the U.S. Food and Drug Administration." No product characterisation data on the public page. - Vitti Labs — https://vittilabs.com/. Company-level page: "an industry leading manufacturer of human cellular and tissue-based products (HCT/Ps)", with AATB, GMP and FDA-registration claims. A "Product Tracking" form is linked. No product specification page was located on the public site this run.
- EriVan Bio — https://www.erivanbio.com/product-page/uc-msc-exosomes-umbilical-cord-derived. Carries "Umbilical Cord Derived", "Exosome Concentration: 1 x 10⁹ / mL", "shipped on dry ice", "$995 per vial (1 mL)".
- Stem Nova Network — https://www.stemnovanetwork.com/pages/exosome-biologics-for-medspas and https://www.stemnovanetwork.com/. Carries "60 billion NTA-verified extracellular vesicles" (also 12B and 150B SKUs), "Nanoparticle Tracking Analysis confirms exosome particle count on every production lot", "Every lot ships with a lot-specific Certificate of Analysis covering particle count, sterility, mycoplasma, endotoxin, and viral screening", "overnight on dry ice with cold-chain logging". Labelled "for topical cosmetic use only — they are not for injection or therapeutic use". The landing page describes "Wharton's jelly sourced" UCT-MSCs as a separate stem cell line and does not state the source cell type of the exosome line.
- BioRegenEx — https://bioregenex.com/what-are-msc-derived-exosomes-physician-guide/. Carries "Umbilical cord-derived MSCs", "Wharton's jelly", "Supplied for Research Use Only", and "BioRegenEx provides complete documentation in all five areas as part of the standard physician portal package". The page's own list of what a COA should contain (including CD63, CD81, HSP70) describes supplier requirements, not disclosed values for this vendor's product; it is not recorded as a marker-panel disclosure.
- Exocel Bio — https://exocelbio.com/ and https://exocelbio.com/exovex/. Homepage carries per-SKU particle counts ("5 TRILLION", "5 BILLION", "12 BILLION", "25 BILLION" for exovex+ Skin; "7 BILLION", "15 BILLION", "30 BILLION" for exovex+ Hair; "100 BILLION", "75 BILLION", "450 BILLION" for research products), "Ethically sourced placental stem cell–derived exosomes", "Infectious disease panel screened, exceeding FDA requirements / transplant grade", "Sterile C-section collection", "our team produces exosomes under rigorous cGMP and ISO standards", "FOR PROFESSIONAL USE ONLY", and "This product is not intended to diagnose, treat, cure, or prevent any disease". The exovex product page itself carries no particle count, no method, and no COA offer.
Provenance
- First run of this agent. Eight vendors is the starting set, not the category — it was assembled from public search for exosome suppliers selling to clinics and practitioners, plus vendors already named elsewhere in this repo. It is not a census and should not be read as one; the denominator in the aggregate table is eight audited pages.
- Not checked this run: Zeo ScientifiX (formerly Organicell —
organicell.com301-redirects tozeoscientifix.com, which returned HTTP 403 to automated fetch). No row was created rather than filling one withnot disclosedvalues that were never actually verified. Also not reached: Rion/Purexo, Predictive Biotech, Regenerative Labs, and the Korean supplier segment. - Every value above comes from a page a prospective buyer can reach without a login. Nothing gated was fetched, no account was created, and no vendor was contacted. Where a vendor states that a document exists behind a portal or on request (BioRegenEx, Stem Nova), the row records the statement, not the document — the document itself was not seen.
- Field values are recorded from the specific pages listed above. A vendor may disclose more in a product insert, a portal, or on request. This audit measures what is published, which is the field a buyer can check before committing.
- Signal's row was compiled independently from the site this run and not
copied from
sgl-001-spec.md, so the two are separate readings of the same public pages rather than one reading cited twice.
Run history
- 2026-08-28Artifact created; half the vendors publish a particle count and one publishes the method behind it, while passage number, isolation method and cell line are absent across all eight.