CellREDM · Re2O and Rejuran — what is different
We are often asked “which is better, CellREDM or Rejuran?” To be honest first: no clinical trial comparing the two under the same conditions exists. So we cannot answer with “this one is better”. There is something we can answer instead — the two families differ in their raw material, in their regulatory route, and in the kind of evidence that has accumulated so far. In particular, one of them had to pass a clinical trial in humans before it could be marketed, and the other had no legal obligation to do so. This piece sets out that difference.
The conclusion, first
The biggest difference is not the ingredient but the regulatory route. Rejuran (PN) is a medical device, so to be approved it had to submit data from a clinical trial in humans, and it did in fact pass a phase 3 trial in 72 people. CellREDM · Re2O (hADM) are human tissue, so they fall under the Act on the Safety and Management of Human Tissue, and had no legal obligation to submit a clinical trial in humans before marketing. A clinical paper on Re2O does in fact exist, but it was carried out after launch (November 2024), it is of 20 people · 20 weeks, and its comparator is not no treatment but hyaluronic acid alone (the test arm being hADM + HA). And looking at the United States makes this formulation's position clearer still — FDA guidance recognises decellularised dermis in sheet form as ‘minimal manipulation’, but states expressly that grinding the dermis into particles goes beyond minimal manipulation. That is, in the United States this formulation cannot be distributed as plain human tissue. Conversely, traceability is tighter on the human tissue side — the record is managed from donor through to distribution. It is an asymmetry in which verification of efficacy is tighter on the PN side and tracing of the record is tighter on the hADM side. This does not mean one is better; it means different things are guaranteed.
What the raw material is
| Item | hADM (CellREDM · Re2O) | PN (Rejuran) |
|---|---|---|
| Source | Dermal matrix from donated human skin with the epidermis stripped and the cells removed | DNA extracted · purified from salmon · trout testes |
| Nature of the substance | A tissue-derived mixed matrix of collagen · elastin · glycoproteins and the like | A single kind of macromolecule |
| The unit it is specified in | Particle size (µm) and mg per vial | Molecular weight (kDa) and concentration (mg/mL) |
| Final formulation | Freeze-dried powder injected after mixing with normal saline or non-cross-linked hyaluronic acid | An aqueous injectable; Rejuran is polynucleotide 20 mg/mL |
Let us first dispose of one question that comes up often — “which one has the larger molecular weight?” The question does not hold. PN is a single macromolecule and so is specified in kDa, but hADM is a piece of tissue made of several substances mixed together, so the value called molecular weight is not defined for it at all. The figure that means something for hADM is how finely it was ground (µm).
Are PN and PDRN different substances
They are not different substances but chain-length bands of the same substance. According to a 2025 summary in Biomolecules, they are divided into PDRN at 50–1,500 kDa and PN above 1,500 kDa, with recent formulations reported up to 8,000 kDa. However, the 1,500 kDa boundary itself was only proposed in 2025, so the literature before that uses the two terms interchangeably. We were not able to confirm a published molecular weight figure for Rejuran itself from a primary source.
The manufacturing process for PN includes a step called controlled depurination. Its purpose is “to remove the capacity to transmit genetic information”. It then goes through sterilisation at 121°C. The reason salmon testes are used is explained as being that they are a good source of high-purity DNA with few impurities such as peptides · lipids.
What level of evidence is the mechanism — neither has been confirmed in humans
| Family | The usual explanation | The ceiling of the evidence |
|---|---|---|
| PN | The broken-down nucleotides stimulate the adenosine A2A receptor on fibroblasts and are recycled through the salvage pathway, accelerating matrix synthesis | In vitro · animal. Even a 2026 expert review sponsored by the manufacturer states expressly that “robust histological verification in human skin is lacking” |
| hADM | The remaining matrix becomes a scaffold that fibroblasts move into and make new collagen in | In vitro · human skin explant (ex vivo) · one-week rat histology. There is no histological proof that hADM injected into a human actually remains as a scaffold and makes new collagen |
The level of the mechanistic evidence for the two families is effectively the same. Neither has confirmed its mechanism by biopsy of human skin. That said, the preclinical data for each is fairly detailed.
On the hADM side, the preclinical work (2026, International Journal of Molecular Sciences) observed a concentration-dependent increase in proliferation and increased synthesis of collagen I · III · elastin in cultured human dermal fibroblasts; after injection into human skin explants, a homogeneous distribution through the dermis was reported at 24 hours, and in UV-irradiated skin recovery of basement membrane components (nidogen I · collagen IV). In rats, fibroblasts were observed migrating into the grafted matrix and integrating with it one week after injection.
There is a point here that must be read carefully. One often sees the explanation that “hADM works because it contains growth factors”. But what the paper above actually showed is that hADM increased the ‘production’ of VEGF · FGF · PDGF by cells, not that growth factors are ‘contained’ within hADM. We did not find a primary source quantifying residual growth factor content. The two sentences are different claims.
Clinical trials in humans — what has accumulated, and how much
| Item | hADM (Re2O) | PN (Rejuran) |
|---|---|---|
| Principal trial | Int J Mol Sci 2026;27(5):2193, Severance | J Korean Med Sci 2014, Seoul National University Bundang Hospital — the phase 3 for approval |
| Design | Randomised · split-face · double-blind | Phase 3 · randomised · double-blind · split-face · active-controlled |
| Sample | 20 people (mean age 54.7) | 72 enrolled / 70 analysed |
| Comparator | Hyaluronic acid alone (the test arm being hADM + HA) | A non-cross-linked high-concentration HA product |
| Follow-up | 20 weeks | 12 weeks after the final session |
| Primary endpoint | None designated separately; a number of instrument measurements (density · volume · wrinkles · pores · pigmentation · hydration · transepidermal water loss) | Improvement rate based on clinical photographs, non-inferiority design (margin −15%) |
| Result | Superior to the comparator in skin density · volume · wrinkles · pores · hydration (p < 0.05). The comparator arm showed “no meaningful change” | 95.7% (67/70) versus 94.3% (66/70), a difference of 1.4%p — non-inferiority met |
| Adverse events | No serious adverse events. Transient erythema · mild swelling | 31.9% (23/72) versus 48.6% (35/72), p = 0.042 — significantly fewer on the Rejuran side |
What to look at alongside each trial
- The primary endpoint of the Rejuran phase 3 is the ‘improvement rate on photographic reading’, not a measurement of skin elasticity or hydration. And the comparator is not a placebo but a hyaluronic acid product. What this trial proved goes as far as “it is not worse than an HA product”.
- The test arm in the Re2O trial is not hADM alone but hADM + HA, with the comparator arm HA alone. That is, it is a design that looked at “adding hADM to HA is better than HA alone”. We confirmed this fact not from the body of the paper but from a copy of the trial registry record (NCT07155278), and we did not reach the registry's original entry directly.
- The Re2O trial began after the product was launched (November 2024 – May 2025), and its registration was retrospective (registered September 2025). We were not able to confirm the funding source and conflict of interest declaration.
The overall level of evidence on the PN side
A 2025 systematic review assembled 9 studies · 219 people in total. The authors' conclusion is reserved — promising for wrinkles · skin texture · elasticity, but the included studies were “of low and moderate quality only”, the heterogeneity of injection site · technique was large so that “consensus on optimal use is limited”, and they wrote that “rigorous high-quality studies are essential to verify efficacy and safety”.
That is, the ‘quantity’ of evidence is clearly greater on the PN side (9 studies, 219 people versus 1 study, 20 people), and the ‘quality’ of the evidence is still low on both sides. That is the accurate state of things at this point.
Is there a study comparing them directly
There is not. We did not find a randomised clinical trial comparing the hADM family and the PN family in the same people under the same conditions. In both families the comparator is either hyaluronic acid or a collagen filler. The “Re2O versus Rejuran” comparison tables circulating on the internet are all clinic promotional content and not clinical data.
The regulatory route — the most substantial difference in this piece
| Item | hADM (CellREDM · Re2O) | PN (Rejuran) |
|---|---|---|
| Legal classification | Human tissue | Medical device (tissue-repair biomaterial) |
| Governing statute | The Act on the Safety and Management of Human Tissue | The Medical Devices Act |
| Requirement for market entry | A tissue bank licence | Marketing approval — submission of clinical trial data is mandatory |
| Pre-market clinical trial | No legal obligation | Mandatory |
| Adverse event reporting | The tissue bank reports once a year; serious adverse reactions within 7 days | The adverse event reporting system under the Medical Devices Act |
| Traceability | The record is traced from donor → processing → testing → storage → distribution | Manufacturing · distribution records |
In an April 2026 news report an MFDS official explained, of products made by powdering human acellular dermal matrix, that they “fall within human tissue” and that “unlike a medical device, no separate clinical trial is required”. That is the legal fact.
How the United States views the same formulation
This is the firmest piece of evidence in this article. Two examples sit side by side in the original text of the FDA's 2020 guidance “Minimal Manipulation and Homologous Use”.
| Method of processing | The FDA's determination |
|---|---|
| Sheet — the connective tissue left after removal of the epidermis, freeze-dried · packaged | Recognised as minimal manipulation (may be distributed as human tissue) |
| Particles — after removal of the epidermis, the dermis is ground into particles | “More than minimally manipulated” — because the original characteristics of skin relating to its utility as a protective covering have been altered |
That is, in the United States a powdered hADM injectable such as CellREDM · Re2O cannot be distributed as plain human tissue. Once it falls outside the minimal manipulation requirement it is regulated as a drug · medical device · biological product and pre-market approval becomes necessary. On top of that, one of the requirements in the same regulation prohibits “combination with another article”, and a formulation mixing hADM with hyaluronic acid runs into that condition as well.
What this difference actually means for a patient
- Pre-market verification — Rejuran had to pass a trial in humans in order to be sold. The hADM products had no such obligation, and the clinical paper on Re2O came out after launch.
- Tracing the record — conversely, the human tissue framework traces from donor information through to distribution. On this point the requirement is in fact stronger than for a medical device.
- Raw material — at the October 2025 parliamentary audit it was pointed out that more than 90% of human tissue is imported. Re2O's manufacturer stated to the press that it imports 100% of its raw material from the United States and that cosmetic purposes are included and disclosed in the donor consent form.
Reform of the framework is under discussion in Korea. It has been reported that the MFDS is pursuing an amendment to the rules that would tighten adverse event reporting to twice a year and would include restrictions on advertising for cosmetic purposes and disclosure to the patient that a product is of human origin. We were not able to confirm whether this amendment has come into force as of September 2026. In a survey of 1,034 adults presented at a National Assembly forum in April 2026, 69.8% refused the use of cadaver-derived tissue in cosmetic procedures, and 72.9% called for mandatory labelling of components of human origin.
Adverse events — what has been reported and what does not yet exist
| Family | What was reported in clinical trials |
|---|---|
| PN | Local adverse events 31.9% in the phase 3 (lower than 48.6% in the HA comparator arm, p = 0.042). The systematic review writes only that they were “generally mild and transient” and gives no quantified incidence. The manufacturer-sponsored review writes “no reported cases of granuloma or vascular occlusion in the literature” |
| hADM | 0 serious adverse events in the 20-person trial. For reference, in a 202-person trial of another company's product carried out in China, early swelling · pain were significantly more frequent than in the comparator (collagen filler) (p < 0.05) and resolved within a week |
There is something that must be distinguished here. We did not find a single case report paper on adverse effects of injectable hADM. But this does not mean “it is safe”. The product came out after November 2024, and in the human tissue framework adverse events are tallied once a year, so the published data itself does not yet exist. The granuloma · nodule cases that come up when you search are all about hyaluronic acid fillers.
The theoretical risks that come from being of human origin are set out in the statute as well — the items the Human Tissue Act defines as serious adverse reactions are transmissible disease, transfer of malignant tumour and infection from contamination of the tissue. To these are added immune · hypersensitivity reactions from residual DNA and residual detergent as the items experts point to. In a 2026 comparison of decellularisation processes, the residual DNA of three protocols was 4.95 – 11.53 ng/mg, all satisfying the commonly used criterion (below 50 ng/mg). However, the same study also showed that the higher the detergent concentration, the lower the residual collagen IV · elastin · laminin and the more the collagen fibre thickness changed.
A remark by one physician in private practice, given to the press, summarises the present situation most accurately — “rather than there being no problem with safety, it is closer to a stage of not yet knowing.”
What is different between CellREDM and Re2O
We are also often asked whether there is a difference between them, given that they are both hADM. This is a table filled in only with published information.
| Item | CellREDM | Re2O |
|---|---|---|
| Manufacture · distribution | Hans Biomed · Hugel | L&C Bio · Humedix |
| Launch | September 2025 | November 2024 |
| Decellularisation process | Not disclosed | Only the name disclosed — details not disclosed |
| Detergent · concentration · processing time | Not disclosed | Not disclosed |
| Residual DNA · residual detergent | Not disclosed | Not disclosed |
| Particle size | 75 µm or less (manufacturer press release) | Not disclosed |
| Its own clinical paper | We did not find one | Int J Mol Sci 2026 — exists |
Material on which the performance difference between the two products could be judged is effectively not published. What can be compared publicly is only these four things: the manufacturer · the date of launch · CellREDM's particle specification · and the fact that only Re2O has a paper of its own. That the decellularisation method does in fact make a difference to what remains is confirmed, but because which method each of the two products uses is not published, we can say neither that “the components differ” nor that “they are the same”. Assertions of the kind commonly seen in clinic content — “A is more refined”, “B's particles are finer” — have no basis.
Figures of 150 mg for Re2O and 160 mg for CellREDM appear on distributor pages and in personal posts, but we were not able to confirm either figure from an official manufacturer document. So we have not put them in the table.
Is there evidence for the number of sessions and the interval
| Protocol | Evidence |
|---|---|
| PN — 3 sessions at 2-week intervals | This is the protocol the clinical trials actually used (both the phase 3 and the periocular trial). But what those trials verified is “this protocol is non-inferior to HA”, not “3 sessions are better than 2 or 4” |
| PN — 3 sessions at 4-week intervals | This is the protocol put forward by a 2026 expert review, and that same paper states expressly that it is “based on the authors' clinical experience”. It is expert opinion, not trial evidence |
| hADM — 3 or more sessions at 3–4-week intervals | We did not find clinical evidence. The sources are only distributor · clinic pages, and they are inconsistent enough that “3 sessions at 3–4 weeks” and “2–3 sessions at 1–3 months” appear together on the same page |
| hADM — “3 sessions last a year” | The longest follow-up in humans is 20 weeks (about 4.6 months). One-year data does not exist |
We would add that neither family has a dose-response study verifying “how many sessions are optimal”. The session counts in current use are the values adopted by the early studies, hardened into practice.
In summary — what is confirmed, what we could not confirm
| Confirmed | The regulatory routes differ, hADM being human tissue and PN a medical device · hADM had no legal obligation of a pre-market clinical trial (confirmed by an MFDS official) · FDA guidance states expressly that grinding the dermis into particles goes beyond minimal manipulation · the Rejuran phase 3, 70 analysed, 95.7% versus 94.3%, non-inferiority met · the Re2O trial, 20 people · 20 weeks, the test arm being hADM + HA · the PN systematic review, 9 studies, 219 people, quality ‘low to moderate’ · what remains does in fact differ with the decellularisation process · more than 90% of human tissue is imported |
|---|---|
| Could not confirm | A study comparing the two families directly (it does not exist) · the Re2O paper's funding source and conflict of interest declaration, and the number of sessions · interval · dose · the actual decellularisation method · residual DNA · residual detergent · official content of CellREDM · Re2O · whether CellREDM has a clinical study of its own · a published molecular weight for Rejuran's PN · the class · intended use in the original text of Rejuran's marketing approval · whether Europe actually restricts the use of human tissue for cosmetic purposes · whether the amendment tightening domestic adverse event reporting has come into force · the actual tally of adverse events for hADM boosters in Korea |
| Data pointing the other way | In the Rejuran phase 3, adverse events were more frequent in the comparator arm (HA) · in the 202-person Chinese trial, early swelling · pain were more frequent with the hADM-family product than with the comparator · traceability is tighter on the human tissue side — the regulatory route does not only mean “less verified” |
This is what we actually say in the consulting room. The two families are not competitors but options about which different kinds of things are known. If you want “the one that has been verified more”, at present there is more material on the PN side. If you choose the hADM side, it is right to choose knowing that it is a substance of human origin and that a pre-market clinical trial was not a legal requirement. Either way, a sentence such as “it lasts a year” is not supported by the material that exists today.
Frequently asked questions
Which works better, CellREDM or Rejuran?
We cannot answer that. Because no clinical trial comparing the two families in the same people under the same conditions exists. The trials of each product had either hyaluronic acid or a collagen filler as comparator. The comparison tables circulating on the internet are clinic promotional content and not clinical data.
What are CellREDM · Re2O made from?
They are the dermal matrix (collagen · elastin and so on) that remains after the epidermis is stripped from donated human skin and the cells are removed. This is finely ground and freeze-dried into a powder, which is injected after mixing with normal saline or hyaluronic acid. Rejuran is a polynucleotide derived from DNA extracted from salmon · trout testes — different from the source onwards.
Does it matter that the material is of human origin?
The statute already sets out that risk — the Human Tissue Act defines transmissible disease, transfer of malignant tumour, and infection from contamination of the tissue as serious adverse reactions. To prevent these, the record is traced and managed from donor through to distribution. However, how many actual occurrences have been reported is not confirmable from published material — the product came out only recently and adverse events are tallied once a year.
Why were the hADM products able to come out without a clinical trial?
Because their legal classification is human tissue. Human tissue falls under the Act on the Safety and Management of Human Tissue and is distributed under a tissue bank licence, with no obligation to submit pre-market clinical trial data as a medical device would. In an April 2026 report an MFDS official also explained that “unlike a medical device, no separate clinical trial is required”. A clinical paper on Re2O does in fact exist, but it was carried out after launch.
Is the same product used in the United States?
It cannot be distributed in the same way. FDA guidance recognises decellularised dermis in sheet form as “minimal manipulation”, but states expressly that grinding the dermis into particles goes beyond minimal manipulation. Once outside minimal manipulation it is regulated not as human tissue but as a drug · medical device, and pre-market approval is required. A formulation mixed with hyaluronic acid also runs into the “combination with another article” requirement.
Is Rejuran the better-verified product?
There is more material — 9 studies and 219 people by the count of the systematic review, and it passed a phase 3 for approval (70 analysed). But that review's conclusion is reserved. It wrote that the included studies were “of low and moderate quality only” and that “consensus on optimal use is limited”. There is more of it, and the quality is still low on both sides.
Are CellREDM and Re2O different from each other?
The material on which that could be judged is not published. Neither product discloses its decellularisation method · the type and concentration of detergent · residual DNA figures. What can be compared publicly is only the manufacturer, the date of launch, CellREDM's particle specification (75 µm or less), and the fact that only Re2O has a clinical paper of its own. Assertions of the kind “this one's particles are finer” have no basis.
I heard that 3 sessions last a year.
There is no material supporting that number. The longest follow-up in humans in the hADM family is 20 weeks, about 4.6 months. Because no trial observed the one-year point, “maintained for a year” is not something measured but an estimate. On the PN side too, the follow-up in the principal trial is 12 weeks after the final session.
How many weeks apart should the sessions be?
It differs by family. For Rejuran, 3 sessions at 2-week intervals is the protocol the clinical trials actually used. The commonly quoted 4-week interval is the one put forward by a 2026 expert review, and that paper itself states that it is “based on the authors' clinical experience”. On the hADM side we did not find clinical evidence, and even the distributor materials differ from one another.
So how should I choose?
We say it this way. If you want “the one with more material behind it”, at present that is the PN side. If you choose the hADM side, it is right to choose knowing that it is a substance of human origin and that a pre-market clinical trial was not a legal requirement. And either way, please take as the yardstick for your expectations that the period the material now in existence has observed is about four months at most, and about three on the other side.
Who wrote this
Written and reviewed by Lee Chi-Hak, MD, medical director of Miso Clinic in Daegu, South Korea. Every study cited above is given together with its design, its size and the limitations the authors themselves recorded, and where we could not find data, we have said that we could not find any.
| Medical director | Lee Chi-Hak, MD |
|---|---|
| Address | 4F Bombom Building, 125 Dongdeok-ro, Jung-gu, Daegu, South Korea · Exit 1, Kyungpook National University Hospital Station |
| Phone | +82-53-428-2700 |
| Hours | Weekdays 11:00–19:00 (lunch 13:00–14:00) / Saturday 10:00–16:00 (no lunch break) / Closed Sundays and public holidays |
| Columns | All clinical columns |
| Reference library | All booster and device references |
References
- hADM clinical · preclinical — Lee YI, Chau NH, Nguyen NH, Ham S, Baek Y, Kim J, Lee JH. Injectable Particulated Human Acellular Dermal Matrix Booster for Skin Restoration: An Integrated Randomized, Split-Face, Double-Blinded Clinical Trial and Preclinical Study. International Journal of Molecular Sciences 2026;27(5):2193. Randomised · split-face · double-blind, 20 people (mean age 54.7), November 2024–May 2025, 20 weeks of follow-up. Superior to the comparator in skin density · volume · wrinkles · pores · hydration · transepidermal water loss (p < 0.05), with no significant change in the comparator arm. 0 serious adverse events. Preclinical — concentration-dependent proliferation and increased collagen I · III · elastin in human dermal fibroblasts, homogeneous distribution and recovery of basement membrane components at 24 hours in human skin explants, migration of fibroblasts into the matrix at one week in rats.
- The design of that trial — important — trial registration number NCT07155278. Test arm hADM + hyaluronic acid, comparator arm hyaluronic acid alone. Sponsor Yonsei University, registered September 2025 (the trial having started in November 2024, this is retrospective registration). We did not reach the registry's original entry directly and confirmed it from a copy of the record. We were not able to confirm the funding source and conflict of interest declaration, nor the number of sessions · interval · dose.
- The PN phase 3 for approval — Pak CS, Heo CY et al. Journal of Korean Medical Science 2014;29(Suppl 3):S201–S209 (Seoul National University Bundang Hospital). Phase 3 · randomised · double-blind · matched · active-controlled · split-face. 72 enrolled / 70 analysed. Test article PRM-001 (polynucleotide 20 mg/mL) versus a non-cross-linked high-concentration HA. 3 sessions at 2-week intervals, 12 weeks of follow-up after the final session. Primary endpoint — photograph-based improvement rate at week 12, non-inferiority margin −15%. Result 95.7% (67/70) versus 94.3% (66/70), a difference of 1.4%p (the original paper in places states this difference as 1.5%). Adverse events 31.9% versus 48.6%, p = 0.042.
- The PN periocular trial — Lee YJ et al. Journal of Dermatological Treatment 2022;33(1):254–260. Randomised · double-blind · split-face, 27 people, PN versus non-cross-linked HA, 3 sessions at 2-week intervals. No significant between-group difference in VAS or GAIS; the improvement rate was higher on the PN side for elasticity · hydration · roughness · pore volume, and there was no significant difference in dermal density.
- The PN systematic review — Lampridou S, Bassett S, Cavallini M, Christopoulos G. Journal of Cosmetic Dermatology 2025;24(2). 9 studies · 219 people in total included. The authors' conclusion — the quality of the included studies was “low and moderate only”, the heterogeneity of injection site · technique is large and “consensus on optimal use is limited”, “rigorous high-quality studies are essential”.
- The definition and process of PN · PDRN — Marques et al. Biomolecules 2025;15(1):148. PDRN 50–1,500 kDa, PN above 1,500 kDa (this boundary being the one proposed by this paper). The raw material is rainbow trout · chum salmon. Process — protein digestion → multi-stage separation → controlled depurination → autoclaving at 121°C. Mechanism — the broken-down deoxyribonucleotides bind the fibroblast adenosine A2A receptor and are recycled through the salvage pathway.
- The limits of the evidence for the PN mechanism — a 2026 expert review in Clinical, Cosmetic and Investigational Dermatology (sponsored by PharmaResearch). It places the mechanism at the level of “primarily preclinical · mechanistic studies and clinical observation” and states expressly that “robust histological verification in human skin is lacking”. The 3-sessions-at-4-week-intervals protocol the same paper puts forward is stated by the paper itself to be “based on the authors' clinical experience”.
- The effect of the decellularisation process on what remains — Mbele et al. Frontiers in Bioengineering and Biotechnology 2026;14:1839855. A comparison of three protocols for decellularising human dermis. Residual DNA A 9.25 ± 0.90 / B 4.95 ± 4.71 / C 11.53 ± 1.95 ng/mg — all satisfying the commonly used criterion of below 50 ng/mg. However, the higher the detergent concentration, the lower the residual collagen IV · elastin · laminin and the more the collagen fibre thickness changed (A 0.1388 ± 0.0363 µm versus C 0.4275 ± 0.1142 µm, p < 0.05). This paper does not quantify residual growth factor content.
- US regulation — the core evidence of this piece — FDA Guidance, Regulatory Considerations for Human Cells, Tissues, and Cellular and Tissue-Based Products: Minimal Manipulation and Homologous Use, July 2020. Example 11-3 — decellularised dermis consisting of the connective tissue left after removal of the epidermis, freeze-dried · packaged, is recognised as minimal manipulation. Example 10-4(b) — “grinding the dermis into particles … generally is considered more than minimally manipulated because the processing alters the original relevant characteristics of skin related to its utility as a protective covering.” If even one of the four requirements of 21 CFR 1271.10(a) is unmet, the product is regulated not as a 361 HCT/P but as a drug · medical device · biological product, and one of those requirements prohibits “combination with another article”.
- Korean regulation — the Act on the Safety and Management of Human Tissue and its Enforcement Rule. In an April 2026 Edaily report, an MFDS official — a product made by powdering human acellular dermal matrix “falls within human tissue”, and “unlike a medical device, no separate clinical trial is required”. A tissue bank reports adverse events once a year, with serious adverse reactions (transmissible disease · transfer of malignant tumour · infection from contamination of the tissue) within 7 days. It has been reported that the MFDS is pursuing an amendment that would tighten ordinary adverse event reporting to twice a year and add restrictions on advertising for cosmetic purposes · a duty to disclose human origin — we were not able to confirm whether it has come into force as of September 2026.
- The state of the domestic debate — the October 2025 parliamentary audit — more than 90% of human tissue is imported (Rep. Nam In-soon), “no regulation, no clinical trials and no oversight” (Rep. Lee Su-jin). The April 2026 National Assembly K-Bio Health Forum — in a survey of 1,034 adults, 66.8% consented to donation for therapeutic purposes only, 69.8% refused the use of cadaver-derived tissue in cosmetic procedures, and 72.9% called for mandatory labelling of components of human origin.
- A large trial of another company's hADM-family product (for reference) — Zhang et al. Aesthetic Plastic Surgery 2026;50(10). PMID 41381953. This is a Chinese product and not a Korean one. Double-blind · multicentre · randomised · non-inferiority, 202 allocated / 175 completed, with a cross-linked collagen filler as comparator. 88.4% versus 85.4% at 3 months, 70.9% versus 69.7% at 6 months — non-inferiority met. Early swelling · pain were significantly more frequent on the hADM side (p < 0.05) and resolved within a week.
- Product information — CellREDM (manufactured by Hans Biomed · distributed by Hugel, launched September 2025, particles of 75 µm or less per the manufacturer's press release), Re2O (manufactured by L&C Bio · distributed by Humedix, launched November 2024, the name of its own decellularisation technology disclosed). Neither product discloses the type · concentration · processing time of the detergent, nor residual DNA and residual detergent figures. The content figures (Re2O 150 mg, CellREDM 160 mg) were confirmed only on distributor pages and in personal posts, so they have not been put in the table in the body of this piece.
- What we were not able to confirm — (1) a study comparing the two families directly (it does not exist) (2) the Re2O paper's funding source · conflicts of interest · injection protocol (3) whether the body of the paper names the test product by brand (it is written only as “phADM” in the text) (4) the class · intended use · date of approval in the original text of Rejuran's marketing approval (5) a published molecular weight figure for Rejuran's PN (6) whether CellREDM has a clinical study of its own (7) whether the European SoHO Regulation (EU 2024/1938, applying from August 2027) actually restricts use for cosmetic purposes — there is only secondary reporting and we did not reach the original text of the provisions (8) the actual tally of adverse events for hADM boosters in Korea.
- The regulatory content of this piece is as of 4 September 2026, and because the relevant framework is under discussion for amendment it may differ depending on the date.
Everything in this column is general information and does not replace medical diagnosis or treatment. Effects and side effects vary with individual skin condition, age and underlying illness, and the same result is not guaranteed for everyone. Any decision to proceed should be made in an in-person consultation with a physician.
← Clinical columns · Reference library · Miso Clinic home
한국어 · English · 日本語 · 简体中文 · Español · Tiếng Việt · ภาษาไทย · Bahasa Indonesia