PDRN in creams: how far does the research go, and where does marketing begin?

PDRN originated as an injectable preparation. For about a year now, we have been finding it in creams.

In between, until the end of 2025, there was no published clinical evidence on the topical form.

I am not saying that PDRN does not work. For the injectable, the data exist and are good. I am saying that those data concern something different from what you are being sold, and that the distance between the two is where science ends and storytelling begins.

[INDICE:h2]

What is PDRN

PDRN stands for polydeoxyribonucleotide: a polymer made of DNA fragments, generally extracted from salmon, which has become “salmon DNA” on social media. The proposed mechanism is activation of A2A adenosine purinergic receptors, which may stimulate fibroblasts, collagen synthesis, and the formation of new blood vessels.

It is a serious molecule, not a marketing invention. Some of the literature describes it as a registered drug for tissue repair, and it has been studied in wound healing, tendinopathies, periodontal regeneration, cartilage, and corneal epithelium. These are clinical settings, with clinical administration.

Where the data are solid: injectable use

Here, we need to be honest in the opposite direction. For the injectable, evidence exists and is of good quality.

A randomized split-face study—that is, with one side of the face treated and the other serving as a control, which is the cleanest design possible because it eliminates differences between people—compared an injectable polynucleotide with a hyaluronic acid filler in 27 participants. The side treated with the polynucleotide showed significantly better scores on the Global Aesthetic Improvement Scale at 16 weeks, and a greater reduction in skin roughness at week 28.

These are the data that give PDRN clinical legitimacy. They are also the data cited in serum presentations.

Where the data are thin: topical use

Until the end of 2025, there had been no published clinical study on a topical form of PDRN. Not a few, not conflicting ones: none.

The first is from late 2025 and concerns low-molecular-weight PDRN from peony: it showed improved periocular elasticity after four weeks of application. There is also a randomized, double-blind, split-face study involving 33 women, using an eye cream versus placebo for twenty-eight days.

Two studies, a few dozen participants, the periocular area, four weeks. They are a beginning, and as a beginning they deserve respect. But they are not the foundation on which the way PDRN is discussed today rests.

“An active ingredient that works when injected does not necessarily work when applied topically. It is the first question I ask, and in three cases out of four I get no answer.”

The problem no one talks about: how large is this molecule?

This is the technical point that closes the argument, and you can find it stated in the studies themselves, not in the criticisms.

A study published in July 2026 on a topical preparation of medium-length PDRN describes the material as follows: fragments of approximately 1,200 base pairs, with an average molecular weight of up to 850 kilodaltons. It states, verbatim, that because PDRN is a high-molecular-weight nucleic acid polymer, whether it can reach the viable layers of the skin after topical application remains an open translational question.

Let’s put this into perspective. The most frequently cited rule of thumb for skin penetration places the threshold at around 500 Daltons: above that value, crossing the stratum corneum becomes progressively more difficult. Eight hundred and fifty thousand versus five hundred is a difference of three orders of magnitude.

The stratum corneum is not an imperfect filter that lets a little of everything through. It is a structure with the precise job of letting nothing through, and it is very good at doing so. Injecting means bypassing it entirely; applying something topically means asking it for permission.

It is not impossible to solve: low-molecular-weight fragments, delivery systems, and specially designed formulations exist. In fact, the positive-outcome study used a low-molecular-weight form. But then the question becomes a different one—and the right one: which PDRN is in that cream, what is its molecular weight, and at what concentration? An INCI that says only “sodium DNA” does not answer that.

How I decide whether an active ingredient belongs in a product line

It is not an ideological stance against new developments. It is a method, and it is always the same, for PDRN as for anything else.

First question: do the studies concern the active ingredient or the finished product? They're two different things. An active ingredient can have excellent literature behind it and do nothing inside the wrong formula.

Second: is the route of administration used in the studies the same as mine? Oral, injectable, and topical aren't interchangeable. It's the same distinction that applies to collagen and half of beauty supplements.

Third: at what concentration? An active ingredient present only in trace amounts serves to put a name on the label, not to do anything for the skin.

Fourth: what happens when I put it into a formula? Stability, pH, compatibility with the rest, behavior over time. Supplier documents aren't enough here: you try it, test it, and see what the prototype does after several weeks.

Most of the active ingredients I evaluate fail on this fourth question. That's why the LeLang catalog consists of only a few products: not because there aren't interesting ingredients, but because the distance between interesting and usable is enormous.

What I would do if I were on the other side

If you're thinking of buying a PDRN serum, I'm not telling you not to. I'm telling you to look at three things.

Look at what it promises. If the text talks about regeneration, collagen stimulation, and cites studies, ask—or look up—whether those studies were conducted on injectable preparations. Almost always, they were.

Look at the molecular weight. If the manufacturer specifies that it uses a low-molecular-weight form, that's a good sign: it means they've considered the issue. If they don't mention it, they probably haven't.

Look at the price in relation to your expectations. An expensive cosmetic product with the expectations of a medical treatment is the best way to end up disappointed by both.

And if you're looking for that kind of result, the honest answer is that you need to look where the data exist: with a doctor. A cosmetic product improves the appearance of the skin—that's already a lot, but it's something else.

Why this story keeps repeating

PDRN isn't the first, and it won't be the last. The same thing happened with exosomes, and something else will take its place in six months. The pattern is identical every time: a molecule with genuine clinical data in medicine, a transition into cosmetics that no one spells out, and a narrative that inherits the credibility without inheriting the evidence.

What saddens me is that this mechanism burns through good molecules. In three years, when there may perhaps be truly effective topical formulations with solid data, PDRN will already have become « that thing everyone was talking about », and no one will believe in it anymore.

To help navigate molecules in general, I wrote a guide to cosmetic ingredients, and there is a dedicated guide to peptides — the opposite example, namely actives developed and studied for topical use from the outset.