Most Semax explainers you'll find online say the same thing: it's an ACTH(4-10) fragment that raises BDNF. That mechanism story has been recycled so many times it now reads like settled fact. But the two most interesting Semax papers published in the last twelve months barely mention BDNF at all — one maps the peptide onto the μ-opioid receptor in spinal cord injury, the other tests it in an Alzheimer's mouse model. Both are rodent-only. Neither has been replicated. This is a look at what current semax research is actually chasing, what it does and doesn't show, and where the compound sits legally for Australian readers.

What is Semax, briefly

Semax is a synthetic heptapeptide derived from the ACTH(4-10) fragment, with a Pro-Gly-Pro tail added to slow enzymatic breakdown. It's registered in Russia as a nasal drop formulation and appears on Russia's Vital and Essential Drugs list — a genuine national registration, not a research-chemical footnote. Outside Russia it has no marketing approval anywhere, and there are no internationally registered clinical trials for it. Our full profile sits at /peptides/semax.

The human clinical base is almost entirely Russian-language stroke work, largely from a small network of institutions, with sample sizes in the tens to low hundreds and limited independent replication. That's the honest baseline. Everything below is newer, and all of it is preclinical.

The 2025 spinal cord paper: a target nobody was talking about

In July 2025, British Journal of Pharmacology published work by Liu and colleagues titled, plainly enough, "Semax peptide targets the μ opioid receptor gene Oprm1 to promote deubiquitination and functional recovery after spinal cord injury in female mice."

The reported chain runs: Semax → μ-opioid receptor → the ubiquitin-specific protease USP18 → deubiquitination of FTO (the fat-mass-and-obesity-associated protein) → less lysosomal-membrane-permeabilisation-driven pyroptosis and oxidative stress → better functional recovery after spinal cord injury. The authors used RNA-seq, network pharmacology and molecular docking to nominate the target, then knocked down USP18 to test whether the effect disappeared.

Three things are worth flagging before anyone gets excited:

It's a mouse study, in one sex

The title says female mice. Sex-specific preclinical work is good practice, not a flaw — but it means the result has not been shown in male animals, let alone humans. Spinal cord injury outcomes in rodents are also notoriously sensitive to injury model, timing and scoring method.

The mechanism is new, not confirmed

Network pharmacology plus docking is a hypothesis-generating approach. A single paper proposing μ-opioid receptor engagement for a peptide that has been studied for decades as a melanocortin-adjacent neuromodulator is interesting precisely because it's unexpected — and unexpected findings are the ones most in need of independent replication.

It has nothing to do with "nootropic" use

This is an acute injury model. It says nothing about focus, memory or mood in healthy people — the uses Semax is most often marketed for.

The Alzheimer's model result: first functional data in an AD mouse

The second paper, published in Acta Naturae (Vol. 17, No. 4, 2025) by Radchenko and colleagues, tested Semax and a derivative in APPswe/PS1dE9 transgenic mice — a standard amyloid model. Animals received 50 μg/kg intranasally every other day from six months of age, for 15 doses over one month.

Reported outcomes: improved performance on open field, novel object recognition and Barnes maze testing, plus histology showing fewer amyloid inclusions in cortex and hippocampus for both the parent peptide and the derivative.

That's a more complete preclinical package than Semax has previously had in Alzheimer's-adjacent work, which leaned on in vitro copper-chelation data. It is still a transgenic mouse model — a model class with a long and well-documented history of producing positive results that fail to translate into human dementia trials. A behavioural improvement in APP/PS1 mice is a reason to run more animal work, not a reason to draw conclusions about human cognition.

So where did BDNF go?

Nowhere, exactly. The BDNF/TrkB story for Semax rests on real rodent data — elevated BDNF mRNA in hippocampus and cortex after administration, with downstream MAPK/ERK and PI3K/Akt signalling described in preclinical work. The problem is how that gets used in consumer copy: "raises BDNF" becomes shorthand for "improves cognition," which is a leap across two species and several orders of evidential rigour.

The 2025 papers are a useful corrective. They suggest Semax's pharmacology may be broader and messier than a single neurotrophin pathway — and that the field itself doesn't treat the BDNF mechanism as the whole answer.

Is Semax legal in Australia? The 2026 position

Semax is not scheduled in the Poisons Standard — it isn't individually named. That is where a lot of online commentary stops, and where it goes wrong. Not being scheduled does not mean lawful to import or supply.

Semax is an unapproved therapeutic good: it is not on the Australian Register of Therapeutic Goods, has never been assessed by the TGA for safety, quality or efficacy, and importing, supplying, compounding or advertising it outside a TGA-recognised pathway (such as the Special Access Scheme or Authorised Prescriber arrangements) is unlawful. The TGA has issued explicit warnings about unapproved peptide products and has publicly said it is prioritising compliance action — infringement notices, seizures, import interventions and, where warranted, civil or criminal penalties.

"Research use only" labelling doesn't change that analysis. Neither does a Russian registration.

For athletes: Semax is not named on the WADA 2026 Prohibited List, but anti-doping specialists describe its status as unclear — a non-approved substance argument under S0 is not something to rule out. Athletes under an anti-doping code should treat unlisted as unsettled, not safe.

A registry warning worth repeating

If you go looking for Semax trials, be careful what you count. ClinicalTrials.gov now hosts fictional demonstration records for several peptides — often listed as RECRUITING with plausible-looking sponsor names that aren't real organisations. Before treating any registry entry as evidence, check that the sponsor is a verifiable institution with a real research footprint. As of now, there are no internationally registered Semax trials.

Browse the rest of our compound profiles at /peptides, or join the free updates list for new evidence reviews as they land.

FAQ

What is Semax used for?

Semax is registered in Russia as nasal drops and has been studied there mainly in ischaemic stroke and cognitive/neuroprotective contexts. Outside Russia it has no approval, and its reputation as a nootropic rests largely on preclinical rodent data rather than controlled human trials.

Is Semax legal in Australia?

Semax is unscheduled in the Poisons Standard but is an unapproved therapeutic good not on the ARTG, so importing, supplying or advertising it outside TGA pathways such as the Special Access Scheme is unlawful. The TGA has publicly warned about unapproved peptide products and is pursuing compliance action.

Does Semax increase BDNF in humans?

The BDNF and TrkB signalling evidence for Semax comes from rodent studies measuring brain-region mRNA and protein — not from human trials. Claims that Semax raises BDNF in people, or that any such change improves cognition, go beyond the published evidence.

Are there any registered Semax clinical trials?

There are no internationally registered Semax trials. Some peptide entries on ClinicalTrials.gov are fictional demonstration records, so always verify that a listed sponsor is a real, identifiable organisation before treating a registry entry as evidence.

Sources

This article is information only, not medical advice. Semax is an unapproved therapeutic good in Australia and is not on the ARTG. retatrutide.net.au is independent and does not sell or source peptides.