Dihexa

Origin and structure

Dihexa comes out of work on angiotensin IV that Joseph Harding and Jay Wright at Washington State University had been pursuing since 1992. Angiotensin IV is a six-amino-acid peptide from the blood-pressure hormone system that improved learning in rodent models of dementia. Its analogues were of little use as drugs: they could cross neither the gut wall nor the blood-brain barrier, and the group's lead analogue, Nle1-angiotensin IV, was broken down in rat serum within about two minutes. Building on the finding that this analogue's activity sat in its first three amino acids, the group presented a stabilised derivative in a paper published online in October 2012 (McCoy et al., J Pharmacol Exp Ther 2013) and called it dihexa: N-hexanoic-Tyr-Ile-(6) aminohexanoic amide.

Strictly speaking dihexa is a peptidomimetic rather than a peptide: two amino acids from angiotensin IV, tyrosine and isoleucine, capped at one end with a hexanoic acid group and at the other with 6-aminohexanoic amide. Its formula is C27H44N4O5 and its molar mass 504.7 g/mol (CAS 1401708-83-5); the FDA's substance register also lists it under the names PNB-0408 and ATH-1001. McCoy et al. reported that it worked in rats when given by mouth and that it crossed the blood-brain barrier; in three rats it had a half-life of about 12.7 days after an intravenous injection. That it would be absorbed from the gut was predicted with a computer model, not measured.

Wright and Harding were founders and shareholders of M3 Biotechnology, which was developing drugs based on this work. The company, incorporated in March 2011, was renamed Athira Pharma in 2019 and LeonaBio in January 2026. Its lead candidate, fosgonimeton (ATH-1017), is dihexa with a phosphate group attached; the FDA's substance register lists it as a prodrug whose active form is dihexa.

How it is thought to work

The Washington State group proposed that dihexa works through hepatocyte growth factor (HGF) and its receptor MET, also called c-Met: a growth-factor system that in the brain supports the survival of nerve cells and the formation of new connections. In cultured rat hippocampal neurons, McCoy et al. reported that dihexa at picomolar concentrations nearly tripled the number of dendritic spines, the small protrusions that carry synapses.

The direct evidence for the HGF/MET link came from a 2014 paper by the same group (Benoist et al.), which reported that dihexa binds HGF, helps it activate MET at concentrations too low to work alone, and loses its effect on memory in rats when HGF is blocked in the brain. The journal retracted that paper in April 2025, so the proposed mechanism no longer rests on a valid publication.

An independent team at China Pharmaceutical University linked dihexa's effects in Alzheimer-model mice to the PI3K/AKT signalling pathway: blocking PI3K with wortmannin reversed its anti-inflammatory and anti-apoptotic effects. How dihexa itself acts in the human brain has not been studied.

Is dihexa approved?

No. No medicine containing dihexa is authorised in the EU or the US, and research material sold under the name is not a medicine. Its prodrug fosgonimeton was not approved either.

• EU: neither the EMA's list of authorised medicines nor its Article 57 database of medicines authorised in the member states contains dihexa (checked 28 October 2026).
• Netherlands: the Dutch medicines register (CBG Geneesmiddeleninformatiebank) has no product with dihexa, so it may not be sold, advertised or supplied for use in people.
• US: not FDA-approved. Dihexa acetate was nominated for use by compounding pharmacies, and the FDA placed it in Category 2, its group of bulk substances that may present significant safety risks. The nomination was later withdrawn; the FDA's page, last updated in April 2026, lists dihexa acetate among withdrawn nominations and states that the agency has found no human exposure data for any route and lacks the information to know whether it would harm people.
• US, next step: the FDA has announced that its Pharmacy Compounding Advisory Committee will discuss dihexa acetate for the 503A bulk substances list, together with LL-37, GHK-Cu, melanotan II and PEG-MGF, at a meeting before the end of February 2027. By 28 October 2026 no date and no Federal Register notice had been published. The committee's vote is advice only, and a compounded product would still not be an approved drug.

What the research shows

Almost everything known about dihexa comes from rodents and cell cultures, most of it from the group that developed it, whose leaders had a commercial stake in it. The only human data come from trials of its injectable prodrug fosgonimeton. The main findings:

• Rats, McCoy et al. 2013 (Washington State University; under an expression of concern since 2021): dihexa given by mouth, by injection or directly into the brain reversed scopolamine-induced memory deficits in the water maze, and given by mouth it improved the performance of 24-month-old rats (6 per group).
• Mechanism, Benoist et al. 2014 (same group): the HGF/MET paper described above, retracted in April 2025.
• Alzheimer-model mice, Sun et al. 2021 (China Pharmaceutical University, independent): APP/PS1 mice (12 per group) treated for three months, from six to nine months of age, learned the water maze better than untreated mice, had more neurons and more of the synaptic protein synaptophysin, and lower levels of the inflammatory signals IL-1β and TNF-α.
• Fosgonimeton, phase 2 (ACT-AD): in people with Alzheimer's disease the trial missed its primary and all secondary endpoints; the developer announced the results in June 2022.
• Fosgonimeton, phase 2/3 (LIFT-AD, Porsteinsson et al. 2025): 549 people with mild to moderate Alzheimer's disease received daily injections of fosgonimeton or placebo for 26 weeks. In the main analysis of 287 participants not taking cholinesterase inhibitors, the combined score for cognition and daily function did not differ from placebo (difference −0.08, p = 0.70), and neither did cognition alone (ADAS-Cog11, p = 0.35) or daily activities (ADCS-ADL23, p = 0.61). After the results in September 2024 the developer paused further development.
• Dihexa itself has never been tested in people: the FDA says it has found no human exposure data for dihexa acetate by any route.

Retracted papers and the BDNF claim

In 2021 the Journal of Pharmacology and Experimental Therapeutics placed an expression of concern on McCoy et al. 2013, the paper that introduced dihexa. In April 2025 it retracted two other papers by the same group: Benoist et al. 2014 on the HGF/MET mechanism, and Kawas et al. 2012 on related angiotensin IV analogues as modifiers of HGF and MET.

The retraction notice for the 2014 paper states that an investigation by Washington State University found falsified and/or fabricated data in Figures 1B and 2A/C and in a later erratum submission, and that Leen Kawas and Joseph Harding were solely responsible. The 2025 annual report of LeonaBio, formerly Athira Pharma, states that the university announced a review of claims of potential research misconduct in the doctoral research of the company's former chief executive. McCoy et al. 2013 has not been retracted, but the expression of concern remains attached to it.

The claim that dihexa is 'ten million times more potent than BDNF', repeated on many websites, comes from Washington State University's press release of October 2012. It said that in bench assays with living nerve cells dihexa was seven orders of magnitude more powerful than BDNF at forming new connections, so that it would take ten million times as much BDNF for the same effect. McCoy et al. 2013, the paper the release accompanied, contains no such comparison, and a potency figure from cultured cells says nothing about effects on memory in people.

Safety and the cancer question

There are no human safety data on dihexa itself. The best indirect evidence comes from LIFT-AD, in which fosgonimeton was injected under the skin every day for 26 weeks. Injection-site reactions affected 57% of participants on the lower dose and 73% on the higher dose, against 14% on placebo, and eosinophilia, a rise in one type of white blood cell, was reported in about 7% on fosgonimeton and in no one on placebo. Side effects led 11% and 22% to leave the trial, against 5% on placebo. Serious adverse events were not more frequent than on placebo (4.9% and 2.8% against 6.9%), and no one died, but the investigators judged five serious events to be possibly drug-related, two of them angioedema (swelling of the skin and mucous membranes).

The larger unknown is cancer. MET is a known cancer driver: the EMA has authorised the MET inhibitor Tepmetko (tepotinib) for advanced lung cancer with MET exon 14 skipping alterations. Whether a compound meant to boost HGF/MET signalling raises the risk of cancer with longer use has never been studied; LIFT-AD lasted 26 weeks, and no long-term or carcinogenicity study of dihexa has been published.

Frequently asked questions

Is dihexa legal in the Netherlands?

Dihexa is not a registered medicine in the Netherlands or anywhere else in the EU, so it may not be sold, advertised or supplied for use in people. Trading it as research material is a different status, which our guide on research-use-only products in the Netherlands explains. For athletes it is prohibited however it was obtained.

Is dihexa banned in sport?

Yes, at all times. Dihexa is not named on the WADA Prohibited List, but class S0 bans any pharmacological substance with no approval for human therapeutic use that no other class covers, and dihexa is approved nowhere. Class S2.3 bans hepatocyte growth factor (HGF) and other growth factors or growth-factor modulators that affect, among other things, regenerative capacity. Whichever class applies, it is prohibited in and out of competition; the Dutch Dopingautoriteit uses the same list.

Is dihexa really ten million times more potent than BDNF?

That figure comes from a 2012 university press release about experiments with nerve cells in a dish, not from a peer-reviewed comparison; the paper the release accompanied contains no such comparison. Potency in a cell assay does not show that a substance improves memory, and the later paper on dihexa's mechanism has been retracted.

Has dihexa been tested in humans?

Not as such. The FDA says it has found no human exposure data for dihexa acetate by any route. The closest evidence comes from fosgonimeton, an injectable prodrug that the body converts into dihexa: in the LIFT-AD trial in Alzheimer's disease it did not improve cognition or daily functioning over 26 weeks, an earlier phase 2 trial also missed its endpoints, and its development has been paused.

Why were dihexa studies retracted?

In April 2025 the Journal of Pharmacology and Experimental Therapeutics retracted two papers from the group that developed dihexa, including the 2014 paper on how it works. According to the notice for the 2014 paper, an investigation by Washington State University found falsified and/or fabricated data, for which Leen Kawas and Joseph Harding were solely responsible. The 2013 paper that introduced dihexa carries an expression of concern but has not been retracted.

Can dihexa cause cancer?

Nobody knows, because it has never been tested for that. The concern is theoretical but serious: dihexa is meant to boost signalling through MET, a receptor that drives some cancers and is the target of approved lung-cancer drugs such as tepotinib. The fosgonimeton trials lasted months, too short to answer the question.

Storage and handling

Store according to product-label instructions and batch handling notes. Lyophilized research peptides are commonly kept dry, protected from light, and handled with clean laboratory technique.

References

Checked against these sources on 28 October 2026.

Related peptides