Dihexa Research Guide: Mechanism, Studies and Retractions

The claim behind dihexa's reputation as a nerve-growth-factor peptide comes from one research program at Washington State University. A 2014 paper reported that dihexa binds directly to hepatocyte growth factor (HGF) and that both dihexa and its parent compound, Nle1-angiotensin IV, trigger phosphorylation of HGF's receptor, c-Met, on hippocampal neurons — activating a growth-factor pathway the same paper linked to new dendritic spine formation and synapse growth, effects it reported were blocked by HGF-pathway antagonists in the same experiments.^[3]

That paper no longer stands. The Journal of Pharmacology and Experimental Therapeutics issued a formal Expression of Concern about it in September 2021 and retracted it in April 2025. The retraction notice states that a Washington State University investigation found falsified and/or fabricated data in several of the paper's figures.^[3][4][5]

It was not the only casualty. Two earlier papers from the same laboratory and research program, describing related angiotensin-IV-derived compounds designed to block — rather than activate — HGF's receptor as potential anticancer agents, received the same Expression of Concern in 2021 and were retracted in the same April 2025 action.^[6][7][8][9]

One part of the story stands apart from any of this. The classical binding site through which angiotensin IV and its analogues act, generally called the AT4 receptor, was independently identified by a separate group of researchers as insulin-regulated aminopeptidase (IRAP), an enzyme unconnected to HGF or c-Met. That identification does not rely on, and does not appear in, any of the retracted papers above.^[11]

The best-known behavioural data on dihexa comes from the same 2013 paper that named the compound. Researchers reported that oral dihexa reversed scopolamine-induced deficits in Morris water maze performance in rats, a standard model of drug-induced memory impairment, and separately improved water-maze learning in aged, 24-month-old rats, alongside increased hippocampal synaptogenesis in the same animals.^[1]

That paper has not been retracted, but it has not been cleared either. It has carried an unresolved Expression of Concern since September 2021, issued alongside the one on the mechanism paper that was later retracted. Peptyds treats its specific figures as unverified pending resolution.^[2]

A separate, more recent rodent study is unconnected to the Washington State University laboratory or its retractions. Researchers at China Pharmaceutical University and Nanjing Medical University tested dihexa in APP/PS1 mice, a transgenic model bred to develop Alzheimer's-like amyloid pathology. Four groups of twelve mice each — wild-type, untreated APP/PS1, and APP/PS1 given dihexa at 1.44 or 2.88 mg/kg — were treated by daily stomach gavage for three months. Dihexa-treated APP/PS1 mice had a shorter escape latency in the Morris water maze than untreated APP/PS1 mice and crossed the former platform location more often, alongside reduced inflammatory markers and increased synaptic protein expression; blocking the PI3K/AKT signalling pathway reversed these effects.^[10]

As of September 2026, a ClinicalTrials.gov search for 'dihexa' returns no registered study, and no completed or ongoing human study of the compound appears in PubMed either.^[15]

A related but chemically distinct compound has reached human testing. Fosgonimeton (also called ATH-1017), described in its own published methods paper as a positive modulator of HGF/MET, is not dihexa — it is a differently structured molecule from later work in the same hepatocyte-growth-factor research area. That phase 1 safety and pharmacology paper's acknowledgements thank Dr. Leen Kawas, a co-author of the retracted dihexa mechanism papers described above, for her contribution to the trial's design.^[13]

A larger, completed phase 2/3 study of fosgonimeton — 315 participants with mild-to-moderate Alzheimer's disease, daily subcutaneous dosing against placebo over 26 weeks — has posted results on ClinicalTrials.gov. Neither the study's primary endpoint, a combined cognition-and-function score, nor its key secondary endpoints reached statistical significance against placebo.^[14]

HGF and its receptor, c-Met, are not obscure to cancer biology — they are one of its established subjects. Independent research unconnected to dihexa describes MET as a receptor that drives tissue growth, survival and migration, links overactive HGF/MET signalling to tumour invasion and metastasis across multiple cancer types, and treats inhibiting MET, not activating it, as the anticancer strategy worth pursuing.^[12]

Dihexa's now-retracted mechanism paper proposed the opposite direction: activating, not blocking, this same receptor pathway to promote nerve growth. The same Washington State University laboratory separately published, and in the same 2025 action had retracted, work exploring hepatocyte growth factor mimics as Met-blocking anticancer agents — evidence that this research group itself treated the pathway as cancer-relevant, just from the opposite end.^[6]

None of the rodent studies described in this article screened for or reported tumours, and none ran longer than three months. That is an absence of monitoring over a short window, not a demonstrated safety result, and it leaves the underlying question — what sustained activation of this receptor pathway would do over a longer period — open rather than answered.^[1][10]

Dihexa is not approved as a medicine anywhere. No dihexa-based product holds a marketing authorisation from the European Medicines Agency, and none appears in the FDA's Drugs@FDA database of approved drugs.^[17][18]

Its regulatory position is, however, actively moving. Dihexa acetate is one of five substances FDA is considering for its section 503A bulks list, the list of bulk drug substances that licensed compounding pharmacies may use, alongside cathelicidin (LL-37), GHK-Cu, melanotan II and pegylated mechano growth factor. FDA says its Pharmacy Compounding Advisory Committee will discuss these substances at a meeting before the end of February 2027; no decision had been made as of September 2026.^[16]

Separately, FDA's page on bulk drug substances that may present significant safety risks in compounding (current as of 22 April 2026) lists dihexa acetate among substances previously in category 2 whose nominations were withdrawn by the nominators. Its entry notes that FDA has not identified any human exposure data on drug products containing dihexa acetate, and lacks important information on whether it would cause harm if given to people.^[19]

A 503A listing, if the FDA eventually grants one, would let licensed compounding pharmacies prepare dihexa-based preparations to order for an individual, named patient. It would not be an FDA approval of dihexa as a standalone medicine, and a compounded preparation is not reviewed for safety and effectiveness the way an approved drug is — the two are different regulatory categories, not steps on the same ladder.^[16]

Peptyds does not sell dihexa, in any form or purity. This guide exists to summarise the published research and its recent retractions for readers trying to make sense of what is circulating online about it, not to introduce a product.

Nothing in this guide is an offer, a recommendation, or instructions for personal use. The evidence for dihexa, as it stands in September 2026, is a body of rodent research carrying an unresolved integrity problem at its foundation, no published account of the compound being tested in a person, and an open biological question about the pathway it is proposed to activate.