Dihexa: what is known, and what is not
Reviewed 24 sourcesPublished by PepVault
Dihexa has no human trial: what the rat and mouse studies found, why three of its lab's papers were retracted, how its prodrug's Alzheimer's trial went.

- Last reviewed
- Sources
- 24
- Questions
- 8
- Compounds
- 1
The short answer
Dihexa has never been tested in a published human trial: everything known comes from rats, mice, zebrafish and cells. In rats it reversed memory loss, and in rat brain cells it grew new synapses. In 2025 its journal retracted three of the developers' papers, including the one behind its growth-factor mechanism, and fosgonimeton, a drug the body turns into dihexa, missed every goal of its 287-person Alzheimer's trial.
Evidence: Animal and cell studies only, most from the developer's lab and three from independent teams; three of the lab's papers, including its HGF/c-Met mechanism paper, were retracted in 2025, and no human trial of dihexa itself has been published.

Has dihexa been tested in people?
Not as dihexa. A search of ClinicalTrials.gov on November 20, 2026 for dihexa and its code name, PNB-0408, returns no registered study1, and PubMed holds no human trial of it. The FDA reached the same place from its side: it has not identified any human exposure data on products containing dihexa acetate, by any route2.
Yet people have carried dihexa in their blood. Athira Pharma, the company that developed fosgonimeton (ATH-1017), says the drug is injected in an inactive form and rapidly converted in the blood to an active metabolite, dihexa417. Its structure shows why: fosgonimeton is dihexa with a phosphate group on its tyrosine3. So the closest thing to human data is fosgonimeton's trials, covered below. They tested a different molecule, injected under the skin under medical supervision, not dihexa as it is sold.

Where did dihexa come from?
Dihexa came out of Joseph Harding's lab at Washington State University, which studied angiotensin IV, a six-amino-acid fragment of the blood-pressure hormone angiotensin with a long record of helping memory in animals. As a drug it had two problems: the body breaks it down fast, and it cannot pass the gut or the blood-brain barrier, the filter between blood and brain. The team started from the three amino acids that carried the memory effect in an angiotensin IV analog and rebuilt both ends. The result worked by mouth, reached the brain and resisted breakdown5.
Its chemical name, N-hexanoic-Tyr-Ile-(6) aminohexanoic amide, shows how little of it is peptide: two amino acids, tyrosine and isoleucine, between two carbon-chain caps5. Kawas, Wright and Harding listed the lab's company, M3 Biotechnology, beside the university on the 2014 mechanism paper14, and Leen Kawas, a graduate student in the lab, later became chief executive of Athira Pharma4.
Peptide SkySee Dihexa as a molecule53 peptides as a night sky in 3D. Each star opens into its molecule, one bead per amino acid, with the measured shape where one exists.Open the Peptide SkyWhat did the animal studies find?
The founding paper, McCoy 2013, reported strong anti-dementia activity in rats in two models: memory loss caused by scopolamine, a drug that blocks a memory signal, and memory loss in old age. Both were measured in the Morris water maze, a pool where rats learn to find a hidden platform. In rat neurons from the hippocampus, the brain's memory center, dihexa also grew new synapses, the contact points between neurons5. The later studies, side by side:
| Study | Animal | What was tested | What it found |
|---|---|---|---|
| McCoy 20135 | Rats | Memory loss from scopolamine and from old age | Deficits reversed in the water maze; marked synapse growth |
| Uribe 20156 | Zebrafish larvae | Hearing-type hair cells exposed to the antibiotics neomycin and gentamicin | Hair cells protected, best at 1 µM |
| Sun 20217 | APP/PS1 mice, bred to model Alzheimer's | Memory and brain inflammation | Water-maze learning restored; inflammatory IL-1β and TNF-α down, calming IL-10 up |
| Weiss 20218 | Rats | Walking after a cut and repaired sciatic nerve | Better footprint scores at 8 to 16 weeks when given with stem cells into the calf muscle |
| Martino 202524 | 52 rats | Working memory after repeated mild head injury, dihexa infused into the brain | Working-memory deficits rescued, more at higher doses |
| Wells 20249 | 40 rats | A toxin that causes Huntington's-like damage | No protection from the motor or memory deficits |
Three of these have no author from the Harding lab: Sun 2021 from a Chinese team and Martino 2025 from Rowan University, whose animals improved, and Wells 2024, whose rats did not7249. A 2018 systematic review of angiotensin IV and its analogs, dihexa included, found that eight of nine studies in animals with a memory deficit showed better performance, with the clearest effects when the compounds went straight into the brain close to the time of learning10.
Every result in this section is from animals or cells. Memory drugs that work in rodent mazes often fail in people, which is what the fosgonimeton trial below put to the test.
What happened to the research behind dihexa?
The mechanism most dihexa pages repeat comes from Benoist 2014: dihexa binds hepatocyte growth factor (HGF), a growth signal, and boosts its action at its receptor, c-Met, and blocking HGF in the brain cancelled dihexa's memory effect in rats11. In September 2021 the journal, the Journal of Pharmacology and Experimental Therapeutics, placed notices of concern on that paper and on McCoy 20131213. In April 2025 it retracted Benoist 2014 together with two earlier papers from the lab on its HGF work, from 2011 and 20121415. McCoy 2013 still stands, with its notice of concern.
The background is on the public record. In October 2021 a special committee of Athira's board found that its chief executive, Leen Kawas, had altered images in her 2011 doctoral dissertation and in at least four papers she co-authored as a Washington State graduate student, published from 2011 to 2014; she resigned4. Kawas is an author on McCoy 2013 and on all three retracted papers51115.
What that leaves: the HGF/c-Met explanation has no standing published proof, the founding rat results carry a flag, and the main independent replication, Sun 2021, traced the mouse effect to a different pathway, PI3K/AKT, without testing HGF7.
What did fosgonimeton's Alzheimer's trial find?
Athira screened molecules that boost HGF/MET signaling, picked the most potent, and made it a prodrug, a form the body converts back into the active drug: fosgonimeton16. In its first human trial, 88 people received injections under the skin, a single dose or one a day for 9 days; it was well tolerated, and in 7 people with Alzheimer's against 4 on placebo one brain-wave timing measure, the P300 latency, moved toward normal17.
The real test was LIFT-AD, a randomized phase 2/3 trial. Its main analysis gave 287 people with mild to moderate Alzheimer's, none on cholinesterase-inhibitor memory drugs, 40 mg of fosgonimeton or a placebo under the skin daily for 26 weeks. It missed its primary goal, a combined score of memory tests and daily function (p = 0.70), and every secondary goal; the small differences that favored the drug were within chance18. Across everyone dosed, serious side effects were no more common than on placebo (4.2% against 6.9%), while 14.2% stopped because of side effects against 4.6%, mostly from injection-site reactions18. The open-label extension was ended because LIFT-AD missed its primary goal19.
Fosgonimeton is not dihexa. The trial gave a different molecule by injection, so its 40 mg dose says nothing about dihexa amounts, and its safety record covers 26 weeks of that drug only.
Does dihexa raise cancer risk?
Nobody has measured it. What is known is the pathway. MET, the gene behind the c-Met receptor, is an oncogene: when it is genetically altered it can start and sustain a cancer, and when it is switched on in a tumor it helps cancer cells survive and spread20. The FDA approves MET-blocking drugs such as capmatinib (Tabrecta) for adults with advanced non-small cell lung cancer whose tumors carry a MET exon 14 skipping mutation21. Dihexa was described as amplifying HGF's action at that receptor, in a paper now retracted1114.
No published study has tested whether dihexa makes tumors grow, in animals or in people, and the FDA has no human exposure data on it2. LIFT-AD reported serious side effects as balanced, but 26 weeks is too short, and the trial was not built, to answer a cancer question18. The risk is real as biology and unmeasured as an outcome.
When to get help If you have had cancer or are being treated for one, talk to your oncologist before taking anything meant to switch on a growth-factor pathway, since MET drives some tumors.
What do people who take dihexa report?
Commonly reported: sharper word-finding and verbal fluency within the first days for those who respond, and for many little or no change. The usual complaints are overstimulation, poor sleep, anxiety and vivid dreams; people who get them move it to the morning, take less, or stop.
These are user reports, not trial results: no human study has measured what dihexa does to thinking, sleep or mood, and no trial has set a human dose.
If you track it, PepVault logs each dose and any side effect you note that day, so a change in sleep or mood sits next to the dose before it. Log it in PepVault
Is dihexa approved or legal?
No drug containing dihexa or fosgonimeton is approved in the US: Drugs@FDA lists no application for either22. As of November 2026, FDA's compounding safety page lists dihexa acetate among bulk substances nominated but withdrawn2, and FDA's Pharmacy Compounding Advisory Committee has dihexa acetate on the agenda of a meeting set for before the end of February 2027, among substances being considered for the list pharmacies may compound from23. Products sold online are labelled for research, with no trial and no FDA review behind them.
Sources
- ClinicalTrials.gov API v2. Searches for "dihexa" and "PNB-0408", November 20, 2026: no registered studies. clinicaltrials.gov
- U.S. Food and Drug Administration. Certain bulk drug substances for use in compounding that may present significant safety risks: bulk drug substances nominated but withdrawn, dihexa acetate (content current as of April 22, 2026). fda.gov
- PubChem (NCBI). Fosgonimeton, CID 156596375 (ATH-1017 free acid), and dihexa, CID 129010512: structures and IUPAC names; fosgonimeton is dihexa's tyrosine dihydrogen phosphate. pubchem.ncbi.nlm.nih.gov
- Athira Pharma, Inc. Athira Pharma announces leadership changes (press release, GlobeNewswire, October 21, 2021; filed with the SEC as Form 8-K Exhibit 99.1): the special committee's findings on Dr. Leen Kawas's doctoral research; ATH-1017 converted in plasma to an active tyrosine metabolite (dihexa). biospace.com
- McCoy AT, Benoist CC, Wright JW, et al. Evaluation of metabolically stabilized angiotensin IV analogs as procognitive/antidementia agents. J Pharmacol Exp Ther. 2013;344(1):141-54. PubMed 23055539
- Uribe PM, Kawas LH, Harding JW, Coffin AB. Hepatocyte growth factor mimetic protects lateral line hair cells from aminoglycoside exposure. Front Cell Neurosci. 2015;9:3. PubMed 25674052
- Sun X, Deng Y, Fu X, et al. AngIV-analog dihexa rescues cognitive impairment and recovers memory in the APP/PS1 mouse via the PI3K/AKT signaling pathway. Brain Sci. 2021;11(11):1487. PubMed 34827486
- Weiss JB, Phillips CJ, Malin EW, et al. Stem cell, granulocyte-colony stimulating factor and/or dihexa to promote limb function recovery in a rat sciatic nerve damage-repair model. Ann Med Surg (Lond). 2021;71:102917. PubMed 34703584
- Wells RG, Azzam AF, Hiller AL, Sardinia MF. Effects of an angiotensin IV analog on 3-nitropropionic acid-induced Huntington's disease-like symptoms in rats. J Huntingtons Dis. 2024;13(1):55-66. PubMed 38489193
- Ho JK, Nation DA. Cognitive benefits of angiotensin IV and angiotensin-(1-7): a systematic review of experimental studies. Neurosci Biobehav Rev. 2018;92:209-225. PubMed 29733881
- Benoist CC, Kawas LH, Zhu M, et al. The procognitive and synaptogenic effects of angiotensin IV-derived peptides are dependent on activation of the hepatocyte growth factor/c-met system. J Pharmacol Exp Ther. 2014;351(2):390-402. Retracted 2025. PubMed 25187433
- Notice of concern: Benoist CC, et al. (2014). J Pharmacol Exp Ther. 2021;378(3):311. PubMed 34551987
- Notice of concern: McCoy AT, et al. (2013). J Pharmacol Exp Ther. 2021;378(3):313. PubMed 34551989
- Retraction notice to "The procognitive and synaptogenic effects of angiotensin IV-derived peptides are dependent on activation of the hepatocyte growth factor/c-Met system". J Pharmacol Exp Ther. 2025;392(4):103567. PubMed 40312093
- Kawas LH, McCoy AT, Yamamoto BJ, Wright JW, Harding JW. Development of angiotensin IV analogs as hepatocyte growth factor/Met modifiers. J Pharmacol Exp Ther. 2012;340(3):539-48. Retracted 2025 (J Pharmacol Exp Ther. 2025;392(4):103566); the lab's 2011 paper on HGF dimerization-domain mimics (PMID 21859930) was retracted in the same issue (103568). PubMed 22129598
- Johnston JL, Reda SM, Setti SE, et al. Fosgonimeton, a novel positive modulator of the HGF/MET system, promotes neurotrophic and procognitive effects in models of dementia. Neurotherapeutics. 2023;20(2):431-451. PubMed 36538176
- Hua X, Church K, Walker W, et al. Safety, tolerability, pharmacokinetics, and pharmacodynamics of the positive modulator of HGF/MET, fosgonimeton, in healthy volunteers and subjects with Alzheimer's disease: phase I clinical trial. J Alzheimers Dis. 2022;86(3):1399-1413. PubMed 35180125
- Porsteinsson AP, Sabbagh M, Tariot PN, et al. Fosgonimeton in mild-to-moderate Alzheimer's disease (LIFT-AD, NCT04488419). J Alzheimers Dis Rep. 2025;9:25424823251405817. PubMed 41393340
- ClinicalTrials.gov. NCT04886063: Open label study of ATH-1017 for treatment of mild to moderate Alzheimer's disease. Terminated after LIFT-AD did not meet its primary endpoint. clinicaltrials.gov
- Comoglio PM, Trusolino L, Boccaccio C. Known and novel roles of the MET oncogene in cancer: a coherent approach to targeted therapy. Nat Rev Cancer. 2018;18(6):341-358. PubMed 29674709
- TABRECTA (capmatinib) tablets, prescribing information, section 1 Indications and usage. DailyMed, Novartis. dailymed.nlm.nih.gov
- U.S. Food and Drug Administration. Drugs@FDA (openFDA drugsfda endpoint), searches for dihexa and fosgonimeton, November 20, 2026: no matches. api.fda.gov
- U.S. Food and Drug Administration. Meeting of the Pharmacy Compounding Advisory Committee, before the end of February 2027: agenda (content current as of April 15, 2026). fda.gov
- Martino KA, Nakhre A, Demarest RM, Devilbiss DM. Hepatocyte growth factor/MET activator rescues working memory deficits after repeated mild traumatic brain injury. Neurotrauma Rep. 2025;6(1):991-998. PubMed 42825082
Newsletter · Weekly
The weekly peptide briefing.
New guides, the week’s peptide news, one tool. Free, one email a week, unsubscribe anytime.