Dihexa Research Overview: Rat Studies, a Retraction, No Trials

Dihexa is a small synthetic peptide-like molecule, N-hexanoic-Tyr-Ile-(6) aminohexanoic amide, derived from angiotensin IV. Its published evidence is rodent work, much of it from the group that founded the company that tried to develop it. We found no human trial. The 2014 paper that proposed how it works was retracted in April 2025 [3][5].

Key points

• Dihexa came out of Washington State University (WSU) research published from 2013. All its cognitive and synapse data are from rats, mice and cultured rat cells [1][2][14].
• The paper that tied dihexa to the hepatocyte growth factor (HGF)/c-Met system was retracted by the Journal of Pharmacology and Experimental Therapeutics in April 2025. The 2013 founding paper carries a 2021 expression of concern [3][5][6].
• ClinicalTrials.gov returned zero records for dihexa and for its code name PNB-0408. A PubMed search returned 19 records, none a human trial of dihexa [16][17].
• FDA has not approved it, and its compounding page says it has found no human exposure data on dihexa acetate [20][22].
• A “B tier” placement in a viral cognition ranking is not supported by the evidence.

What dihexa is

Dihexa is a modified version of the three N-terminal amino acids of norleucine-angiotensin IV. The 2013 paper that introduced it reports chemical changes meant to make the molecule survive digestion and cross the blood-brain barrier [1]. Its code name is PNB-0408 [15]. Prices are on our dihexa compound page; this article covers the research.

Where it came from

The papers come from Joseph Harding, John (Jay) Wright and colleagues at WSU [10]. Harding and Wright state that they are founders and shareholders of M3 Biotechnology, which was developing treatments based on this work [2][4].

Athira Pharma’s SEC filing says the company was incorporated as M3 Biotechnology, Inc. in Washington on 2011-03-31, moved its incorporation to Delaware on 2015-10-27 and changed its name to Athira Pharma in April 2019 [9]. In an 8-K dated 2021-10-21 Athira wrote that from 2011 to 2015 it tried to develop dihexa, which WSU had exclusively licensed to it. Its studies showed dihexa enhanced HGF/MET activity but “failed to demonstrate” the characteristics needed for a successful therapeutic, so development was discontinued [8].

The same filing says Athira’s later candidate, ATH-1017 (fosgonimeton), is a prodrug that “metabolizes rapidly in plasma” into dihexa. Per the filing, it was chosen from more than 50 compounds for better drug-like properties [8]. That is a company description, and ATH-1017 is a different drug. Its Phase 2/3 trial LIFT-AD (312 people with mild-to-moderate Alzheimer’s disease, 26 weeks) missed its primary endpoint (p=0.70), according to the company’s 2024-09-03 release [19].

Integrity problems around the early papers

The 8-K reports a special committee finding that Athira’s former CEO Leen Kawas, a co-author of the early papers, altered images in her 2011 doctoral dissertation and in at least four WSU papers published 2011 to 2014. It also found that WSU’s dihexa patent includes images from papers she co-authored, some altered. WSU’s own investigation was ongoing in October 2021 [8]. We could not verify its outcome.

In September 2021 the journal issued notices of concern on four papers from the group, including the 2013 and 2014 dihexa papers [6][7]. Retraction Watch quotes the notices as citing “possible image manipulation” [7]. In April 2025 the 2014 paper was retracted [3][5]. We could not open the retraction notice, so we cannot say what reason it gives.

What the key preclinical papers found

Study	Model and size	Route (as reported)	Result
McCoy et al., 2013, scopolamine model [2]	Male Sprague-Dawley rats, 8-10 per group	Intracerebroventricular, intraperitoneal, oral gavage	At the highest tested dose by each route, dihexa reversed scopolamine-impaired water-maze learning to vehicle level
McCoy et al., 2013, aged rats [2]	24-month-old rats, 6 (3 male, 3 female)	Oral	Better performance on most test days. Authors note the rats were not pre-screened for deficits
McCoy et al., 2013, synapses [2]	Cultured neonatal rat hippocampal neurons; rat brain slices	Added to culture	41 spines per 50 micrometres versus 15 for vehicle (200 segments); 11 versus 7 in slices
Benoist et al., 2014 (retracted) [3][4]	Rats, 8-10 per group; cell assays	Oral dihexa, intracerebroventricular HGF antagonist	Reported that an HGF antagonist blocked dihexa’s effect
Sun et al., 2021, APP/PS1 mice [14]	Male transgenic mice, about 6 months old, 6 or 12 per group, about 3 months of treatment	Methods say both intragastric and intraperitoneal	Better water-maze performance, lower inflammatory markers
2024 Huntington’s-like model [15]	40 male Wistar rats, 5 weeks	Not stated in the abstract	“did not protect rats” from the induced deficits

The 2013 paper was funded by NIH and others and discloses founder ownership of M3 [2]. Scopolamine blocks acetylcholine receptors and impairs memory; reversing it shows an effect in that model, not on dementia. The 2021 mouse study came from Chinese institutions, declared no conflict of interest and gives inconsistent routes in its Methods [14].

A 2018 systematic review of animal studies of angiotensin IV and its analogs, dihexa among them, screened 450 articles and included 32. In cognitive-deficit models eight of nine studies found improvement. The review concluded that these peptides work best when given into the brain ventricles close to learning or testing [13]. It covers no human work.

Is there any human evidence?

We found none for dihexa itself. ClinicalTrials.gov searched on 2026-11-08 returned zero studies for “dihexa”, for “PNB-0408” and for the full chemical name [16]. The PubMed search for “dihexa” returned 19 records. We read the titles of all 19: animal studies, reviews and unrelated cell or chemistry papers, with no report of dihexa given to people [17]. The only human data nearby are the fosgonimeton trials above, which concern a different compound, and the trial record shows 5 registered studies [18].

Theoretical concerns the literature raises

The concern is mechanistic, not observed harm. The group’s own review of Parkinson’s disease work states that “c-Met has been shown to play a role in multiple types of cancer” and says it “remains to be seen” whether long-term treatment with small-molecule HGF mimetics is possible and effective [10]. An oncology review describes c-MET as a receptor that, once HGF binds, drives proliferation, motility, migration and invasion, and that is aberrantly activated in human cancers through mutation, amplification or overexpression [11].

Whether dihexa activates c-Met in people, and whether doing so would promote tumours, is untested. We found no carcinogenicity study of dihexa. The mechanism itself is contested. The retracted paper was the main direct evidence for the HGF/c-Met route [3]. A 2015 review from the same group mentions “controversy” over what the angiotensin IV receptor is, with another group proposing insulin-regulated aminopeptidase [12]. The 2021 mouse study proposed PI3K/AKT signalling instead [14].

Regulatory status

• United States, checked 2026-11-08: Not an approved drug. A query of openFDA’s Drugs@FDA data for “dihexa” returned no match [22]. FDA’s page on bulk substances that may present significant safety risks, current as of 2026-04-22, lists “dihexa acetate” under “nominated but withdrawn”, a list of substances “previously in category 2” that “were withdrawn by the nominators”. FDA states it “has not identified any human exposure data” on drug products containing it and “lacks important information regarding any safety issues” [20]. So FDA currently does not list it as a category 2 substance, and it does not clear it either.
• UK, Germany, EU/EMA, and US scheduling: not checked.

What “research chemical” labelling means

Peptide sellers often label products “for research use only” or “not for human consumption”. That wording states the seller’s purpose. It does not mean a regulator reviewed a product’s identity, purity or safety. We did not review any dihexa product label. In a letter dated 2026-03-31 to a peptide seller, FDA said such labelling did not outweigh website evidence that products were intended as human drugs [21]. That letter concerns other products and does not name dihexa. We found no FDA letter naming dihexa, but we did not search exhaustively.

Risks and unknowns

• Documented adverse effects in people: none, because no human study exists [16][17].
• Cancer and tissue growth: theoretical and untested for dihexa, as above [10][11].
• Interactions, pregnancy, children, older adults, people with a cancer history: not studied.
• Healthy people: rodent models used a scopolamine deficit, aged animals or disease models. We found no study showing cognitive enhancement in healthy animals or people.
• Replication: the one independent efficacy paper we found, in an Alzheimer’s-type mouse model, was positive [14]. The one Huntington’s-like rat study was negative [15].

What the source post claims and how it holds up

On 2026-10-20 the account @dex7official posted “THE ONLY TRUE COGNITIVE ENHANCEMENT TIER LIST (NO COPES EDITION)”, an image with no supporting text. It had 6,522 likes and about 1.2 million views at fetch [23]. In the image a vial labelled “Offline Peptides / Dihexa” sits in tier B, above CDP-choline and noopept (C) and below phenylpiracetam (A). The poster cites no evidence. Our full tier-list check tests the list as a whole.

Verdict: not supported. A tier ranking implies a comparative cognitive benefit. For dihexa, no human data exist, the mechanism paper is retracted, the main safety question is untested, and the strongest results are from disease-model rodents rather than healthy people [2][3][14][16]. The placement is an opinion. It is not a finding.

Evidence gaps

No human pharmacokinetic, safety or efficacy data. No carcinogenicity or long-term toxicity studies. An unresolved mechanism. Few independent replications. We did not verify what products sold under the name contain.

Frequently asked questions

Is dihexa approved by the FDA?

No. We found no approved product, and FDA’s compounding page says it has identified no human exposure data on dihexa acetate. UK, German and EMA status were not checked [20][22].

Has dihexa been tested in humans?

We found no registered or published trial. ClinicalTrials.gov returned zero studies, and PubMed showed no human dihexa study. Athira’s prodrug fosgonimeton was tested in people, but it is a different compound [16][17][19].

Why was the 2014 dihexa paper retracted?

PubMed lists it as retracted in April 2025, after a 2021 expression of concern over possible image manipulation. We could not open the retraction notice and cannot state its reason [3][5][7].

Does dihexa cause cancer?

No study has shown that, and none has tested it. The worry comes from c-Met’s role in cancer biology, which the developers’ own review acknowledges [10][11].

Sources

1. NCBI PubMed (data read via NCBI E-utilities), McCoy AT et al., “Evaluation of metabolically stabilized angiotensin IV analogs as procognitive/antidementia agents”, J Pharmacol Exp Ther (2013-01). https://pubmed.ncbi.nlm.nih.gov/23055539/ . Abstract; link to 2021 notice of concern.
2. PubMed Central, same paper, full text. https://pmc.ncbi.nlm.nih.gov/articles/PMC3533412 . Animal numbers, routes, results, funding and competing interests.
3. NCBI PubMed (via E-utilities), Benoist CC 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 (2014-11). https://pubmed.ncbi.nlm.nih.gov/25187433/ . Abstract; record flagged “Retracted Publication”.
4. PubMed Central, same paper, full text. https://pmc.ncbi.nlm.nih.gov/articles/PMC4201273 . Model, sample size, competing interests, retraction banner.
5. NCBI PubMed (via E-utilities), “Retraction notice to ‘The Procognitive and Synaptogenic Effects of Angiotensin IV-Derived Peptides…’”, J Pharmacol Exp Ther 392(4):103567 (2025-04). https://pubmed.ncbi.nlm.nih.gov/40312093/ . Title and date only; the notice text was not accessible.
6. NCBI PubMed (via E-utilities), two “Notice of Concern” records for the 2013 and 2014 papers, J Pharmacol Exp Ther 378(3) (2021-09). PMIDs 34551989 and 34551987. Titles only.
7. Retraction Watch, “Four papers by Athira CEO earn expressions of concern” (2021-09-24). https://retractionwatch.com/2021/09/24/four-papers-by-athira-ceo-earn-expressions-of-concern/ . Wording of the notices.
8. Athira Pharma, Form 8-K (2021-10-21). https://www.sec.gov/Archives/edgar/data/1620463/000119312521304002/d179221d8k.htm . Special committee findings; dihexa development history; ATH-1017 description.
9. Athira Pharma, SEC EDGAR filing 0000950170-22-017421, organization note (filed 2022). https://www.sec.gov/Archives/edgar/data/1620463/000095017022017421/R7.htm . Incorporation and name-change dates.
10. Wright JW, Harding JW, “Importance of the brain angiotensin system in Parkinson’s disease”, Parkinson’s Disease (2012). https://pmc.ncbi.nlm.nih.gov/articles/PMC3503402 . WSU affiliation; c-Met and cancer statement.
11. Organ SL, Tsao MS, “An overview of the c-MET signaling pathway”, Ther Adv Med Oncol (2011-11). https://pubmed.ncbi.nlm.nih.gov/22128289/ . Abstract.
12. Wright JW, Harding JW, “The development of small molecule angiotensin IV analogs to treat Alzheimer’s and Parkinson’s diseases”, Prog Neurobiol (2015-02). https://pubmed.ncbi.nlm.nih.gov/25455861/ . Abstract; receptor identity controversy.
13. “Cognitive benefits of angiotensin IV and angiotensin-(1-7): a systematic review of experimental studies”, Neurosci Biobehav Rev (2018-09). https://pubmed.ncbi.nlm.nih.gov/29733881/ . Abstract.
14. Sun et al., “AngIV-analog dihexa rescues cognitive impairment and recovers memory in the APP/PS1 mouse via the PI3K/AKT signaling pathway”, Brain Sciences (2021-11). https://pmc.ncbi.nlm.nih.gov/articles/PMC8615599/ . Full text.
15. “Effects of an angiotensin IV analog on 3-nitropropionic acid-induced Huntington’s disease-like symptoms in rats”, J Huntingtons Dis (2024). https://pubmed.ncbi.nlm.nih.gov/38489193/ . Abstract; PNB-0408 naming.
16. ClinicalTrials.gov API v2, searches for “dihexa”, “PNB-0408”, full chemical name (run 2026-11-08). https://clinicaltrials.gov/api/v2/studies?query.term=dihexa . Zero results each.
17. NCBI PubMed E-utilities search “dihexa” (run 2026-11-08): 19 records, titles reviewed. https://pubmed.ncbi.nlm.nih.gov/?term=dihexa
18. ClinicalTrials.gov, NCT04488419, ATH-1017 Phase 2/3 (record read 2026-11-08), plus search for fosgonimeton or ATH-1017 returning 5 studies. https://clinicaltrials.gov/study/NCT04488419
19. Athira Pharma, “Athira Pharma announces topline results from Phase 2/3 LIFT-AD clinical trial of fosgonimeton” (2024-09-03), read via BioSpace copy. https://www.biospace.com/drug-development/athira-pharma-announces-topline-results-from-phase-2-3-lift-ad-clinical-trial-of-fosgonimeton-for-mild-to-moderate-alzheimers-disease
20. U.S. FDA, “Certain bulk drug substances for use in compounding that may present significant safety risks” (content current as of 2026-04-22). https://www.fda.gov/drugs/human-drug-compounding/certain-bulk-drug-substances-use-compounding-may-present-significant-safety-risks
21. U.S. FDA, warning letter to Lovega LLC dba Pink Pony Peptides (2026-03-31). https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/warning-letters/lovega-llc-dba-pink-pony-peptides-721088-03312026
22. openFDA Drugs@FDA API query for “dihexa” (run 2026-11-08). https://api.fda.gov/drug/drugsfda.json . No match.
23. @dex7official, “THE ONLY TRUE COGNITIVE ENHANCEMENT TIER LIST (NO COPES EDITION)” (2026-10-20). https://x.com/dex7official/status/2112691493261484516 . Text and image read through the fxtwitter mirror.

This article is information, not medical advice.