1

Overview

Dihexa is the short name for N-hexanoic-Tyr-Ile-(6) aminohexanoic amide, a peptide related to angiotensin IV that was reported by researchers at Washington State University.1 The 2013 paper describes it as "an orally active, blood-barrier permeant, metabolically stabilized analog".1 A 2015 review by Wright and Harding, whose affiliations include M3 Biotechnology, calls dihexa "a first-in-class compound that is orally active, penetrates the blood-brain barrier, and facilitates memory consolidation and retrieval".2

The app files dihexa as a cognitive neuropeptide with oral and nasal routes and grades its evidence as anecdotal. The studies described below are in rats, mice and cultured cells. No controlled trial in people was identified in PubMed (19 records returned for "dihexa") or in ClinicalTrials.gov (no study returned).

The app's notes describe dihexa as an angiotensin IV receptor agonist. Benoist and colleagues (Washington State University, Journal of Pharmacology and Experimental Therapeutics, 2014) titled their paper to say that the procognitive and synaptogenic effects of angiotensin IV-derived peptides "Are Dependent on Activation of the Hepatocyte Growth Factor/c-Met System".3 The record for this paper carries the notice "This article has been retracted. See J Pharmacol Exp Ther. 2025 Apr 29;392(4):103567".3 They report that radiolabelled dihexa bound human hepatocyte growth factor (HGF) in a cell-free assay "saturably and with high affinity (Kd = 65 pM)".3 In HEK-293 cells, dihexa alone at 10–10 M and 10–12 M did not activate c-Met, but "dihexa at both concentrations markedly augmented the capacity of HGF at 1.25 and 2.5 ng/ml to activate c-Met".3 In MDCK cell scattering, dihexa at 10–10 M facilitated HGF-dependent scattering, while 10–12 M and 10–14 M produced a trend that "did not reach statistical significance", and dihexa alone at 10–12 M augmented scattering, which the authors say suggests "the presence of endogenous HGF in the preparation".3 The authors also report that dihexa "effectively inhibited HGF dimerization at 10–10 M concentration", and suggest, while noting this was not directly addressed, that it "might act allosterically to stabilize an active HGF conformation".3 In their summary, the procognitive capacity of orally delivered dihexa in rats "was blocked by an HGF antagonist delivered intracerebroventricularly".3

In APP/PS1 mice, Sun and colleagues (China Pharmaceutical University, Brain Sciences, 2021) report that "Dihexa activated the PI3K/AKT signaling pathway" and that a PI3K inhibitor reversed anti-inflammatory and anti-apoptotic effects.4 In rats after repeated mild traumatic brain injury, Martino and colleagues (Rowan University, Neurotrauma Reports, 2025) report that the procognitive effects of dihexa "were blocked by Hinge, an HGF/MET dimerization antagonist".5

3

Half-life and pharmacokinetics

Values from the Milligram app's model.

Half-life
~2 h
Bioavailability
50%
Typical dose
10–20 mg
Route
Oral / Nasal
Time to peak
~1 h

Sources in this part

  1. 1McCoy et al., 2013

The only measured pharmacokinetic figures located are from rats, in McCoy and colleagues (Washington State University, Journal of Pharmacology and Experimental Therapeutics, 2013).1 Adult male Sprague-Dawley rats received 10 mg/kg intravenously, and the plasma data were modelled by noncompartmental analysis.1 The paper reports "a long half-life (t1/2) of 12.68 days following i.v. administration and similarly when delivered intraperitoneally", with the intraperitoneal figure given as 8.83 ± 2.41 days (mean ± S.E.M., N = 4).1 The table of intravenous parameters is headed "Dihexa pharmacological parameters (N = 3)" and gives the terminal half-life as 18,256 ± 7787 minutes.1 The authors write that dihexa "appeared to be extensively distributed outside the central blood compartment and/or bound within the tissues as evidenced by its large Vd".1 In pooled male rat liver microsomes, phase I metabolism was "very low", with "an average half-life of 509.4 minutes".1 They call their own pharmacokinetic work "cursory".1

The app's figure for half-life is about 2 hours, 150 times shorter than the rat intravenous figure of 12.68 days (about 304 hours); no human half-life was identified in the sources.1 For oral bioavailability, the app's figure is about 50%; the McCoy paper gives no measured percentage and writes that dihexa "was predicted to be orally bioavailable", based on modelling of its properties.1 Its brain entry was tested by infusing radiolabelled dihexa into the carotid artery of rats, after which "all of the brain regions examined avidly concentrated dihexa, attesting to its ability to cross the BBB".1 No figure for a nasal route, a time to peak or a duration of action was identified in the sources.

Table 3.1Half-life and pharmacokinetics5 rows
ParameterApp's figureReported in the sources
Half-life~2 hRat, intravenous 10 mg/kg: 12.68 days (18,256 ± 7787 min, N = 3); rat, intraperitoneal: 8.83 ± 2.41 days (N = 4)1
Bioavailability50%No measured figure; predicted to be orally bioavailable from modelling1
Time to peak~1 hNo figure in the sources
Duration~6 hNo figure in the sources
RoutesOral / NasalOral gavage, intraperitoneal, intravenous and intracerebroventricular routes were used in rats; no nasal study was identified1
4

Human studies

No controlled trial in people was identified in PubMed or ClinicalTrials.gov. The FDA page on bulk drug substances that may present significant safety risks states: "FDA has not identified any human exposure data on drug products containing dihexa acetate administered via any route of administration".6 The table below lists the animal and cell studies that were read; none is a human study.

Table 4.1Human studies6 rows
Study and yearDesignPeopleDose and lengthMain result
McCoy and colleagues, J Pharmacol Exp Ther 20131Rat scopolamine memory model, Morris water maze; pharmacokinetics; cultured neuronsNone (rats; n = 8–10 per group)Dihexa 0.1 or 1 nmol i.c.v., 0.05 to 0.50 mg/kg i.p., 1.25/kg and 2.0 mg/kg oral; 8 training daysIn scopolamine-treated rats, the high dose of each route performed significantly better than the scopolamine groups and was indistinguishable from vehicle controls; the 0.05 mg/kg i.p. group did not differ from scopolamine alone1
McCoy and colleagues, aged-rat arm, 20131Daily oral dosing, water maze, no scopolamineNone (24-month-old rats, 3 male and 3 female per group)2 mg/kg by gavage daily; 8 daysThe learning curve of treated rats was significantly different from that of untreated rats (Mann-Whitney U test)1
Benoist and colleagues, J Pharmacol Exp Ther 2014 (record carries a retraction notice)3Cell-free binding, HEK-293 cells, cultured neurons, rat scopolamine water maze with antagonist (N = 8–10)None (cells and rats)10–10 M and 10–12 M in cells; oral dihexa 2 mg/kg per day in rats, 8 daysSaturable HGF binding (Kd = 65 pM); dihexa alone did not activate c-Met; effects blocked by an HGF antagonist and a c-Met shRNA3
Sun and colleagues, Brain Sci 20214APP/PS1 Alzheimer's-model mice, three arms, Morris water mazeNone (male mice, n = 12 per group in the maze)1.44 or 2.88 mg/kg; three monthsEscape latency fell "to varying degrees"; the figure legend says dihexa "tended to increase" target-quadrant crossings4
Martino and colleagues, Neurotrauma Rep 20255Repeated mild traumatic brain injury in rats, T-maze working memoryNone (52 Long-Evans rats, 33 male and 19 female)1 nmol/day i.c.v. at the highest dose; days 1 to 581.5% against 67.1% correct with vehicle; blocked by Hinge (68.6%)5
Wells and colleagues, J Huntingtons Dis 20247Rat 3-nitropropionic acid model of Huntington's disease-like symptomsNone (40 male Wistar rats, three groups)Five weeks; dose not stated in the recordNull: the compound "did not protect rats from the deficits induced by 3-NP neurotoxicity"7

Deeper analysis: McCoy and colleagues, 2013

McCoy and colleagues (Washington State University, Journal of Pharmacology and Experimental Therapeutics, 2013) set out to modify an angiotensin IV fragment for stability and barrier permeability, then tested the candidate in rats and cultured neurons.1 In the memory model, rats received scopolamine into the brain 20 minutes before training and dihexa afterwards, with five trials a day for 8 days.1 For oral dosing, "The high oral dose (2 mg/kg) of dihexa completely reversed the scopolamine-dependent learning deficit", whereas the lower dose partly reversed it.1 In cultured rat hippocampal neurons treated for 5 days at 10−12 M, the average was 41 spines per 50-µm dendrite length with dihexa against 15 for vehicle.1 The mean frequency of AMPA-mediated miniature excitatory postsynaptic currents was 4.82 ± 0.34 Hz (29 cells) with dihexa against 3.06 ± 0.23 Hz (33 cells) with vehicle.1 The aged-rat arm used 6 animals per group (3 male and 3 female), compared treated rats with untreated rats by a Mann-Whitney U test (P < 0.03), and the authors note that these aged rats were not prescreened for cognitive deficits.1 The study was supported in part by a Washington State endowment, an NIH grant (MH086032) and a Hope for Depression Research Foundation grant.1 The paper states: "J.W.W. and J.W.H. are founders and shareholders of M3 Biotechnology, LLC, which is developing pharmaceuticals based on this technology".1 The record carries a notice that "An expression of concern has been published for this article. See J Pharmacol Exp Ther. 2021 Sep;378(3):313".1 The sources read do not say what the concern was.

Deeper analysis: Benoist and colleagues, 2014

Benoist and colleagues (Washington State University, Journal of Pharmacology and Experimental Therapeutics, 2014) aimed "to elucidate the mechanism that underlies dihexa's procognitive activity".3 The binding result came from [3H]dihexa and human HGF in a soluble assay, a composite of three experiments run in quadruplicate.3 In HEK-293 cells, unlike dihexa, Nle1-AngIV alone at 10–10 M and 10–12 M "was capable of activating c-Met".3 The paper introduces the dimerization finding with the words "Somewhat surprisingly, dihexa effectively inhibited HGF dimerization".3 The work was supported by grants from the Michael J. Fox Foundation and M3 Biotechnology, Inc. (to J.W.H.), among others.3 The disclosures state: "J.W.H. and J.W.W. are cofounders and major shareholders of M3 Biotechnology, Inc." and "L.H.K. is the CEO of M3 Biotechnology, Inc."; L.H.K. is an author of the paper.3 The record carries a notice that "An expression of concern has been published for this article. See J Pharmacol Exp Ther. 2021 Sep;378(3):311" and a later notice that "This article has been retracted. See J Pharmacol Exp Ther. 2025 Apr 29;392(4):103567".3

Deeper analysis: Sun and colleagues, 2021

Sun and colleagues gave dihexa to male APP/PS1 mice from six to nine months of age, with groups at 1.44 and 2.88 mg/kg, and tested them in the Morris water maze after three months of dosing.4 The methods text states both that the drug was "administered intragastrically" and that it "was administered intraperitoneally", so the route is not consistent within the paper.4 The results text says dihexa-treated mice "performed significantly better than APP/PS1 controls for the number of platform crossings", while the figure legend says dihexa "tended to increase" those crossings.4 Swimming speed "was not significantly different among the groups".4 The authors name two limitations: "whether doses larger than 2.88 mg/kg would further improve MWM performance is worth exploring in future studies", and that an "only wortmanin" control group was missing from the inhibitor experiment.4 The funders were Chinese national and Jiangsu provincial programmes, and the authors "declare that they have no conflict of interest".4

Deeper analysis: Martino and colleagues, 2025

Martino and colleagues used 52 Long-Evans rats in a closed-head repeated injury model and a T-maze working memory task, giving dihexa by intracerebroventricular infusion.5 At 1 nmol/day, working memory in repeatedly injured animals was 81.5% against 67.1% with vehicle, and 68.6% when the HGF/MET antagonist Hinge was given first.5 The authors report that dihexa at 1.0 nmol i.c.v. limited the increase in total session time compared with saline after repeated mild injury, although this effect was not statistically significant (HSD, p = 0.9777).5 The authors report "No sex-dependent differences".5 They note that "The study design and analytic plan were not preregistered".5 They also write that "The mechanisms supporting the procognitive actions of MET are poorly understood".5 Because dihexa was infused into the brain, the study does not address oral or peripheral dosing.5 The funders were the Rowan University School of Osteopathic Medicine Heritage Foundation and the New Jersey Commission on Brain Injury Research, and the disclosure reads "No competing financial interests exist".5

5

Doses reported in studies and on labels

No label for dihexa and no human dose from a trial were identified in the sources, so the table lists only the doses used in animals and the app's figure.

Table 5.1Doses reported in studies and on labels8 rows
SourceSpecies and routeDoseLengthn
McCoy 20131Rat, intravenous (pharmacokinetics)10 mg/kgSingle dose, sampled to 120 hoursN = 3
McCoy 20131Rat, intracerebroventricular0.1 or 1 nmolDaily, 8 daysn = 8–10
McCoy 20131Rat, intraperitoneal0.05, 0.25 or 0.50 mg/kgDaily, 8 daysn = 8–10
McCoy 20131Rat, oral gavage1.25/kg and 2.0 mg/kg (as printed)Daily, 8 daysn = 8–10
McCoy 2013, aged rats1Rat, oral gavage2 mg/kgDaily, 8 daysn = 6 per group
Sun 20214Mouse (APP/PS1), intragastric or intraperitoneal as printed1.44 or 2.88 mg/kgThree monthsn = 12 per group in the maze
Martino 20255Rat, intracerebroventricularUp to 1 nmol/dayDays 1 to 552 rats in the study
The app's figureOral or nasal, in people10–20 mgNot statedNot stated
6

Side effects reported

No adverse event data from people were identified. The FDA states that it "lacks important information regarding any safety issues raised by dihexa acetate, including whether it would cause harm if administered to humans".6 In the rat study by Wells and colleagues, only the short PubMed summary was available; it reports that the compound did not protect against the neurotoxin-induced deficits and does not mention adverse events.7 Wright and Harding describe c-Met receptor activation as stimulating "mitogenesis, motogenesis, morphogenesis"; whether this bears on dihexa safety is not addressed in the sources read.2

7

Regulatory and anti-doping status

The FDA page on bulk drug substances that may present significant safety risks, with content current as of 22 April 2026, includes an entry for dihexa acetate.6 The law firm Frier Levitt reports that on April 15 the FDA published a notice "announcing its intent to remove twelve peptide bulk drug substances from Category 2", and its list of the twelve includes Dihexa Acetate.8 Frier Levitt states that the twelve are being removed "as their nominations have been withdrawn by the respective nominators".8 Frier Levitt adds: "Removal from Category 2 does not render these bulk drug substances eligible for compounding under section 503A".8 Goodwin lists dihexa acetate among substances for "later consultation with the PCAC at an advisory committee meeting expected before February 2027".9 Frier Levitt likewise places dihexa acetate among substances not scheduled for the July meeting.8 The app's note says a July 2026 FDA meeting will explore a full approval pathway; the two commentaries instead place dihexa acetate outside the July agenda.8,9 Both are law-firm summaries; the FDA's own notice was not among the sources read. No FDA approval of dihexa was identified in the sources. The 2026 WADA Prohibited List does not name dihexa in the text read.10 Its S0 class covers "Any pharmacological substance which is not addressed by any of the subsequent sections of the List and with no current approval by any governmental regulatory health authority for human therapeutic use", and names BPC-157 as an example.10

No human pharmacokinetic figure, dose, efficacy result or safety result was identified in the sources searched. The rat half-life of 12.68 days and the app's 2 hours differ by a wide margin, and the sources do not report which, if either, applies in people.1 The mechanism papers come largely from authors with ties to M3 Biotechnology; the McCoy record carries an expression of concern, and the Benoist record carries an expression of concern and a later retraction notice.1,3 The Wells study of a Huntington's disease-like rat model reported no protective effect, so the memory results are not shown to extend to other models.7 Martino and colleagues give dihexa into the brain, so oral and nasal delivery in injured animals is not covered by that study.5

What is the half-life of dihexa?

In rats, McCoy and colleagues reported a half-life of 12.68 days after intravenous dosing, while the app's figure is about 2 hours.1 No human half-life was identified in the sources.

Is dihexa banned by WADA?

The 2026 WADA Prohibited List does not name dihexa in the text read.10 Its S0 class covers pharmacological substances with no current approval by any governmental regulatory health authority for human therapeutic use.10

Is dihexa approved by the FDA?

No FDA approval of dihexa was identified in the sources. The FDA page states that it "has not identified any human exposure data on drug products containing dihexa acetate".6

What is dihexa's FDA compounding status?

Law-firm commentary reports that the FDA announced its intent to remove dihexa acetate from Category 2 in April 2026.8 The same commentary says removal "does not render these bulk drug substances eligible for compounding under section 503A".8

Has dihexa been tested in humans?

No controlled trial in people was identified in PubMed or ClinicalTrials.gov. The studies that were read are in rats, mice and cultured cells.1,4

What does dihexa bind to?

Benoist and colleagues report that dihexa binds hepatocyte growth factor (HGF) with a Kd of 65 pM in a cell-free assay.3 The paper's record carries an expression of concern dated September 2021 and a retraction notice dated 29 April 2025.3

Does dihexa cross the blood-brain barrier?

In rats given radiolabelled dihexa into the carotid artery, "all of the brain regions examined avidly concentrated dihexa".1 No measurement of brain entry in people was identified.

Did dihexa work in every animal model?

No: in the Wells study of a 3-nitropropionic acid rat model, dihexa "did not protect rats from the deficits induced by 3-NP neurotoxicity".7 Other rat and mouse studies reported memory benefits in the scopolamine, aged-rat, APP/PS1 and repeated-injury models.1,4,5

What side effects has dihexa been reported to cause?

No adverse event data from people were identified. The FDA states that it "lacks important information regarding any safety issues raised by dihexa acetate".6

11

References

  1. McCoy AT, et al. Evaluation of metabolically stabilized angiotensin IV analogs as procognitive/antidementia agents. J Pharmacol Exp Ther 2013. PMID 23055539
  2. Wright JW, Harding JW. The brain hepatocyte growth factor/c-Met receptor system: a new target for the treatment of Alzheimer's disease. J Alzheimers Dis 2015. PMID 25649658
  3. 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. PMID 25187433
  4. Sun 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. PMID 34827486
  5. Martino KA, et al. Hepatocyte growth factor/MET activator rescues working memory deficits after repeated mild traumatic brain injury. Neurotrauma Rep 2025. PMID 42825082
  6. U.S. Food and Drug Administration. Certain bulk drug substances for use in compounding that may present significant safety risks. Content current as of 22 April 2026. FDA page
  7. Wells RG, et al. Effects of an angiotensin IV analog on 3-nitropropionic acid-induced Huntington's disease-like symptoms in rats. J Huntingtons Dis 2024. PMID 38489193
  8. Frier Levitt. FDA to remove 12 popular peptides from the Category 2 "Do Not Compound" list. Law-firm commentary, 2026. Frier Levitt commentary
  9. Goodwin. FDA signals potentially evolving stance toward compounding of certain peptides. Law-firm commentary, May 2026. Goodwin commentary
  10. World Anti-Doping Agency. World Anti-Doping Code International Standard: Prohibited List 2026. WADA list