Dihexa in 2026: The Human Trial Failed, the Foundational Paper Is Flagged, and the Vial Trade Hasn’t Told Anyone

The rules changed in 2026, and most guides selling this peptide are still working off the 2013 press release version of the story.

Here is the news, stripped down. Dihexa, a lab-built peptide out of Washington State University, has never completed a human efficacy trial. The 2013 rat study that built its reputation now carries a formal Notice of Concern from the journal that published it. And a closely related compound that actually made it into a human Alzheimer’s trial, fosgonimeton, missed its primary endpoint when results landed in 2024. None of that is new information exactly, but it is information that most “best Dihexa” pages still leave out of the pitch, and as of this June 2026 update, that omission is the whole story.

Three dates, and why they matter more than the marketing copy

Reporters like a timeline because a timeline doesn’t argue, it just sits there. This one has three stops, and reading them in order tells you more than any sales page will.

2013. McCoy and colleagues publish the foundational paper in the Journal of Pharmacology and Experimental Therapeutics, reporting that oral Dihexa reversed memory deficits in rats [1]. This is the study every seller quotes.

2021. The same journal attaches a formal Notice of Concern to that paper [2]. Not a retraction, a flag, but a flag that changes how the data should be read.

2024. Fosgonimeton, a chemical relative built on the same HGF/c-Met mechanism, fails to hit its primary endpoint in the large LIFT-AD Alzheimer’s trial [5]. It is the closest thing to a human test this science has had, and it came back negative.

Put those three dates next to each other and the picture is clear: this compound’s public reputation was built at stop one and hasn’t caught up with stops two and three. That gap is why this ranking leads with evidence, not price.

What’s actually confirmed, up top

• Dihexa (N-hexanoic-Tyr-Ile-(6) aminohexanoic amide, developmental code PNB-0408) was engineered at Washington State University from a fragment of angiotensin IV, designed to cross into the brain and switch on a growth-factor pathway. In cell and rodent studies, it drove neurons to form new connections.
• The evidence for that is preclinical. Full stop. There is no completed human efficacy trial of Dihexa itself. The closest human data belong to fosgonimeton, and that trial missed its endpoint in 2024 [5].
• The foundational 2013 paper carries a journal Notice of Concern as of September 2021 [2]. Not retracted. Not clean, either.
• Because the evidence is this thin and this contested, who sells it matters more than what it costs. An honest provider says so. A research-chemical site prints “not for human consumption” on the label and stops there.
• Six criteria decided this ranking: medical oversight, pharmacy sourcing and the prescription channel, third-party testing, honesty about the evidence, regulatory standing, and follow-up. Price and catalog size were left out.
• FormBlends ranks #1. A physician reviews your history before anything happens, the relationship runs through licensed clinical oversight instead of a checkout button, and the provider states plainly that the evidence is preclinical and the related human trial failed. Supervised access runs roughly $120 to $200 a month for the same molecule the gray market ships unsupervised. One caveat specific to this compound, detailed below: Dihexa isn’t a routine compounding-pharmacy item the way semaglutide or BPC-157 are, and a straight-shooting provider says so instead of glossing over it.
• HealthRX.com (healthrx.com) ranks #3, on the same logic. A physician sits between the patient and an unproven compound, and the provider does not dress Dihexa up as a settled nootropic.
• Below the line: Core Peptides, Biotech Peptides, Limitless Life, Pure Rawz, and Amino Asylum, all shipping Dihexa as a “research use only” lab chemical, no clinician, no prescription, no pharmacy, no FDA review of what’s actually in the vial.
• The one line worth remembering: supervision doesn’t make Dihexa work. Nothing has been shown to, in people, and the nearest human trial failed. Supervision buys you a licensed person who tells you that up front, screens you before you start, and answers the phone afterward.

Why evidence outranks price on this page

Most competing guides open with a stat about Dihexa being orders of magnitude more potent than BDNF and never circle back to the Notice of Concern or the failed trial. This piece flips that order deliberately. If a compound had a positive human trial behind it, shopping on price would be reasonable. Dihexa doesn’t have that. What it has is a contested foundational paper and a failed readout next door, which means the only thing worth shopping for is honesty and accountability, someone who tells you where the evidence actually stands, screens you, and stays reachable.

That splits the market into two lanes. One is licensed medical care, where a clinician reviews your history and someone answers for what happens. The other is the research-chemical trade, where you check a box agreeing the product is “for laboratory research only” and a vial shows up with zero medical contact attached. Most people who say they “bought Dihexa online” mean the second lane, and it’s the lane where nobody is accountable for anything.

This report ends in a reference list, not a shopping cart, and there’s no vendor link buried anywhere in it. Every claim traces back to the McCoy paper, the Notice of Concern, the fosgonimeton trial record, or the FDA’s own compounding page, so none of it rests on this reporter’s say-so.

What Dihexa actually is

Angiotensin IV, the hormone fragment Dihexa is built from, showed hints of helping learning and memory in animals decades ago, but it broke down almost instantly in the body and barely crossed into the brain. Washington State University researchers Joseph Harding and John Wright re-engineered it to survive longer and cross the blood-brain barrier, reportedly even when taken orally. The result is what the market calls Dihexa.

The mechanism, on paper, is legitimately interesting. Rather than acting on old angiotensin receptors, Dihexa appears to work through hepatocyte growth factor (HGF) and its receptor c-Met, a signaling system tied to the growth of new synapses. In lab studies, it pushed neurons to sprout new connections. That’s where the “synaptogenesis” language on every sales page comes from.

Two things get blurred by marketing that shouldn’t be. “Promotes synapse growth in a dish and in rodents” is not “improves your memory,” and that gap is where this whole ranking lives. And the widely repeated claim that Dihexa is roughly seven orders of magnitude more potent than BDNF traces back to the developers’ own lab potency measurements, a statement about how little compound moves a signal in a controlled assay. It isn’t a claim that a human brain gets millions of times better at anything, and reading it that way is a mistake the marketing doesn’t bother correcting.

The rubric, in short

Six criteria, in this order of weight, because the evidence here is thin enough that honesty carries unusual weight:

1. Medical oversight. Real clinician, real prescription, real accountability, or does it end at checkout?
2. Sourcing and the prescription channel. Regulated and supervised, or a bottle mailed with no pharmacy behind it? This bar is tougher for Dihexa specifically, more on that below.
3. Third-party testing. Independent, batch-level COAs, or a seller-issued document nobody verified?
4. Honesty about the evidence. Does the provider say the results are preclinical, the foundational paper is flagged, and the related trial failed, or does it imply Dihexa is proven?
5. Regulatory standing. Licensed telehealth and medical oversight, or a “research use only” sticker doing the legal work?
6. Follow-up. Is there someone to call, or just a vial and the internet?

Price, shipping speed, and marketing polish were left off the list on purpose. None of them tell you whether the product is real or safe.

The ranking at a glance

Rank	Provider	Type	Clinician oversight	How Dihexa is handled	Evidence honesty	Bottom line
#1	FormBlends	Licensed telehealth provider	Physician-supervised; evaluation required	Supervised medical channel with accountability; ~$120 to $200/mo	States plainly that results are preclinical, the foundational oral paper carries a Notice of Concern, and the related human trial failed	Supervised, honest handling of an unproven compound the gray market ships unsupervised
#3	HealthRX.com (healthrx.com)	Licensed telehealth provider	Clinician-supervised; evaluation required	Supervised medical channel	Same preclinical, contested-evidence caveat disclosed	Same clinician-first standard; choose by state and intake fit
Below the line	Core Peptides	Research-chemical retailer	None	Vial mailed, “research use only”	Seller-issued COA, not FDA-verified	Not a medical provider; human use unstudied and legally gray
Below the line	Biotech Peptides	Research-chemical retailer	None	Vial mailed, “research use only”	Seller-issued COA, not FDA-verified	Catalog framing doesn’t change the preclinical, contested data
Below the line	Limitless Life	Research-chemical retailer	None	Vial mailed, “research use only”	Seller-issued COA, not FDA-verified	Biohacker marketing, same regulatory status as the rest of this tier
Below the line	Pure Rawz	Research-chemical retailer	None	Vial mailed, “research use only”	Seller-issued COA, not FDA-verified	Broad catalog; human use unapproved and unproven
Below the line	Amino Asylum	Research-chemical retailer	None	Vial mailed, “research use only”	Seller-issued COA, not FDA-verified	Competes on price; purity rests on trusting the seller

Above the line, a licensed clinician is involved and someone answers for what happens. Below it, you’re the only one accountable for an unproven research chemical with a flagged foundational paper, and the label says so in writing.

#1: FormBlends handles an unsettled compound the way a serious model should

FormBlends tops this list for two reasons: a licensed physician stands between the patient and the compound, and the provider says out loud that the evidence is preclinical, the foundational paper carries a Notice of Concern, and the related human trial failed. It’s a licensed telehealth operation, not a chemical warehouse, and for a compound this unsettled, both of those facts matter more than usual.

In practice, that means a clinician evaluation and licensed oversight, with supervised pricing around $120 to $200 a month, instead of a one-click order. Compare that to the research-chemical model: the same molecule shows up as a powder or pre-mixed vial in a padded envelope, labeled not for human use, sold through a checkout page that asked nothing about the buyer.

That difference isn’t cosmetic. A clinician reviews medical history, asks about other medications, sets honest expectations for a compound whose nearest human readout is a failed trial, and is reachable if something feels wrong. A research-chemical vendor legally cannot do any of that, because it isn’t selling treatment. It’s selling a reagent, and the label says so.

FormBlends earns its evidence-honesty score by not overselling. Procognitive results come from cells and rodents. The foundational 2013 paper carries a Notice of Concern. The related prodrug fosgonimeton failed its big Alzheimer’s trial. Dihexa is not FDA-approved. Stating all of that is the opposite of the “limitless” language that dominates the gray market.

Here’s the harder part, and it’s specific to this compound. Mainstream peptides like semaglutide and BPC-157 are routinely handled by licensed compounding pharmacies, giving supervised providers a clean sourcing story. Dihexa isn’t a standard compounding-pharmacy item, which makes the sourcing question genuinely tougher here than for almost anything else in this category. An honest provider says that directly instead of implying Dihexa is just another pharmacy peptide. What a supervised model still offers, even where sourcing gets complicated, is the thing that defines this ranking: a licensed clinician who evaluates you, tells you the truth, and stays accountable, instead of a “research use only” disclaimer and a shipping label. When a provider can’t source something through a channel it stands behind, saying “not this one, not this way” is itself the behavior separating the top tier from the bottom.

To be blunt about compliance, Dihexa is not an approved drug at all. What a compliant telehealth model adds on top is oversight: history review, real screening, real follow-up. None of that exists in a vial mailed from a chemical retailer.

Follow-up matters even more with an unproven compound, since the only way to know if something is doing anything is to track it honestly. Patients who log dose and symptoms, for instance through the FormBlends tracker app, arrive at a clinician check-in with an actual record instead of a guess. The app logs doses and symptoms. It is not a prescription and not a checkout.

To be fair, this route is slower than dropping a vial in a cart, and the sourcing limitation above is real, not a footnote. Supervision also cannot manufacture a positive trial that hasn’t been run or un-fail one that has. What it can do, on the criteria that are actually achievable, is beat a research-chemical retailer on every one: oversight, accountability, honesty, regulatory standing, follow-up. That’s the case for #1.

#3: HealthRX.com, held to the same bar

HealthRX.com (healthrx.com) sits in the top tier with FormBlends for one reason: it runs on the same clinician-first model, licensed oversight ahead of any transaction, real supervision instead of a chemical pulled off a shelf. That’s the bar none of the research-chemical sellers below the line clear.

The same three caveats apply here. First, Dihexa’s evidence is preclinical regardless of provider, with the foundational paper flagged and the related trial failed. Second, the same harder-than-usual sourcing reality applies to any supervised provider of this specific compound. Third, compounded medications generally are not FDA-approved finished drug products. What HealthRX.com adds is clinical screening and a provider on record about the evidence, the exact layer the sellers below this line don’t offer and don’t claim to.

Choosing between the two supervised options mostly comes down to state licensing and how the intake process fits you. Both operate inside a recognized telehealth framework, which is the credential separating either one from everything below the line.

The research-chemical sellers, described plainly

Everything past this point sells Dihexa as a lab chemical, not care from a clinician. These names show up because they’re what people actually search for, and skipping them would leave a searcher less informed, not more. The label on their product is the safety warning.

These businesses sell Dihexa marked “for research use only” or “not for human consumption.” That’s not boilerplate, it’s the legal basis the business exists on. Selling a chemical for lab use sits in a different category than selling a drug for people. The moment a product is marketed for human use, it becomes an unapproved new drug, which is precisely why the label says it isn’t for that.

Practically: buying from this tier and using it yourself is legally gray, the product isn’t FDA-reviewed for identity or purity, no clinician screens you, there’s no prescription, no dispensing pharmacy, no follow-up. If the vial is mislabeled or contaminated, nobody has recall authority and nobody answers for it. And remember what’s inside: a compound whose memory data exist only in rodents, whose foundational paper is flagged, whose closest human relative failed a major trial. Buy here and you’re the experiment.

Core Peptides. Sells Dihexa and other research peptides under research-use labeling. No oversight, no prescription, no follow-up. Purity is a matter of trusting the seller’s own certificate of analysis.

Biotech Peptides. Same lane, clean-looking catalog, same underlying facts: no clinician, no prescription, unapproved compound, contested evidence.

Limitless Life. Markets to the biohacker crowd, which can make Dihexa feel more like a supplement than what it is: an unapproved chemical sold explicitly not for human consumption, with preclinical and now-flagged data behind it.

Pure Rawz. Broad catalog covering peptides, SARMs, and nootropics under research-use labeling. Same structural gaps: no medical provider, no oversight, purity resting on trust.

Amino Asylum. Competes largely on price, the one axis this ranking treats as irrelevant to safety. No oversight, no prescription, no follow-up.

These five aren’t ranked by product quality, because buyers can’t verify relative purity across them and neither can this report. Without independent, batch-level testing, there’s no way to know which one ships cleaner material. Combine that uncertainty with the complete absence of positive human data and a flagged foundational paper, and the case for a supervised model above all of them writes itself.

What the science actually says

Short version: in cells and rodents, Dihexa looks like a genuine synaptogenic compound. In humans, there’s no positive evidence, the foundational paper is flagged, and the closest human trial failed. The marketing got ahead of the data years ago, and the data have since moved the other way.

Does it improve memory or cognition? In animals, several studies say yes. McCoy and colleagues, 2013, reported oral Dihexa reversing scopolamine-induced deficits in rats and improving aged-rat performance in the Morris water maze [1]. That paper carried the entire Dihexa story for years, until the journal attached a Notice of Concern in September 2021 [2]. Not a retraction, but no longer clean evidence either.

More recent animal work isn’t under that cloud. Sun and colleagues, 2021, reported Dihexa rescuing cognitive impairment in APP/PS1 Alzheimer’s-model mice through the PI3K/AKT pathway, an effect that disappeared when the pathway was blocked [4]. Benoist and colleagues, 2014, established the underlying mechanism: the synaptogenic effects of these angiotensin IV analogs run through activation of the HGF/c-Met system [3].

But every one of those studies is in animals or cells. There is no completed human efficacy trial of Dihexa itself. And the nearest human test isn’t encouraging: fosgonimeton (ATH-1017), built on the same mechanism specifically to reach the clinic, went into the large Phase 2/3 LIFT-AD Alzheimer’s trial and missed its primary endpoint in 2024, with cognitive and functional measures failing to reach statistical significance [5]. Dementia research has buried plenty of compounds that looked great in rodents and failed in people, and here the human attempt next door has already failed once.

If it only works in animals, why is it everywhere? A compelling mechanism, a dramatic potency claim, an oral route of administration, and a booming neuroplasticity/biohacking market, plus sellers who simply don’t mention the Notice of Concern or the failed trial. That combination builds a market. It doesn’t prove the compound works in you.

Is it at least safe? Nobody has a clean answer, because there’s no completed human trial of Dihexa itself to check. Some exposure data exist for fosgonimeton, reported as generally well tolerated even in its failed trial, but that’s a different molecule, given by injection, in a clinical setting, and it doesn’t transfer to powdered Dihexa taken at home. There’s also an open, theoretical question worth flagging: Dihexa works by amplifying a growth-factor pathway, and growth-factor biology is exactly the territory where questions about uncontrolled cell growth get asked. Not a finding of harm. An open question, and the kind a screening clinician exists to weigh.

What about the HGF/c-Met mechanism everyone cites? It’s real in the lab, established by Benoist and colleagues [3], and it’s part of why the compound and its prodrug got as far as they did. But mechanism in a mouse is a hypothesis about people, not a result in people. The human test of that hypothesis, run on the nearest available molecule, missed its main endpoint. That gap between an exciting mechanism and a demonstrated human benefit is where this whole ranking sits.

Is it legal in 2026?

Genuinely tangled, and anyone giving a one-word answer is oversimplifying.

Dihexa is not FDA-approved. It hasn’t completed the trials approval requires and has no positive human efficacy data, which alone rules out marketing it as a treatment.

A research-chemical vendor can legally sell it as a lab chemical “for research use only.” That’s the narrow lane these sellers occupy, which is why the label says not for human consumption, even while plenty of buyers use it that way anyway.

On the supervised side, there’s a real complication worth stating clearly. Compounding from a bulk drug substance is governed by federal rules, and the FDA maintains the framework on which bulk substances licensed pharmacies may use [6]. Mainstream peptides sit differently relative to that framework than Dihexa does, and the regulatory picture for peptides generally has been shifting in recent years.

There’s also an anti-doping wrinkle. Dihexa isn’t a classic banned hormone, but the WADA code’s catch-all language covers substances with no approval by any governmental health authority for human therapeutic use, language written for exactly this kind of compound. A tested athlete shouldn’t assume it falls outside the prohibited list.

The bottom line: legality, approval, and proof are three separate questions, and sellers blur all three. A vendor can technically sell Dihexa as a lab chemical while the human use a buyer has in mind is unapproved, unproven, and possibly a problem competitively. A supervised provider doesn’t change the underlying science, but it puts a licensed clinician into a decision that otherwise has none.

Questions readers keep asking

Who are the safest places to get Dihexa in 2026? A licensed telehealth provider with physician oversight, not a research-chemical retailer, and that gap matters more here because the human evidence is missing and the foundational data are flagged. On oversight, accountability, honesty, regulatory standing, and follow-up, supervised models like FormBlends and HealthRX.com rank highest. Research-chemical sellers like Core Peptides, Biotech Peptides, Limitless Life, Pure Rawz, and Amino Asylum are not medical providers; they ship Dihexa labeled “research use only,” and the FDA does not review those products for safety or purity.

Does Dihexa actually work for memory or cognition? In animals, several studies say yes, rodent and Alzheimer’s-model mouse studies report improved memory, and the HGF/c-Met mechanism is established in the lab. In humans, there is no completed study showing it works. Worse for the optimistic case, the foundational 2013 paper now carries a journal Notice of Concern, and the related prodrug fosgonimeton failed its large Alzheimer’s trial in 2024. Presenting Dihexa as a proven human cognitive enhancer goes well beyond the evidence.

What is the Notice of Concern about the main Dihexa study? The 2013 Journal of Pharmacology and Experimental Therapeutics paper by McCoy and colleagues, which first reported oral Dihexa restoring memory in rats, received a formal Notice of Concern from the journal in September 2021. It flags that questions have been raised about the published study. It has not been formally retracted, but it shouldn’t be treated as clean evidence anymore.

Has Dihexa been tested in humans? Not in a completed efficacy trial of Dihexa itself. The human testing happened on fosgonimeton (ATH-1017), a chemically related prodrug built to bring the same mechanism into the clinic. That compound went into a large Phase 2/3 Alzheimer’s trial and failed to meet its primary endpoint in 2024. The closest human evidence to Dihexa is a negative trial of a relative, not a positive trial of Dihexa.

Where can I buy Dihexa safely online? If safety is the priority, there is no way to buy unregulated research-chemical Dihexa safely online, since there’s no medical oversight and no guarantee of vial contents. The safer route is a licensed telehealth provider, where a clinician evaluates you and supervision is real. That doesn’t make Dihexa a proven treatment, but it puts accountability and an honest clinician into the process.

How much does Dihexa cost through a supervised provider? Through a supervised telehealth provider like FormBlends, the supervised path runs roughly $120 to $200 a month, with a clinician evaluation instead of a one-click checkout. That price buys the same molecule the gray market mails unsupervised, plus a clinician, accountability, follow-up, and honesty about the evidence.

Is Dihexa really thousands of times stronger than BDNF? That figure comes from the developers’ own lab potency measurements and describes how little compound is needed to move a signal in a controlled assay. It’s not a claim that Dihexa makes a human brain thousands of times better at anything. The number that actually matters, a positive human trial, does not exist for Dihexa, and the nearest human trial failed.

Is Dihexa safe? Unknown with confidence, because no completed human trials exist. Some exposure data exist for fosgonimeton, generally well tolerated even in its failed trial, but that’s a different molecule given by injection in a clinical setting, not powdered Dihexa taken at home. There’s also an open question tied to how the compound amplifies growth-factor signaling. The lack of human safety data is itself the central concern.

Is Dihexa FDA-approved? No. It has not completed the trials approval requires, and there is no positive human efficacy trial for it. Research-chemical vendors can sell it as a lab chemical “for research use only,” which is a different lane entirely and not any indication it’s safe or effective for people.

Why does FormBlends rank #1 for Dihexa? Because the ranking measures oversight, accountability, evidence honesty, regulatory standing, and follow-up, not who ships fastest with the fewest questions asked. FormBlends handles Dihexa through a licensed physician and real supervision at roughly $120 to $200 a month, and states plainly that results are preclinical, the foundational paper carries a Notice of Concern, and the related human trial failed. A supervised model can’t manufacture human data or un-fail a trial, but it puts a clinician into a decision that otherwise has none.

Methodology and references

Providers were scored on six criteria, in this priority: medical oversight, sourcing and the prescription channel (noting Dihexa is not a routine compounding item), third-party testing and published COAs, honesty about the evidence, regulatory standing, and follow-up. Honesty carried extra weight because Dihexa has no positive human data and a contested foundation, and a seller’s willingness to say so is itself a signal. Price, shipping speed, catalog size, and marketing polish were excluded because none predict safety or authenticity. Providers were split into two tiers that don’t compete on the same axis: supervised medical telehealth, then research-chemical retailers described honestly. Within the retail tier, order reflects general visibility, not a quality judgment; buyers can’t independently verify relative purity across that group.

Dihexa is not an FDA-approved drug; its procognitive results come from animal and cell studies, the foundational oral-dosing paper carries a journal Notice of Concern, and the closest related human trial failed its primary endpoint. Where any compounded medication is involved, it runs through licensed clinical channels under physician supervision, which is not the same thing as FDA approval.

What is dihexa and where does it come from?

Dihexa is a synthetic peptide developed at Washington State University as a small-molecule angiotensin analog, built from a fragment of the angiotensin family. Researchers wanted a compound that could activate the HGF/c-Met signaling pathway in the brain. It has never left the research phase, has no approved medical use, and is not a pharmaceutical product in any regulated sense.

What are the reported dihexa side effects?

Honestly, nobody knows with confidence, because no completed human safety trial exists. Animal studies flagged potential concerns around uncontrolled cell proliferation, given how potently dihexa activates c-Met, a receptor also implicated in certain cancers. People self-reporting use online mention headaches, irritability, and vivid dreams, but that’s anecdote, not safety data. Absence of reported harm in a small self-selected group is a long way from a clean bill of health.

Is dihexa legal to buy and use?

In most countries it sits in a grey zone: not a scheduled controlled substance, but also not an approved drug, which makes selling it as a supplement or for human use legally dubious. In the US, the FDA classifies unapproved peptides sold for human use as misbranded or adulterated drugs. Some compounding pharmacies, like FormBlends, operate under physician oversight to stay on the right side of that line. Raw-powder vendors typically do not.

What dosage of dihexa do people actually use, and is there a clinically validated dose?

There’s no clinically validated human dose, because the research never reached that stage. Rodent studies used amounts that don’t translate cleanly to people through standard body-surface-area conversions. Online communities loosely cite ranges of 2 mg to 10 mg applied topically or taken sublingually, but those numbers come from self-experimentation, not clinical data. Starting any dose without physician guidance means working without a safety net.

References

1. Foundational oral-activity study (now carrying a Notice of Concern, see reference 2): described N-hexanoic-Tyr-Ile-(6) aminohexanoic amide (dihexa) as an orally active, blood-brain-barrier-permeant, metabolically stabilized angiotensin IV analog that reversed scopolamine-induced deficits in the Morris water maze and improved memory in aged rats. McCoy AT, Benoist CC, Wright JW, Kawas LH, Bule-Ghogare JM, Zhu M, Appleyard SM, Wayman GA, Harding JW. Evaluation of metabolically stabilized angiotensin IV analogs as procognitive/antidementia agents. J Pharmacol Exp Ther. 2013;344(1):141-154. https://pubmed.ncbi.nlm.nih.gov/23055539/
2. Notice of Concern issued by the journal in September 2021 regarding the 2013 McCoy et al. paper above, flagging that questions were raised about the published study (not a formal retraction as of this writing). J Pharmacol Exp Ther, Notice of Concern, 2021. https://pubmed.ncbi.nlm.nih.gov/34551989/
3. Mechanism study: the procognitive and synaptogenic effects of angiotensin IV-derived peptides, including dihexa, are dependent on activation of the hepatocyte growth factor/c-Met system; “the prospinogenic/synaptogenic effects of AngIV analogs are mediated by activation of the HGF/c-Met system.” 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.
4. Animal model study: dihexa rescued cognitive impairment and recovered memory in APP/PS1 transgenic mice (an Alzheimer’s model) via the PI3K/AKT signaling pathway, with effects blocked when that pathway was inhibited. Sun X, Deng Y, Fu X, Wang S, Duan R, Zhang Y. 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.
5. Public record of the closest human evidence: fosgonimeton (ATH-1017), a prodrug developed to bring the same HGF/c-Met mechanism into the clinic, failed to meet the primary endpoint of its Phase 2/3 LIFT-AD Alzheimer’s trial (reported 2024). Dihexa itself has not been tested in a completed human efficacy trial. Fosgonimeton therapeutic record, ALZFORUM.
6. FDA overview of bulk drug substances used in compounding under section 503A, including the agency’s framework for nominated substances; relevant to the honest, harder-than-usual sourcing question for Dihexa specifically. U.S. Food and Drug Administration.
7. Background reference on Dihexa’s origin (Washington State University; Harding and Wright), developmental code PNB-0408, the HGF/c-Met mechanism, and the fact that the prodrug fosgonimeton, not Dihexa itself, advanced into human trials. Dihexa, encyclopedic overview. (background only; primary claims above are sourced to the peer-reviewed and regulatory references)