Does Dihexa Improve Memory?

What is Dihexa?

Dihexa is the common name for N-hexanoic-Tyr-Ile-(6)-aminohexanoic amide, a synthetic peptide also known by its developmental code PNB-0408 and identified by CAS number 1401708-83-5.

It is not a naturally occurring hormone or supplement ingredient but a lab-designed peptidomimetic, chemically derived from angiotensin IV, a fragment of the renin-angiotensin system best known for its role in blood pressure regulation.

Researchers Joseph Harding and John Wright, working at Washington State University, engineered dihexa specifically to retain angiotensin IV’s cognitive effects while improving its stability and delivery.

The proposed mechanism, based entirely on preclinical research, is that dihexa binds to hepatocyte growth factor (HGF) and amplifies its signaling through the c-Met receptor, a pathway linked to neuronal survival and synaptic plasticity.

This mechanistic story has shifted over time (detailed later in this article), and part of the foundational literature supporting it has since been retracted.

One of dihexa’s most-cited properties is that it appears to be orally bioavailable and able to cross the blood-brain barrier, both in animal studies.

This is considered unusual for a peptide, since peptides are typically broken down in the digestive tract or blocked from entering the brain, which is why the compound draws attention from peptide researchers and nootropic communities alike.

Marketing and research materials frequently describe dihexa as “seven orders of magnitude” or “10 million times” more potent than BDNF (brain-derived neurotrophic factor) in promoting neuronal growth.

This figure comes from in-vitro molar-potency assays, meaning it reflects how much less of the compound was needed to produce a comparable effect in a lab dish, not a measurement of clinical benefit in a living organism. Presented without that context, the claim implies a real-world effect size that has never been demonstrated.

The honest evidence picture

Every finding about dihexa’s effects on memory, learning, or neuroprotection comes from rodent studies: scopolamine-treated rats, aged rats, and Alzheimer’s-model mice. No completed human trials of dihexa exist. Some of the foundational biochemistry papers describing its HGF/c-Met mechanism have also been formally retracted, a detail covered in full later in this article.

Readers should treat any description of dihexa’s benefits as “clinically observed” with skepticism unless it explicitly distinguishes rodent data from human data, since no human clinical evidence currently exists for this compound.

how we rate evidence

What the research shows

Alzheimer’s-model spatial learning and cognitive restoration

Evidence tier: Limited. The strongest single data point in dihexa’s research base comes from an independent replication, not the original Harding lab. In APP/PS1 Alzheimer’s-model mice, dihexa restored spatial learning performance, increased neuronal density and synaptophysin expression, and shifted neuroinflammatory markers in a favorable direction (IL-1beta and TNF-alpha reduced, IL-10 increased).

Researchers confirmed the proposed PI3K/AKT mechanism by co-administering wortmannin, a PI3K inhibitor, which reversed dihexa’s effects, an important causal check that most animal studies skip. Because this work came from a separate research group in China rather than the original WSU team, it carries more weight than a simple confirmation would.

Still, this is one animal study. No human trials exist; translation from rodent to human is unconfirmed.

Age-related and scopolamine-induced memory deficits

Evidence tier: Limited. The foundational behavioral data for dihexa comes from rats: reversal of scopolamine-induced spatial memory deficits and improved learning in aged, 24-month-old animals. This is the study most commercial pages lean on to imply cognitive benefit.

What they generally omit is that this paper carries a 2021 Notice of Concern over data integrity questions that remain unresolved. No human trials exist; translation from rodent to human is unconfirmed.

General human cognitive enhancement / nootropic use

Evidence tier: Emerging. This is the use case driving most consumer searches, and it is also the one with the least direct support. There is no dihexa study in humans of any kind, no dosing trial, no safety trial, no cognitive-outcome trial. Every nootropic claim in circulation is extrapolated from the animal findings above. No human trials exist; translation from rodent to human is unconfirmed.

Huntington’s disease neuroprotection

Evidence tier: Emerging. A 2024 rat study using 3-nitropropionic acid to induce Huntington’s-like neurotoxic damage found dihexa did not protect against the resulting deficits, a negative result that runs counter to the compound’s broader neuroprotective narrative. No human trials exist; translation from rodent to human is unconfirmed.

Peripheral nerve regeneration

Evidence tier: Emerging. A 2021 study examining dihexa for nerve regeneration produced inconclusive findings, with no clear benefit demonstrated. This result is rarely mentioned in marketing material but belongs in any complete accounting of the evidence. No human trials exist; translation from rodent to human is unconfirmed.

The Retracted Research Behind Dihexa's Mechanism

How the mechanism story changed over time

Before dihexa was framed as an HGF/c-Met compound, researchers believed angiotensin IV worked through a different target entirely: the AT4 receptor, later identified as insulin-regulated aminopeptidase (IRAP). This theory ran into a paradox.

In animal studies, mice with the IRAP gene knocked out did not show improved memory, as the theory might predict, but instead showed worse age-related memory decline.

That inconsistency helped push the Harding lab toward a new explanation, one in which angiotensin IV analogs like dihexa work by binding hepatocyte growth factor (HGF) and amplifying its signaling through the c-Met receptor.

Two papers anchored this shift: Kawas et al. (2012), which proposed the HGF-dimerization mechanism, and Benoist et al. (2014), which reported that dihexa’s procognitive and synaptogenic effects depended on HGF/c-Met activation.

What changed in 2025

Both papers were formally retracted in April 2025, following Notices of Concern first issued in 2021. Reported concerns center on image manipulation within the published data.

McCoy et al. (2013), the foundational rat behavioral study describing dihexa’s reversal of scopolamine-induced memory deficits, was also flagged with a Notice of Concern in 2021, but as of the retractions it remains unretracted.

This distinction matters: a Notice of Concern signals unresolved questions under review, while a retraction is a formal withdrawal of the paper’s findings from the scientific record.

What still stands

Not all of the underlying research is affected. Sun et al. (2021), the independent APP/PS1 mouse replication conducted outside the Harding lab, is not implicated in the retractions and remains part of the evidence base supporting a PI3K/AKT-linked mechanism.

What this means in practice

The retractions do not erase the rodent behavioral findings themselves, animals in these studies did show measurable improvements in maze performance and memory tasks. What they remove is much of the molecular explanation for why those improvements occurred.

The specific claim that dihexa works by binding HGF and potentiating c-Met signaling now rests on thinner, partially withdrawn evidence than most commercial and clinic-facing sources acknowledge.

Readers evaluating dihexa should treat the HGF/c-Met mechanism as a proposed, weakened explanation rather than an established fact, even where the underlying behavioral data remains part of the record.

Dihexa vs. Fosgonimeton: Where Is the Human Data?

The pro-drug that made it further

Dihexa itself has never been tested in a human clinical trial. What has advanced into human testing is fosgonimeton, a phosphate pro-drug of dihexa developed by Athira Pharma, which has been studied in human trials for Alzheimer’s disease and Parkinson’s disease.

A pro-drug is a chemically modified version of a molecule designed to convert into the active compound once inside the body, so fosgonimeton is not a separate drug with its own unrelated mechanism, it is engineered to deliver dihexa’s proposed activity through a different formulation.

Why that distinction matters

This creates a situation where the actual human proof-of-concept for the HGF/c-Met mechanism, if the mechanism holds up at all, is being tested in fosgonimeton’s trials rather than in dihexa.

Any positive or negative signal from those trials speaks to fosgonimeton’s performance in humans, not to dihexa’s, since dosing, pharmacokinetics, and conversion to active compound in the body can all differ between a parent molecule and its pro-drug.

The takeaway for readers

Marketing or forum claims that reference fosgonimeton’s clinical trials as evidence that “dihexa is in human trials” are conflating two distinct compounds. As it stands, dihexa carries zero human trials of its own, while its derivative carries an active human testing program.

Readers evaluating dihexa’s evidence base should treat fosgonimeton’s trial data as relevant context about the underlying mechanism’s plausibility, not as human evidence for dihexa itself.

The History and Development of dihexa

Dihexa’s origin traces to Washington State University, where researchers Joseph Harding and John Wright led a research program investigating angiotensin IV analogs for cognitive effects.

The compound was designed as a metabolically stabilized derivative in this lineage, building on decades of earlier angiotensin-system research that had proposed the AT4/IRAP receptor as the relevant mechanism before the field’s later pivot toward HGF/c-Met signaling (detailed in this article’s retraction section).

Patents and early commercialization

The research produced two notable patent filings. US Patent 8598118, covering “hepatocyte growth factor mimics as therapeutic agents” and naming Harding, Wright, Benoist, Kawas, and Wayman as inventors, was issued December 3, 2013. A separate patent application covering dihexa’s use as an otoprotection lead compound followed in 2015.

Around this period, M3 Biotechnology was founded to advance dihexa commercially, and Wikipedia’s entry on the compound was created in October 2015, reflecting its growing visibility in both scientific and biohacker circles.

Publication timeline

The core publication record runs as follows: Kawas et al. published the original HGF-dimerization mimetic paper in 2012; McCoy et al.’s procognitive rat study and the foundational patent both arrived in 2013; Benoist et al. published the HGF/c-Met mechanism paper in 2014.

By 2021, Notices of Concern had been issued for the Kawas, Benoist, and McCoy papers, the same year Sun et al. published an independent APP/PS1 mouse replication unconnected to the original WSU lab. A negative-result Huntington’s disease model study followed in 2024.

In April 2025, the Kawas 2012 and Benoist 2014 papers were formally retracted, and a market surveillance study (Vanhee et al.) examining nootropic product quality was published the same year.

The successor compound

Separately from dihexa’s own trajectory, Athira Pharma later developed fosgonimeton, a phosphate pro-drug of dihexa, and advanced it into human clinical trials for Alzheimer’s and Parkinson’s disease, a distinction covered in more depth elsewhere in this article.

Dosage and administration

Animal dosing data below describes what researchers administered to rodents in laboratory settings. It is not a human protocol, and no established human pharmacokinetic data exists for dihexa.

Clinic sources handle dosing differently from the laboratory studies above. Cenegenics lists oral capsule and topical or transdermal compounded formats but does not publish a specific dose.

AgeRejuvenation goes further, stating explicitly that it does not publish dosing information because none is established for humans, and frames any protocol decision as individualized, made between a provider and patient rather than following a standardized regimen.

Neither clinic source cites human pharmacokinetic data to justify a given dose or frequency, because none appears to exist in the research record.

The gap between rodent milligram-per-kilogram dosing and undisclosed clinic protocols reflects the absence of human trials rather than an oversight, and readers should not treat the animal figures above as a template for anything.

Safety and side effects

Because no completed human trials of dihexa exist, there is no established human adverse-event profile. Every safety signal in the record so far comes from animal studies or from the patent filing itself, not from monitored human use.

What the patent claims, and why that’s not the same as safety data

The dihexa patent (Harding, Wright, Benoist, Kawas, Wayman; issued 2013) states there was “no apparent toxicity” and a “lack of neoplastic induction” in the animal work underlying the filing.

This is worth noting because it is often the closest thing to a safety statement available on the compound, but it is a patent-holder’s self-reported claim, made by the same team that developed and stood to commercialize the molecule, not an independent toxicology study or a peer-reviewed safety trial. It should be read as a legal filing, not as evidence.

An unresolved theoretical concern

Dihexa’s proposed mechanism involves potentiating c-Met receptor signaling. c-Met is a well-characterized oncogene in other biological contexts, meaning its overactivation is linked to tumor growth in some settings.

Whether sustained activation of the HGF/c-Met pathway carries any long-term cancer-related risk in the context of dihexa use has not been studied and remains an open, unresolved question rather than a documented finding either way.

The overall safety picture

Cenegenics, AgeRejuvenation, and Superpower each state plainly that long-term human safety for dihexa is unestablished. None can point to monitored human dosing data, and none offer a documented human side-effect list.

Product-quality risk for anyone sourcing it outside a clinic

Separate from biological safety is the question of what is actually in a given vial or capsule. Vanhee et al. (2025) surveyed nootropic and “smart drug” products from illegal/gray-market sources and found that 69% of sampled products failed basic quality or labeling standards.

This is not dihexa-specific data, but it directly bears on anyone considering research-use-only or unsupervised gray-market product, since mislabeling or contamination risk exists independent of whatever is known or unknown about the molecule itself.

Who should not use Dihexa

Legal status

No, dihexa is not FDA-approved, and there is no Investigational New Drug (IND) application or New Drug Application (NDA) on record for it. That single fact underlies everything else in this section: every legal pathway for human use in the United States remains closed.

Chemical vendors sell dihexa exclusively as a "research use only" product, with labeling that explicitly states it is not intended for human consumption. That framing is a real regulatory line, not boilerplate legal cover. Buying dihexa from a research-chemical supplier is buying a laboratory reagent, not a medication.

Clinics that offer dihexa as part of compounded peptide therapy occupy murkier ground. Compounding pharmacies typically operate under 503A or 503B frameworks, but no source confirms dihexa holds a recognized status under either pathway.

Clinic marketing pages rarely address this directly, focusing instead on "physician supervision" language that sidesteps the underlying question of whether the compounding itself sits on solid legal footing.

A timeline worth tracking

The regulatory and evidentiary picture has shifted meaningfully in the past few years, and it may keep shifting:

• 2013: The foundational US patent covering dihexa's HGF-mimetic mechanism was issued.
• 2021: Notices of Concern were issued for the papers underlying dihexa's mechanism.
• April 2025: Two of those foundational papers were formally retracted.
• As of the most recent check, no dihexa human trials are registered on ClinicalTrials.gov or indexed in PubMed.

Anyone researching dihexa's legal status should treat it as time-sensitive rather than settled.

The gray-market quality problem

Legal ambiguity compounds a separate, practical risk. Vanhee et al. (2025) surveyed nootropic and "smart drug" products sourced from unregulated markets and found that 69% of samples failed basic quality or labeling standards.

That figure applies to the broader gray-market nootropic category dihexa sits within, and it means that even setting the legal question aside, sourcing dihexa outside a clinical or regulated research setting carries a real chance of receiving a product that does not match its label.

What would actually change this

Legitimate, approved human access to dihexa's mechanism would most plausibly require one of three things: completed human trials (fosgonimeton, dihexa's prodrug, is the compound actually in that pipeline), a formal FDA review process for dihexa itself, or explicit compounding-pathway clarification confirming its status under 503A or 503B rules. None of the three exists today.

Often used with

At least one clinic-marketing source lists dihexa alongside a cluster of other peptides often offered under the same "cognitive peptide therapy" umbrella: Semax, Selank, PE 22-28, methylene blue, FGL, and DSIP. These names appear in clinic content as comparator or companion offerings within a broader peptide-therapy category, not as compounds studied together with dihexa in any trial, rodent or human. Why this list is informational only None of the research reviewed for this article includes ind

Frequently asked questions