Dihexa: Cognitive Peptide Research Profile
In 2012, a small lab at Washington State University [published a paper](https://pubmed.ncbi.nlm.nih.gov/23055539/) that made a bold claim: researchers had created a synthetic peptide that was 10 million times more potent than brain-derived neurotrophic factor (BDNF) at promoting the formation of new
In 2012, a small lab at Washington State University published a paper that made a bold claim: researchers had created a synthetic peptide that was 10 million times more potent than brain-derived neurotrophic factor (BDNF) at promoting the formation of new synaptic connections. The peptide was called Dihexa, and if the data held up, it represented one of the most powerful neurotrophic compounds ever tested.
The claim caught the attention of the nootropic community, Alzheimer's researchers, and biotech investors alike. A company was spun out of the university. Clinical trials were planned. Then came the data integrity scandal, CEO resignation, failed Phase 2/3 trials, and expressions of concern on the foundational papers. The Dihexa story is one of genuine scientific promise tangled with serious credibility questions — and it is far from resolved.
Here is what the published research actually shows, what has been called into question, and what anyone considering this compound should understand.
Table of Contents
- Quick Facts
- What Is Dihexa?
- Development History
- How Dihexa Works: Mechanisms of Action
- Research Evidence
- Administration and Dosing
- Safety Profile and Side Effects
- Legal and Regulatory Status
- Limitations of Current Research
- Frequently Asked Questions
- The Bottom Line
Quick Facts
| Property | Detail |
|---|---|
| Full Chemical Name | N-hexanoic-Tyr-Ile-(6) aminohexanoic amide |
| Developmental Code | PNB-0408 |
| Molecular Formula | C₂₇H₄₄N₄O₅ |
| Molecular Weight | 504.67 g/mol |
| CAS Number | 1401708-83-5 |
| Peptide Type | Synthetic oligopeptide (angiotensin IV analog) |
| Primary Target | Hepatocyte growth factor (HGF) / c-Met receptor |
| Route of Administration | Oral (in animal studies); also injectable |
| Blood-Brain Barrier | Permeable (confirmed in rodent models) |
| Half-Life (rat, IV) | ~12.7 days |
| FDA Status | Not approved for any indication |
| Human Clinical Trials | None for Dihexa itself; pro-drug fosgonimeton tested in Phase 2/3 |
| Research Stage | Preclinical (animal and in vitro only) |
What Is Dihexa?
Dihexa is a small synthetic peptide — technically an oligopeptide — designed to mimic and amplify the activity of angiotensin IV, a fragment of the renin-angiotensin system with known effects on memory and cognition.
Unlike angiotensin IV itself, which breaks down quickly and cannot cross the blood-brain barrier, Dihexa was engineered for stability. Its structure includes an N-hexanoyl cap on a tyrosine-isoleucine dipeptide core, with a 6-aminohexanoic acid amide tail. These modifications give it three properties that natural angiotensin IV lacks: resistance to enzymatic breakdown, blood-brain barrier penetration, and oral bioavailability.
Dihexa does not work through traditional angiotensin receptors. Research from the Harding laboratory at Washington State University identified its primary target as hepatocyte growth factor (HGF), a protein that — when it binds to the c-Met receptor — triggers downstream signaling involved in cell survival, growth, and synapse formation.
This puts Dihexa in a different category from most nootropics. Rather than temporarily adjusting neurotransmitter levels (as racetams or stimulants do), Dihexa aims to structurally change the brain by promoting synaptogenesis — the physical creation of new connections between neurons. That distinction generates both the excitement and the concern.
For readers interested in how Dihexa compares to other peptides being studied for cognitive applications, our guides on Semax, Selank, and PE-22-28 cover peptides with different mechanisms targeting brain function.
Development History
Dihexa emerged from decades of work on the brain's renin-angiotensin system, led primarily by Joseph Harding, PhD, and Jay Wright, PhD, at Washington State University in Pullman, Washington.
The backstory starts in the 1980s and 1990s, when researchers discovered that angiotensin IV — a short peptide fragment produced when the body processes angiotensin II (the blood-pressure hormone) — had unexpected effects on memory. Animals given angiotensin IV performed better on learning tasks. But the peptide was impractical as a drug: it degraded in minutes, could not cross the blood-brain barrier, and required direct brain injection.
Harding's lab set out to build a better version. Working from the three N-terminal amino acids of angiotensin IV (Nle-Tyr-Ile), which they identified as the minimum unit needed for cognitive activity, the team modified the molecule's ends to block enzymatic degradation and increase fat solubility.
The result, published in the Journal of Pharmacology and Experimental Therapeutics in January 2013, was Dihexa (McCoy et al., 2013). Oral Dihexa reversed scopolamine-induced memory deficits in rats at 2 mg/kg per day and improved performance in aged rats with natural cognitive decline. In hippocampal neuron cultures, it tripled dendritic spines — the tiny protrusions where synapses form — within five days.
A follow-up paper in 2014 (Benoist et al.) identified HGF/c-Met as the molecular target, showing that an HGF antagonist blocked Dihexa's cognitive effects in the Morris water maze test.
Based on this work, M3 Biotechnology was formed as a WSU spin-off. Leen Kawas, a former PhD student in Harding's lab, became CEO. The company later renamed itself Athira Pharma and developed fosgonimeton (ATH-1017), a phosphate pro-drug of Dihexa, which entered clinical trials for Alzheimer's disease. We will return to what happened with those trials — and the serious data integrity questions that surfaced in 2021 — in the Limitations section.
How Dihexa Works: Mechanisms of Action
The HGF/c-Met Pathway
Dihexa's proposed mechanism centers on its interaction with hepatocyte growth factor (HGF) and the c-Met receptor.
HGF is a large protein that, in the brain, promotes neuron survival, stimulates the growth of dendrites (the branching arms of neurons), and supports synaptogenesis. It does this by binding to c-Met, a receptor tyrosine kinase, which triggers intracellular signaling cascades including the PI3K/Akt and MAPK/ERK pathways.
According to the Harding lab's research, Dihexa binds to HGF with a dissociation constant (Kd) of 65 picomolar — extremely tight binding. It appears to act as an allosteric potentiator: rather than activating c-Met on its own, Dihexa amplifies HGF's activity at subthreshold concentrations. In cell-based assays, Dihexa plus low-dose HGF produced c-Met phosphorylation, cell scattering, and increased motility — effects that neither compound could produce alone.
Synaptogenesis and Spine Formation
The downstream consequence that matters most for cognition is synaptogenesis. In hippocampal neuron cultures, Dihexa increased dendritic spine density from about 15 spines per 50 micrometers of dendrite to 41 spines — a nearly three-fold increase. It also increased spine-head width, which is associated with stronger, more mature synapses.
This is a fundamentally different approach from most nootropic compounds. Where a drug like Semax modulates BDNF expression and neurotransmitter activity, Dihexa's proposed effect is structural: building new hardware rather than adjusting software.
The "10 Million Times More Potent Than BDNF" Claim
This attention-grabbing number requires context. In a neurotrophic activity assay — specifically, the ability to promote dendritic spine formation — Dihexa produced effects at picomolar concentrations, while BDNF required nanomolar concentrations to achieve comparable results. That is a seven-order-of-magnitude difference in potency within that specific assay.
This does not mean Dihexa is "10 million times better for your brain" than BDNF. BDNF is an endogenous protein with broad functions across the entire nervous system. Dihexa's potency in one assay reflects its tight binding to HGF, not comprehensive superiority over BDNF. The comparison is real but narrow, and it is frequently stripped of context in marketing material.
Connection to the Angiotensin IV System
The brain's angiotensin IV / AT4 receptor system has well-established links to cognition. Activation of this system is cerebroprotective, boosts long-term potentiation (the cellular basis of learning), and stimulates hippocampal neurogenesis. Dihexa was designed to harness these effects through a more druglike molecule, but the Harding lab ultimately concluded that the cognitive benefits operate through HGF/c-Met rather than the classical AT4 receptor.
Research Evidence
Scopolamine-Induced Memory Deficit Model (Rats)
In the 2013 McCoy et al. study, rats were treated with scopolamine (a cholinergic blocker that impairs memory) and then given Dihexa either by injection or orally. Both routes showed clear dose-response relationships. At 2 mg/kg per day orally, Dihexa-treated rats performed indistinguishably from healthy controls in the Morris water maze. On probe trials (which test memory retention), Dihexa-treated animals spent significantly more time in the target quadrant versus scopolamine-impaired controls.
Aged Rat Model
The same 2013 study also tested Dihexa in naturally aged rats. Results showed improvement in water maze performance that reached statistical significance on most test days, though variability was higher because not all aged rats exhibit cognitive decline. This is a more realistic model than scopolamine amnesia, but the data were less robust.
APP/PS1 Alzheimer's Mouse Model
A 2021 study by Sun et al., conducted independently at China Pharmaceutical University and Nanjing Medical University, tested Dihexa in APP/PS1 transgenic mice, a genetic model of Alzheimer's disease. At oral doses of 1.44 and 2.88 mg/kg, Dihexa:
- Restored spatial learning and memory in the Morris water maze
- Increased neuronal cell counts and synaptophysin (SYP) expression (a marker of synaptic density)
- Reduced activation of astrocytes and microglia (markers of neuroinflammation)
- Lowered pro-inflammatory cytokines IL-1β and TNF-α while raising anti-inflammatory IL-10
- Activated the PI3K/AKT signaling pathway (confirmed by reversal with the PI3K inhibitor wortmannin)
This study is notable because it comes from a different research group than the Harding lab, providing partial independent replication of Dihexa's cognitive effects. The full text is available on PMC.
Synaptogenesis (In Vitro)
Studies from the Harding lab showed that Dihexa increases dendritic spine density and spine-head width in hippocampal neuron cultures. A 30-minute acute application increased spines and spine-head size, and five-day treatment nearly tripled spine counts. These effects were blocked by HGF antagonists, confirming dependence on the HGF/c-Met pathway.
What About Normal Brains?
Dihexa did not improve cognitive performance in rats with normal cognition. The HGF antagonist alone also had no effect in healthy animals. This suggests that HGF/c-Met signaling is not strongly engaged during normal learning, and Dihexa's benefits may be limited to states of cognitive impairment — a significant distinction from the "limitless pill" image projected in nootropic communities.
Fosgonimeton (ATH-1017) — The Pro-Drug in Human Trials
While Dihexa itself has never been tested in humans, its phosphate pro-drug fosgonimeton entered clinical trials under Athira Pharma. These trials tested the HGF/c-Met mechanism in patients.
- Phase 1 (2017–2018): Safe and well-tolerated across dose ranges in 88 healthy volunteers and Alzheimer's patients via subcutaneous injection.
- Phase 2 ACT-AD (2020–2022): Did not meet its primary ERP P300 biomarker endpoint. A subgroup analysis suggested that acetylcholinesterase inhibitors (like donepezil) may interfere with fosgonimeton's mechanism.
- Phase 2/3 LIFT-AD (2024): The largest trial. 312 patients with mild-to-moderate Alzheimer's disease. Fosgonimeton failed to meet its primary endpoint (Global Statistical Test) or secondary endpoints (ADAS-Cog11, ADCS-ADL23) versus placebo. Some directional trends favored treatment in APOE4 carriers and patients with more advanced disease.
Athira discontinued fosgonimeton development in late 2024 and pivoted to ATH-1105, a next-generation oral HGF/MET modulator being evaluated for ALS.
The LIFT-AD failure does not necessarily invalidate Dihexa's preclinical data — fosgonimeton is a different molecule delivered by a different route to a complex patient population. But it does mean the HGF/c-Met approach has not yet demonstrated clinical benefit in humans.
For context on other neuroprotective peptides that have reached further stages of research, see our profiles on Cerebrolysin and BPC-157.
Administration and Dosing
No established human dose exists for Dihexa. All dosing information comes from animal studies or anecdotal reports. The following is provided for informational context about the research — not as a recommendation.
Animal Research Doses:
- Oral: 2 mg/kg per day in rats (the effective dose in scopolamine studies)
- Oral: 1.44–2.88 mg/kg per day in mice (the APP/PS1 Alzheimer's model)
- Intraperitoneal: 0.5 mg/kg per day in rats
Anecdotal Human Reports:
Online communities report doses ranging from 10–30 mg per day, typically in cycles of 4–8 weeks. These doses have no basis in controlled research and carry unknown risk.
Route:
Dihexa is orally bioavailable in animal models — a rare property for a peptide. Its high hydrophobicity (LogP estimated at 177.8) and small size contribute to gut absorption and blood-brain barrier penetration. Some sources also report subcutaneous injection use.
Half-Life:
In rats, Dihexa has an unusually long half-life: approximately 12.7 days following intravenous administration and 8.8 days following intraperitoneal injection. This means the compound persists in the body far longer than most peptides. If adverse effects occur, they cannot be quickly reversed by stopping the dose.
Safety Profile and Side Effects
This is where honesty matters most, because the safety data for Dihexa is thin.
What We Know
According to patent filings from the original researchers, short-duration safety studies in animals "uncovered no apparent toxicity," including no evidence of tumor formation (neoplastic induction). But these were short-term observations, not long-term safety evaluations.
The FDA has stated it "has not identified any human exposure data" for Dihexa. No controlled human safety studies have been published.
Theoretical Concerns
Cancer risk. The HGF/c-Met pathway is a recognized oncogenic signaling axis. In healthy tissue, c-Met activation promotes cell survival and growth — functions that tumors hijack. Whether Dihexa could promote tumor growth, accelerate existing cancers, or increase metastatic potential is unknown. No long-term animal cancer studies have been conducted.
Maladaptive neural wiring. Synaptogenesis is not inherently beneficial. The brain actively prunes excess synapses throughout life. Artificially stimulating new connections could produce maladaptive wiring rather than functional improvements. This concern is theoretical but entirely unaddressed by current research.
Systemic effects. HGF and c-Met are expressed throughout the body — not just the brain. Dihexa's impact on non-neural tissues has not been characterized.
Accumulation. With a half-life of nearly two weeks in rats, Dihexa could accumulate with repeated dosing in ways that are difficult to predict or reverse.
Reported Side Effects (Anecdotal Only)
Self-experimenters in online forums have reported mild headaches, anxiety, overstimulation, dizziness, and sleep disruption. None of these reports come from controlled settings, and they cannot be reliably attributed to Dihexa.
The Bottom Line on Safety
We do not know whether Dihexa is safe for humans. The absence of reported toxicity in short-term animal studies is not evidence of safety — it is evidence of limited testing. Anyone considering this compound is operating with essentially no safety net.
Legal and Regulatory Status
Dihexa is not FDA-approved for any medical condition. It is not a controlled substance in the United States.
Currently, Dihexa exists in a regulatory gray area. It can be legally sold as a "research chemical" for investigational purposes, but marketing it for human consumption would violate FDA regulations. Despite this, some clinics prescribe Dihexa off-label under physician supervision.
The FDA's broader crackdown on compounded peptides in 2024 — which targeted several commonly used research peptides — has affected availability, though Dihexa's status is less clearcut than more widely used compounds like BPC-157 or thymosin alpha-1.
In international markets, Dihexa is generally unregulated. It is available from research peptide suppliers in many countries.
Limitations of Current Research
This section is long because the limitations of Dihexa research are substantial. Anyone evaluating this compound should read it carefully.
Almost Entirely Preclinical
No human clinical trials of Dihexa have been conducted. Every claim about its cognitive effects comes from rat and mouse studies or cell cultures. Animal models of cognition have a poor track record of translating to human therapies — especially in Alzheimer's disease, where dozens of compounds that worked in mice have failed in patients.
Data Integrity Problems in Foundational Papers
This is the most serious issue. In 2021, both foundational Dihexa papers from the Harding lab received expressions of concern from the Journal of Pharmacology and Experimental Therapeutics:
- McCoy et al. (2013) — the paper establishing Dihexa's cognitive effects in rats
- Benoist et al. (2014) — the paper identifying HGF/c-Met as the molecular target
The concerns centered on western blot images that appeared to have been manipulated — portions of images appeared to have been copied and pasted from other datasets. These problems were first flagged on PubPeer, the scientific integrity platform. Four papers by the group received expressions of concern in September 2021.
A subsequent investigation found that Leen Kawas, first author on related papers and then-CEO of Athira Pharma, had altered images in her 2011 doctoral dissertation and at least four research papers. Research integrity consultants found problems in 19 out of 30 images from her dissertation. Manipulations included copying and pasting data between experiments, digitally altering western blot band intensity, and reusing identical images to represent different experimental conditions.
The 2014 Benoist et al. paper was fully retracted in April 2025 after Washington State University's investigation confirmed that figures contained "falsified and/or fabricated data," with Kawas and Harding found "solely responsible."
Kawas resigned as Athira CEO in October 2021. Athira's stock dropped 67% that year. In January 2025, Athira agreed to pay over $4 million to settle False Claims Act allegations related to NIH grants referencing the compromised research.
What this means for Dihexa: The foundational claim that Dihexa works through HGF/c-Met — and the western blot data supporting it — is now compromised. The behavioral data (water maze performance) was not directly implicated, and the 2021 Sun et al. study in APP/PS1 mice provides some independent support. But the mechanistic story is damaged.
Pro-Drug Failed in Clinical Trials
Fosgonimeton, Dihexa's pro-drug, failed to meet primary or secondary endpoints in the Phase 2/3 LIFT-AD trial in Alzheimer's patients. While this does not directly disprove Dihexa's preclinical effects, it raises questions about whether HGF/c-Met modulation produces meaningful cognitive benefits in humans.
No Long-Term Safety Data
No study — in animals or humans — has evaluated Dihexa's long-term effects. For a compound that activates a known oncogenic pathway and has a two-week half-life, this gap is significant.
No Effect in Healthy Brains
Dihexa did not improve cognition in rats with normal brain function. This is important context for anyone in the nootropic community considering it as a general cognitive booster. The available evidence suggests it may only work in states of impairment.
Very Small Research Base
Outside the Harding lab and the 2021 Chinese study, there is almost no independent research on Dihexa. The compound has not been widely studied by the broader neuroscience community.
Frequently Asked Questions
Is Dihexa a nootropic?
It is classified as one in the biohacking community, but it does not work like a typical nootropic. Most nootropics modulate neurotransmitters; Dihexa promotes the physical formation of new synapses through HGF/c-Met signaling. It is better described as an experimental neurotrophic compound. Notably, it did not improve cognition in rats with normal brain function.
Is Dihexa FDA-approved?
No. Dihexa has not been approved by the FDA or any regulatory body worldwide. It remains an experimental research compound.
Has Dihexa been tested in humans?
Dihexa itself has not. Its pro-drug fosgonimeton (ATH-1017) was tested in Phase 1, 2, and 2/3 clinical trials in Alzheimer's and Parkinson's patients, but failed to meet its primary endpoints and development was discontinued.
Is Dihexa really 10 million times more potent than BDNF?
In one specific assay measuring dendritic spine formation in hippocampal neuron cultures, Dihexa worked at concentrations 10 million times lower than BDNF. This comparison is accurate within that narrow context but does not mean Dihexa is broadly superior to BDNF for brain function.
Can Dihexa cause cancer?
Unknown, but the concern is scientifically grounded. HGF/c-Met is a known oncogenic pathway, and long-term activation could theoretically promote tumor growth. No long-term cancer studies have been conducted with Dihexa.
How does Dihexa compare to other cognitive peptides?
Dihexa has a unique mechanism (HGF/c-Met potentiation) and is among the most potent neurotrophic compounds described in preclinical research. However, it also has the thinnest safety data and most serious credibility concerns. Semax and Selank have longer track records in human use, particularly in Russia. Cerebrolysin has the most clinical trial data of any neuropeptide. See our guide to best peptides for cognitive improvement for a broader comparison.
What happened with the research fraud allegations?
The CEO of Athira Pharma (the company commercializing Dihexa) was found to have manipulated western blot images in foundational Dihexa papers from her doctoral research. One paper has been retracted, others received expressions of concern, and Athira settled with the DOJ over NIH grants referencing the compromised work.
The Bottom Line
Dihexa occupies an unusual position in peptide research. The core idea — using a small, orally bioavailable molecule to activate HGF/c-Met signaling and promote synaptogenesis — is scientifically interesting. The preclinical behavioral data showing cognitive rescue in impaired animals is consistent across studies, including at least one from an independent group.
But the problems are serious. The foundational papers have been tainted by confirmed data fabrication. The pro-drug failed in clinical trials. No human has been tested with Dihexa under controlled conditions. The safety profile is almost completely unknown, and the mechanism raises legitimate cancer concerns.
For readers exploring peptides for cognitive support, compounds with more established research bases — such as Cerebrolysin, Semax, or Selank — may offer more transparent risk-benefit profiles. The emerging compound PE-22-28 is another nootropic peptide worth tracking, and Epitalon addresses a different longevity pathway entirely.
Dihexa is not a scam, and it is not a miracle drug. It is an early-stage research compound with a complicated history, and the honest assessment is that we do not yet know what it can do — or what harm it might cause — in humans.