Dihexa (PNB-0408) is a small molecule built from a fragment of the hormone angiotensin IV at Washington State University, where it improved learning in rats.[1][2] It is sold online as a nootropic, but no published study has tested Dihexa in people, and FDA says it has found no human exposure data.[22] A key paper describing how it works was retracted in 2025.[4]
What is Dihexa?
Dihexa’s chemical name is N-hexanoic-Tyr-Ile-(6) aminohexanoic amide. Researchers shortened an angiotensin IV analog to its first few amino acids and modified both ends so it would resist breakdown, cross the blood-brain barrier and work when swallowed.[1] It is technically a modified two-amino-acid peptide, which is why it is often grouped with peptides.
The developers, Joseph Harding and John Wright, described it in 2012 and planned to fund development through their company, M3 Biotechnology.[2] That company, later renamed Athira Pharma, held an exclusive license to Dihexa from the university.[8][20] Athira later said its own studies failed to show that Dihexa had the characteristics needed for a successful drug, and it stopped developing it.[8] In a 2021 SEC filing, Athira also said its Alzheimer’s drug fosgonimeton is rapidly converted to Dihexa in the blood.[8]
How Dihexa is thought to work
The proposed mechanism is that Dihexa binds hepatocyte growth factor (HGF) and boosts signaling through its receptor, c-Met. In brain cells, that pathway can promote new dendritic spines and synapses.[9][10]
The main paper showing this, Benoist and colleagues (2014), was retracted in April 2025 at the editor’s request after a Washington State University investigation found falsified or fabricated data in several figures, for which it held Leen Kawas and Joseph Harding solely responsible.[3][4] The journal had flagged it and three other papers from the lab in 2021 over possible image manipulation.[7] A special committee of Athira’s board then found that Kawas, Athira’s then-CEO and a co-author, had altered images in her 2011 doctoral dissertation and at least four research papers she co-authored as a graduate student.[8] Two other papers from that 2021 group were also retracted in 2025.[5][6]
A 2012 news release said Dihexa was “seven orders of magnitude” more powerful than BDNF in lab tests on nerve cells. This “10 million times BDNF” figure is often repeated online. It came from the developers’ own bench assays, not from people.[2]
What the research shows
All Dihexa studies are in animals, fish or cells:
| Model | Study | Result |
|---|---|---|
| Rats with drug-induced amnesia | Oral 2 mg/kg per day | Learning in the water maze returned to normal[1] |
| 24-month-old rats | Oral 2 mg/kg per day | Better water maze learning than untreated old rats[1] |
| Alzheimer’s model mice (China) | Several doses | Better memory, less brain inflammation[11] |
| Huntington’s-like toxin in rats | Given with the toxin for 5 weeks | No protection of movement or memory[12] |
| Repeated mild brain injury in rats | Infused into the brain | Working memory deficits improved at the higher dose[13] |
| Zebrafish hair cells | Exposed to antibiotics that damage hearing | Fewer hair cells lost[14] |
| Rat sciatic nerve repair | Given with stem cells into muscle | Better walking scores than some comparison groups[15] |
The original 2013 study, which first described Dihexa, has carried an expression of concern since 2021 but has not been retracted.[1][7] The developers also reported benefits in Parkinson’s disease models in animals.[10]
Has Dihexa been tested in humans?
Not directly. The closest human data come from fosgonimeton (ATH-1017), an injectable drug Athira developed after dropping Dihexa. In 2021, Athira told investors that fosgonimeton is a prodrug that is rapidly converted in the blood into Dihexa.[8] Athira’s later research papers call the active form “fosgo-AM” rather than Dihexa.[27]
- Phase 1 (88 people): single doses up to 90 mg and 9 days of daily doses up to 80 mg were well tolerated, and a small group with Alzheimer’s showed a change in a brain-wave measure (P300 latency).[16]
- ACT-AD (phase 2, 77 people, 26 weeks): missed its primary endpoint, P300 latency. Athira pointed to a better result in patients not taking donepezil-type drugs, but that was a post hoc subgroup analysis and was not statistically significant.[17][20]
- LIFT-AD (phase 2/3, 26 weeks): fosgonimeton 40 mg did not beat placebo on its combined memory and daily-function score (p = 0.70), or on memory (p = 0.35) or function (p = 0.61) alone.[18][19]
After these results, the company paused fosgonimeton and in January 2026 renamed itself LeonaBio to focus on other drugs.[20]
Dihexa vs fosgonimeton
| Dihexa | Fosgonimeton (ATH-1017) | |
|---|---|---|
| Form | Small molecule; oral in rats[1] | Daily injection under the skin[16] |
| Human trials | None published | Phase 1, 2 and 2/3[16][17][19] |
| Alzheimer’s results | Rat and mouse studies only[1][11] | Missed main goals in ACT-AD and LIFT-AD[17][19] |
| Status | Gray-market research chemical | Development paused[20] |
Dihexa vs Semax and Selank
Semax and Selank are Russian nasal peptides that have been tested in small human studies and are approved in Russia. Dihexa works through a different pathway (HGF/c-Met rather than BDNF or the stress and immune systems) and has no human studies of its own.[9] None of the three has been compared with the others in a trial.
Dihexa dosage: what was studied
There is no studied human dose. In rats, effective doses were 0.25 to 0.5 mg/kg injected into the abdomen and 2 mg/kg by mouth each day.[1] Animal doses do not translate directly to people, and online dosing advice is not based on human research.
One forum user reported taking 30 mg by mouth daily for four days, then 30 mg about every week and a half, with no way to check what was in the product.[26]
How long does Dihexa last?
In rats, Dihexa had a very long half-life: about 12.7 days after an intravenous dose and 8.8 days after an injection into the abdomen. It also resisted breakdown by liver enzymes and spread widely outside the bloodstream.[1] No human data exist, so it is unknown how long it stays in a person’s body.
Dihexa side effects and safety concerns
- No human safety data: FDA says it lacks important information on whether dihexa acetate would cause harm if given to people.[22]
- Theoretical cancer concern: HGF and c-Met drive cell growth, survival, blood-vessel growth and spread in many cancers, and drugs that block c-Met have shown benefit against some tumors in clinical studies.[21] Dihexa is designed to boost that same pathway. No study has tested whether it raises cancer risk.
- Long exposure: the long half-life seen in rats means any unwanted effect could last for days or weeks.[1]
- Related drug: in LIFT-AD, 14.2% of people on fosgonimeton stopped because of side effects, mostly injection-site reactions, compared with 4.6% on placebo. Serious side effects were similar (4.2% vs 6.9%).[19]
- Product quality: gray-market products are not tested for identity, purity or dose.
Is Dihexa legal?
- United States: not FDA-approved. FDA placed dihexa acetate among compounding substances that may present significant safety risks, noting it had found no human exposure data.[22] In April 2026 FDA removed it from that category because the nomination had been withdrawn, which is not an approval. FDA said it intends to consult its Pharmacy Compounding Advisory Committee about Dihexa before the end of February 2027.[23]
- Sports: not named on WADA’s 2026 list. In our reading it would still be banned at all times, either as a growth factor modulator (class S2.3, which names HGF) or under S0, which covers drugs with no government approval for human use. Athletes should check with their anti-doping organization.[24]
This is general information, not legal advice.
