- Dihexa is a six-amino-acid compound derived from angiotensin IV, developed at Washington State University.
- In a cell-culture test that counts new synapses per unit of compound, it outperformed BDNF by about seven orders of magnitude (McCoy et al., 2013). That is a lab measurement, not a clinical one.
- In preclinical models of memory impairment and Alzheimer's-like disease, it improved memory tasks and increased hippocampal synapse counts.
- No human clinical trial of Dihexa has been completed, so there is no human safety or efficacy data.
- It acts through the HGF/c-Met pathway, which cancer cells use to grow and spread; that is the main safety question and has never been tested in people.
Dihexa has one of the most striking preclinical numbers in the brain-peptide literature and no human trial at all. This page explains what was measured, in what, and why the same mechanism that made researchers excited has kept it out of human studies for over a decade. It is a research summary, not a guide to using Dihexa, which is not an approved drug.
How much more potent Dihexa was than BDNF (brain-derived neurotrophic factor) at triggering new connections between neurons in cell culture. Source: McCoy et al., Journal of Pharmacology and Experimental Therapeutics, 2013 (PMID 23055539). Potency in a dish says nothing on its own about effects or safety in a person.
BDNF is a growth factor the brain uses to build and maintain synapses, the connections between neurons. Dihexa does not supply BDNF. In lab studies it switches on a pathway one step upstream, so the brain's own growth signalling produces more synapse-building activity.
The proposed mechanism
Dihexa is a six-amino-acid peptide built from angiotensin IV, a fragment of the hormone system that controls blood pressure and also has a separate role in the brain. A Washington State University team led by pharmacologist Joe Harding was studying angiotensin IV analogs for memory effects when one derivative, which became Dihexa, showed an unexpectedly potent effect.
It does not bind the BDNF receptor directly. In lab studies it activates hepatocyte growth factor (HGF) and its receptor, c-Met. That pathway sits upstream of TrkB, the receptor BDNF uses to drive synapse formation. The potency figure comes from one specific test, counting how many new connections formed per unit of compound applied to neurons in a dish, in which Dihexa outperformed both NGF and BDNF.
What the preclinical studies found
The headline result from the Washington State lab was reversal of drug-induced memory impairment in a preclinical maze task, with the effect persisting after the compound had cleared, which the authors read as structural change rather than a short-lived chemical effect. A later study in a preclinical model carrying human Alzheimer's mutations found better performance on several memory tasks and more synapses in the hippocampus, the region that forms new memories. A companion paper (Benoist et al., Journal of Pharmacology and Experimental Therapeutics, 2014, PMID 25187433) studied how the compound moves through the body and reaches the brain in preclinical models.
Preclinical Alzheimer's models are poor predictors of human outcomes. Most compounds that improved memory in these models have failed in human trials. That is a limitation of the whole field, not a criticism specific to Dihexa, and it is why these results are not evidence of a human effect.
| Finding | Evidence level | What it tells us about people |
|---|---|---|
| Millions of times more potent than BDNF at synapse formation | Cell culture | Mechanism plausible; potency in a living person unknown |
| Reversal of drug-induced memory impairment | Preclinical study | Suggests a biological effect; nothing about human benefit |
| Better memory in an Alzheimer's-mutation model | Preclinical disease model | Such models rarely predict human outcomes |
| More hippocampal synapses | Preclinical tissue study | Not measured in people |
| Human cognitive effect | No published data | No human trial has been completed |
Why no human trial has been completed
Dihexa was first characterized in 2012. More than a decade later there is no published human trial. The main reason is the pathway. HGF and c-Met drive cell growth, survival and migration. In healthy tissue that supports wound healing and liver regeneration. In cancer, tumours use the same signal to grow and spread, and elevated HGF/c-Met signalling is associated with metastasis in several tumour types. Drug companies have spent heavily developing c-Met inhibitors as cancer treatments. Dihexa activates the pathway those drugs block.
No human adverse event from Dihexa has been documented, because no human study has been run. That is the absence of an experiment, not evidence of safety. A small number of papers related to early Dihexa characterization have also come under scrutiny in recent years; the core mechanism and potency finding have independent support, but some older effect sizes deserve caution.
Regulatory status
Dihexa is not an approved drug anywhere and is not on the FDA's list of bulk substances that compounding pharmacies may use. FDA treats unapproved products marketed for people as unapproved drugs.
What genetics research does and does not show
Dihexa is discussed alongside BDNF Val66Met (rs6265) because of the pathway it acts on. Met carriers release less BDNF on demand than Val/Val carriers in human studies (Egan et al., Cell, 2003), and roughly a third of people of European descent carry at least one Met copy. APOE4, the variant most strongly linked to Alzheimer's risk, is also discussed in this context. No study has tested whether either genotype changes response to Dihexa, and no Dihexa data exists in people at all.
Consumer DNA files also do not assess cancer risk in the HGF/c-Met pathway. No genotype makes an untested compound safer.
How it compares with better-studied brain peptides
Semax and Selank, two other peptides discussed for cognition, were developed and registered as medicines in Russia and have published human studies, though they are not approved in the US. Dihexa has none. Semax vs Selank covers what their human research shows.
A striking lab result, a cancer-pathway mechanism, and no human data.
Dihexa's synapse-forming potency in cell culture and its preclinical memory results are real findings. It has never been tested in a person, so there is no human safety record, and the pathway it activates is one cancer biologists work to block. For the wider picture, see peptides studied for brain fog.
Watch
What clinicians and researchers say about Dihexa
Frequently asked questions
What is Dihexa?
Dihexa is a synthetic six-amino-acid peptide derived from angiotensin IV, developed at Washington State University. In lab studies it activates the HGF/c-Met pathway upstream of TrkB, the receptor BDNF uses, and in a cell-culture test it was millions of times more potent than BDNF at forming new synapses (McCoy et al., 2013).
Has Dihexa been tested in humans?
No. No human clinical trial of Dihexa has been completed. All efficacy data comes from cell culture and preclinical models. A main reason is safety: the HGF/c-Met pathway it activates is one that tumours use to grow and spread.
Is there any human data on Dihexa?
No. Dihexa has never been given to people in a published study, so there is no human pharmacokinetic data and no human safety data. It is not an approved drug anywhere.
What are the safety concerns with Dihexa?
The main concern is its mechanism. Dihexa activates HGF/c-Met signalling, which supports normal tissue repair but is also used by cancer cells to grow and spread. Drug companies develop c-Met inhibitors as cancer treatments. Because no human study exists, the size of any risk is unknown.
How does Dihexa compare with Semax?
Both are discussed for BDNF-related brain effects. Semax was registered as a medicine in Russia and has published human studies, including stroke trials. Dihexa has only cell-culture and preclinical data. Neither is approved in the US.
Does genetics change how Dihexa works?
No study has tested that. BDNF Val66Met and APOE4 are discussed in this context because of the pathway Dihexa acts on and its Alzheimer's research, but Dihexa has no human data in any genotype.
This article is for informational and educational purposes only. It reports what published research found; it is not medical advice, it does not diagnose, treat, cure or prevent any disease, and nothing in it is a dose or a recommendation to use any compound. Talk to a licensed clinician about any medical decision.