Why this article frames the evidence, not the benefits
Dihexa is a Tier 2 transitional research peptide. There are no completed, peer-reviewed human clinical trials of Dihexa in any indication. Speaking of Dihexa “benefits” in the absence of human evidence would imply clinical effect that has not been demonstrated, in a regulatory area (Alzheimer’s disease, cognitive disorders) that is among the most heavily scrutinized in pharmaceutical medicine.
The honest framing is research findings. What the preclinical literature documents in cell culture and rodent models is real and worth discussing carefully. It is not a basis for marketing claims, and it is not a substitute for the human safety and efficacy data that would normally support any cognitive therapeutic. Treat everything that follows as research context, educational only, and not a recommendation to use Dihexa.
The HGF / c-Met mechanism in detail
Understanding what the preclinical findings mean requires understanding the pharmacological hypothesis they are meant to support.
Hepatocyte growth factor (HGF)
HGF is a multifunctional growth factor first characterized in liver biology (hence the name) but expressed widely across tissues, including the central nervous system. In neurons, HGF has been characterized as a neurotrophic signal that supports neuronal survival, axonal growth, and synaptic development. HGF acts at a single receptor: c-Met, a receptor tyrosine kinase.
c-Met receptor signaling
When HGF binds c-Met, the receptor dimerizes and autophosphorylates, triggering downstream pathways that include PI3K / Akt, RAS / MAPK, and others involved in cell survival, growth, and motility. In the CNS specifically, HGF / c-Met signaling has been associated with synaptic plasticity and dendritic spine dynamics. This is the literature backdrop that makes a CNS-active HGF / c-Met agonist mechanistically interesting in the context of neurodegenerative cognitive decline.
The Dihexa proposal, and the retracted paper that anchored it
The Harding lab’s argument is that Dihexa potentiates HGF activity at c-Met in CNS tissue. The cleanest piece of mechanistic support for that argument was the HGF-dependence experiment in Benoist et al. (2014), Journal of Pharmacology and Experimental Therapeutics351:390-402: in cell-culture systems where HGF / c-Met signaling was blocked (HGF antagonist or shRNA against c-Met), Dihexa exposure no longer produced the spine-density effect, which the authors took as evidence that the compound’s pharmacology is HGF-dependent rather than receptor-independent.
That paper was formally retracted in April 2025 after a Washington State University investigation concluded that several figures had been manipulated. The retraction does not, by itself, settle whether the HGF / c-Met hypothesis is correct. It does mean the most-cited piece of mechanistic evidence for that hypothesis has been withdrawn from the peer-reviewed record. Anyone still describing the HGF / c-Met dependence claim as “cleanly established” is overstating what the surviving literature supports.
What the AT4 / IRAP connection adds
Dihexa is descended from angiotensin IV, the AT4 receptor ligand. AT4 has been characterized as insulin-regulated aminopeptidase (IRAP), and IRAP itself is associated with memory and learning in rodent models. The relationship between AT4 / IRAP pharmacology and the HGF / c-Met effect Dihexa is reported to produce remains an area of active research interpretation. The clean summary, suitable for an educational article, is: Dihexa was designed at the AT4 ligand pharmacophore, but the mechanism the lab argues drives its cognitive effects in cell culture is HGF-dependent c-Met signaling.
Rodent synaptogenesis findings
Synaptogenesis is the formation of new synaptic connections between neurons. Dendritic spine density, the number of small protrusions on a neuron’s dendritic arbor, is a structural proxy for the number of excitatory synaptic inputs the cell receives. Loss of dendritic spines is a recognized correlate of cognitive decline in neurodegenerative disease.
Cell-culture spine density
In dissociated rat hippocampal neuron cultures, the Harding lab and collaborators reported that Dihexa exposure produced a dose-dependent increase in dendritic spine density. The reported potency in this in-vitro assay was substantially greater than the parent angiotensin IV molecule. The spine-density observation is the foundational neurobiological finding the rest of the program is built around.
This is a cell-culture result. It is genuinely interesting at the laboratory level. It is not a clinical outcome, and the leap from a spine-density signal in dissociated neuron cultures to a measurable cognitive benefit in a living human is not a leap the published literature supports.
In-vivo extension: living rodents
The lab extended the cell-culture work into living rodent brain tissue, examining dendritic spine density in hippocampal regions of treated animals. Reported findings include increased spine density associated with Dihexa administration in rodent models. Again, this is structural neurobiology in a rodent, not clinical efficacy data in a human.
The scopolamine-induced amnesia models
Scopolamine is a muscarinic acetylcholine receptor antagonist. Administered to rodents, it produces a pharmacologically induced cognitive deficit, primarily affecting tasks that depend on cholinergic signaling. The scopolamine model is a standard preclinical screen for cognitive-enhancing compounds.
In published preclinical work, Dihexa-treated rodents showed improved performance compared to controls on cognitive tasks following scopolamine-induced cognitive deficit. The interpretation in the published discussion is that Dihexa partially counteracted the cholinergic-deficit impairment, presumably through its effects on synaptic structure and function.
The scopolamine-amnesia model is useful in preclinical screening but should not be confused with a clinical Alzheimer’s disease model. Many compounds reverse scopolamine-induced cognitive deficit in rodents and fail to produce clinically meaningful benefit in human Alzheimer’s disease. The cholinergic-deficit-only model is a narrow assay. It is not a substitute for the multi-mechanism, longitudinal degeneration that characterizes human neurocognitive disorders.
Aging-rodent memory paradigms
A separate set of preclinical experiments tested Dihexa in aged rodents, animals selected for naturally diminished cognitive performance relative to young controls. In these aging cohorts, Dihexa-treated animals showed improved performance in spatial and working memory paradigms compared to untreated aged controls.
The Morris water maze is the most-cited paradigm in this literature. It tests a rodent’s ability to learn the location of a hidden submerged platform in a circular pool, using only distal visual cues. It is sensitive to hippocampal-dependent spatial learning, and it is one of the workhorses of preclinical cognition research. Dihexa-treated aged rats reportedly performed better than untreated aged controls in this paradigm.
As with the scopolamine model, the aging-rodent paradigm is a useful preclinical tool, not a model of human Alzheimer’s disease or human cognitive aging. Many compounds improve maze performance in aged rats and do not translate to clinical benefit in older humans. The translation question is not whether the rodent finding is real, it is whether the rodent finding generalizes.
The dendritic spine density question, more carefully
Dendritic spine density is widely treated, in popular peptide writing, as if it were a clinical endpoint. It is not. It is a structural neurobiological measurement.
The relationship between dendritic spine density and cognitive function is real but indirect:
- Loss of spines is a well-documented correlate of cognitive decline in neurodegenerative disease.
- Increase in spines in a preclinical model is not the same thing as restoration of cognitive function. New spines must integrate into functional circuits, receive appropriate input, and survive long enough to participate in memory encoding.
- Spine density in cell culture is a particularly distant proxy. It tells us a compound can drive a structural change in a dish. It does not tell us how the same compound behaves in an intact brain over months of treatment, or whether the new spines are the right ones.
This is not a dismissal of the spine-density work. It is an honest framing of what that work demonstrates and what it does not.
The “10 million times more potent than BDNF” claim, contextualized
The single most repeated claim about Dihexa is that it is roughly 10 million times more potent than BDNF (brain-derived neurotrophic factor). Some derivative writing renders the same finding as “100,000 times.” The primary literature is closer to the larger figure. This number has a real source.
Where the number comes from
In the Harding lab’s in-vitro spine-density assay in dissociated rat hippocampal cultures, Dihexa drove a measurable spine-density effect at picomolar concentrations (around 10-12M), while BDNF required micromolar concentrations to produce a comparable effect in the same system. That ratio of roughly seven orders of magnitude in molar concentration is the source of the “10 million times” figure circulating in derivative writing about the compound.
What the number does not mean
It is not a head-to-head clinical trial. It is not a measure of cognitive efficacy in humans. It is not a measure of cognitive efficacy in rodents. It is a relative-potency value from one in-vitro assay context, in cell culture, with a structural readout (spine density) rather than a behavioral or clinical outcome. Treating it as a clinical claim, or even as a proxy clinical claim, is a category error.
Why this matters
Potency is not efficacy. A compound can be staggeringly potent in vitro at a particular molecular readout and produce no meaningful clinical effect in humans. It can be highly potent in vitro and produce serious adverse effects in vivo. The history of pharmacology is filled with high-potency in-vitro molecules that failed in clinical translation. The seven-orders-of-magnitude figure is a reasonable preclinical observation. It is not a marketing claim, and the published authors did not present it as one. The integrity caveat is also relevant here: the spine-density assay program sits inside the same cluster of papers that drew expressions of concern, with one closely related paper (Benoist et al., 2014) retracted in April 2025.
Why human translation is uncertain
Even if every preclinical finding above were taken at face value, translation to human cognitive disorders would not be guaranteed. Several specific obstacles deserve to be named.
Blood-brain barrier penetration: rodent vs human
The published characterization of Dihexa as blood-brain-barrier-penetrant is based on rodent models. The rodent and human blood-brain barriers differ in tight-junction protein composition, transporter expression, and metabolic handling of small molecules. A compound that is BBB-penetrant in a rat is not automatically BBB-penetrant in a human, and rodent CNS exposure does not automatically predict human CNS exposure at any given oral dose.
Dose extrapolation
Allometric scaling from rodent doses to human doses is an imprecise science, particularly for CNS-active compounds where target engagement at the right tissue concentration matters. The doses circulating in nootropic-community discussion are extrapolations from rodent work without the human pharmacokinetic baseline that any standard drug development program would generate. There is no published, peer-reviewed human pharmacokinetic profile for Dihexa.
Rodent cognitive tasks vs human cognition
Rodent cognitive paradigms, including the Morris water maze and scopolamine-induced amnesia models, are validated preclinical screens. They are not models of human cognition. They are narrow, task-specific, short-timescale assays of behavior in a small-brained species. Many compounds have shown unambiguous effects in these paradigms and have failed to deliver cognitive benefit in human trials. The translation gap between rodent memory tasks and human Alzheimer’s disease is one of the largest in pharmaceutical development.
The disease-process question
Human Alzheimer’s disease involves amyloid pathology, tau pathology, neuroinflammation, vascular dysfunction, and progressive synaptic loss accumulating over years to decades. Driving spine formation in a healthy rodent neuron culture, or rescuing scopolamine-induced cognitive deficit in a young rat, addresses one piece of one mechanism. It does not address the disease process. A compound can drive synaptogenesis and still fail to alter the trajectory of human dementia.
Areas where the evidence is essentially absent
Several questions a reasonable reader would ask have no published answer in Dihexa’s case. Acknowledging the absences is part of an honest evidence review.
- Human pharmacokinetics. No peer-reviewed published characterization of Dihexa absorption, distribution, metabolism, or excretion in humans.
- Human dose-response. No controlled trial that titrates Dihexa exposure against any cognitive, imaging, or biomarker endpoint in humans.
- Human safety baseline. No completed phase 1 safety study published in the peer-reviewed literature.
- Long-term effects. No long-term rodent or human safety data published for chronic Dihexa exposure.
- Effects on the HGF / c-Met oncology axis. Given that c-Met activation is a documented driver of cell proliferation in multiple solid tumors (see Trusolino, Bertotti, and Comoglio, 2010, Nature Reviews Molecular Cell Biology, for a canonical review of the oncology biology), the absence of clinical safety work specifically characterizing this risk is a meaningful gap.
- Drug interactions. Not systematically studied in humans.
- Population-specific risk. Effects in older adults, in patients with cardiovascular disease, in patients with prior or active malignancy, in pregnant or nursing individuals, are not characterized.
How researchers and clinicians frame the gap
Inside academic neuropharmacology, Dihexa is generally regarded as a candidate molecule of mechanistic interest that has not progressed clinically. It is cited in review articles on AT4 / IRAP pharmacology and on procognitive small molecules. It is not, in 2026, listed in any major Alzheimer’s disease clinical trial registry as the active investigational product of a sponsor-led program reaching peer-reviewed publication.
Inside the nootropic community, the framing is dramatically more enthusiastic. Forum threads, vendor copy, and influencer content frequently treat the 10-million-times BDNF figure and the rodent maze findings as if they were established clinical evidence. Some of this content is well-meaning. Some of it is commercial. None of it constitutes the clinical evidence base it is sometimes presented as.
Among compliance-minded telehealth practitioners, the responsible posture is: acknowledge the preclinical literature is real, decline to make outcome-promising claims about a compound that has not been studied in humans, and avoid commercial promotion of a Tier 2 transitional substance. That is PepScribe’s posture, and it is the posture this article reflects.
What this means for someone evaluating Dihexa today
If you are considering Dihexa for a cognitive concern, the responsible answer is straightforward, even if it is not the answer the gray-market vendor ecosystem encourages.
- There is not enough evidence to recommend Dihexa. The absence of published human clinical data means there is no responsible basis for a clinician to prescribe it for any cognitive indication. PepScribe does not currently offer Dihexa.
- Self-administering research-channel material is not an alternative. Research-grade Dihexa is not pharmaceutical-grade. There is no potency assay, no impurity profile, no sterility testing, no chain-of-custody documentation. Adding an unverified compound to a body without a clinical safety baseline is a meaningful risk profile.
- A genuine cognitive concern deserves medical evaluation. Persistent cognitive symptoms, memory changes, or concerns about cognitive aging warrant evaluation by a clinician. There are evidence-supported avenues, including comprehensive medical workup, cardiovascular risk reduction, sleep evaluation, mood evaluation, and approved cognitive treatments where appropriate, that should precede any consideration of an unstudied research molecule.
- For peptide therapy generally, there are Tier 1 options with established compounding pathways. Sermorelin, for example, is a growth-hormone-releasing hormone analog that is available through licensed 503A compounding pharmacies under clinician supervision. It supports deep sleep, the phase during which the brain consolidates memory, and is part of a clinician-led conversation about recovery and wellness rather than a gray-market self-experiment. Learn more about Sermorelin.
Frequently asked questions
Has Dihexa been shown to improve memory in humans?
No completed, peer-reviewed human clinical trial has demonstrated a memory benefit from Dihexa. Reported benefits in the popular discussion are either preclinical findings in rodents or self-reported anecdote-grade observations from forum users. Neither is a substitute for the controlled clinical evidence that would normally support a cognitive claim.
Does Dihexa treat Alzheimer’s disease?
No. Dihexa is not an FDA-approved treatment for Alzheimer’s disease, has not completed a published human clinical trial in Alzheimer’s disease, and should not be characterized as a treatment for any cognitive disorder. The compound has been studied in preclinical models as a candidate for research, not as an established therapy.
Is Dihexa 10 million times more potent than BDNF?
The figure (sometimes paraphrased as 100,000 times in derivative writing) originates from a specific in-vitro spine-density assay in cell culture, where Dihexa was active at picomolar concentrations and BDNF required micromolar concentrations to produce a comparable effect. It is a relative-potency value in that assay, not a clinical claim and not a measure of comparative therapeutic effect. Treat it as a preclinical observation, not a marketing claim, and weight it against the fact that several papers in the underlying program carry expressions of concern.
Is Dihexa safe?
There are no completed phase 1 human safety trials of Dihexa published in peer-reviewed journals. The mechanism (HGF / c-Met agonism) raises a theoretical concern about pro-tumor signaling that has not been clinically addressed. Self-reported adverse effects in nootropic-community use have included headache, lightheadedness, and reported blood pressure changes, but these are anecdote-grade reports, not safety data. The honest answer is: human safety has not been established.
Can I get Dihexa through PepScribe?
No. PepScribe does not currently offer Dihexa. Our pharmacy standard is 503A compounding only, and the regulatory posture of Dihexa Acetate after the April 15, 2026 reshuffle has not been affirmatively cleared for 503A use. We evaluate transitional compounds in clinician-led consultation, not by selling them as commercial products.
The honest summary
Dihexa is a compound with a coherent mechanistic hypothesis, a preclinical evidence base in rodents that is partially under expression-of-concern review, an interesting in-vitro neurobiological signal, the April 2025 retraction of its most-cited mechanism paper (Benoist et al., 2014), and zero published completed human clinical trials. It is mechanism without medicine, on a body of work whose integrity has been actively questioned by the journal that published it. The surviving findings are still worth taking seriously as research, with calibrated skepticism. They are not a basis for self-administration, not a basis for purchase from research-chemical vendors, and not a basis for treating a CNS-active, pro-growth signaling agent as if its safety in humans had been established.
If the human clinical program ever runs, the picture will sharpen. Until then, the responsible position is the one this article has tried to model: acknowledge what the research shows, be explicit about what it does not, and treat the gap between rodent neurobiology and human cognition with the seriousness it deserves.