Dihexa Peptide: What Research Actually Shows in 2026

Dihexa is the most-hyped cognitive peptide on the internet and one of the least validated in humans. The molecule has a coherent preclinical story, a retracted mechanistic paper, an independent replication in an Alzheimer's mouse model, and zero completed human trials. This guide separates what the published record actually supports from what the nootropic underground has projected onto it.

Key takeaways#

• Dihexa (developmental code PNB-0408) is a six-amino-acid angiotensin IV analog engineered to be orally active and to cross the blood-brain barrier, two properties most peptides lack.
• Its proposed mechanism is potentiation of hepatocyte growth factor (HGF) signaling at the c-Met receptor, driving synaptogenesis and spinogenesis at picomolar concentrations in vitro.
• The 2014 Benoist paper that established the Kd = 65 pM HGF binding figure was retracted in April 2025 after a Washington State University investigation found falsified data; the 2013 McCoy paper carries a Notice of Concern.
• An independent 2021 study in APP/PS1 Alzheimer's-model mice replicated dihexa's procognitive effects through PI3K/AKT signaling, providing the cleanest available corroboration.
• No human clinical trials have been completed. Long-term safety, particularly around chronic activation of a pathway involved in tumor biology, is entirely unstudied.

Dihexa was engineered to solve two problems most peptides cannot#

Most peptides fail before they reach the brain. They are degraded by stomach enzymes if taken orally, and even when injected they cannot cross the blood-brain barrier in meaningful quantities. Dihexa was specifically designed to bypass both obstacles. Its full chemical name is N-(1-Oxohexyl)-l-tyrosyl-N-(6-amino-6-oxohexyl)-l-isoleucinamide, with molecular formula C27H44N4O5 and molar mass 504.672 g/mol.

Dihexa is a small synthetic peptide built from angiotensin IV, engineered at Washington State University to be orally active and to cross into the brain. It has been studied as a procognitive compound that may rebuild synaptic connections, and lab claims of extreme potency made it a darling of the nootropic underground. The structural logic is straightforward: native angiotensin IV clears plasma almost instantly, so the WSU team added a hexanoyl group at the N-terminus and an aminohexanoic amide at the C-terminus to shield the molecule from peptidases and increase lipophilicity.

The pharmacokinetic payoff was substantial. In Sprague-Dawley rat studies, dihexa demonstrated a circulating half-life of 12.68 days after intravenous administration and 8.83 days after intraperitoneal administration, which is extraordinary for a peptide of this size.

The proposed mechanism centers on HGF and c-Met#

The original mechanistic story is clean on paper. Dihexa is an oligopeptide and angiotensin IV-derived drug that binds with high affinity to hepatocyte growth factor (HGF) and potentiates its activity at its receptor, c-Met. The c-Met receptor, when activated by HGF, drives a signaling cascade implicated in neuronal survival, synaptogenesis, and neuroplasticity throughout the hippocampus and cortex.

In the 2014 Benoist study, researchers reported that dihexa binds with high affinity to HGF and that both dihexa and its parent compound Nle1-AngIV induce c-Met phosphorylation in the presence of subthreshold concentrations of HGF and augment HGF-dependent cell scattering. They further reported that dihexa and Nle1-AngIV induce hippocampal spinogenesis and synaptogenesis similar to HGF itself.

Research suggests two downstream consequences. First, new dendritic spines that form functional synapses, measured electrophysiologically as increased miniature excitatory postsynaptic current frequencies. Second, activation of PI3K/AKT signaling, which has been shown to support neuronal survival and reduce apoptosis.

The retraction problem cannot be ignored#

Any honest account of dihexa has to address what happened to the foundational papers. The 2014 Benoist paper, which provided the primary evidence that dihexa's mechanism operates through HGF/c-Met binding and established the Kd = 65 pM binding affinity, was formally retracted in April 2025 after Washington State University's investigation confirmed the figures contained falsified and/or fabricated data, with Kawas and Harding found solely responsible.

The McCoy 2013 paper, which reported the behavioral and synaptogenic effects but preceded the mechanistic HGF binding characterization, received an expression of concern but has not been retracted as of March 2026.

This is not a minor footnote. The most-cited number associated with dihexa, the picomolar HGF binding affinity, came from a paper whose figures were judged fabricated. Preliminary evidence for the mechanism therefore rests on a weaker foundation than the popular literature acknowledges.

The picture is not entirely bleak. These integrity issues do not necessarily invalidate all findings related to dihexa: the independent 2021 study by Sun et al. in APP/PS1 mice provides separate corroboration of cognitive benefits and PI3K/AKT pathway involvement, but they cast significant doubt on the specific HGF binding kinetics and mechanistic details reported in the retracted paper. Studies have shown that the broader pharmacology survives, even if the specific molecular numbers do not.

Independent replication exists in one Alzheimer's mouse model#

The most important counterweight to the retraction is a 2021 paper from China Pharmaceutical University, completely independent of the original WSU group. Sun et al. tested dihexa in six-month-old APP/PS1 mice, a genetic model of Alzheimer disease amyloid pathology. Three months of intragastric dihexa at 1.44 or 2.88 mg/kg/day reduced escape latency, increased platform crossings in probe trials, preserved neuronal density, increased synaptophysin expression, and reduced neuroinflammatory markers. This study provided important independent replication of dihexa's procognitive effects and identified PI3K/AKT as the mediating signaling pathway.

The behavioral results were striking. In the Morris water maze, dihexa restored spatial learning and cognitive functions, increased neuronal cells and SYP protein expression, decreased activation of astrocytes and microglia, reduced levels of pro-inflammatory cytokines IL-1β and TNF-α, and increased the anti-inflammatory cytokine IL-10.

These findings are associated with a coherent neuroprotective profile in one rodent model. They do not establish efficacy in humans, and they do not replace the retracted mechanistic data.

Human evidence is absent and safety questions remain open#

This is the part of the dihexa story that the nootropic blogosphere consistently understates. There are no completed human clinical trials. All efficacy and safety claims rest on animal data and an unverified community record of self-experimentation.

The most serious open question is oncological. The HGF/c-Met axis is not just a synaptic regulator; it is also a well-characterised driver of tumor invasion and metastasis in multiple cancer types. While the hypothesis is that controlled HGF/c-Met activation promotes neuronal survival and synaptogenesis, the systemic consequences of chronically enhancing a pro-oncogenic pathway are entirely unknown. No carcinogenicity, tumor promotion, or long-term safety studies have been published.

This is not a hypothetical concern. It is the standard pharmacological caveat that applies whenever a research-grade compound potentiates a pathway that oncology has spent decades trying to inhibit. Preliminary evidence in healthy adults does not exist either way.

How dihexa compares to better-validated cognitive peptides#

For readers comparing options in the cognitive cluster, the honest framing is that dihexa has the most exciting mechanism and the thinnest human record. Semax is the substantially more evidence-based choice, with decades of clinical use in Russia, multiple human trials, regulatory approval, and a well-characterized safety profile. Dihexa has a fascinating preclinical mechanism (HGF/c-Met pathway activation for synaptogenesis) but zero human clinical data. For anyone considering a nootropic peptide, semax offers proven safety and established dosing, while dihexa remains an intriguing but unvalidated research compound with unknown human safety.

This is the trade-off the editorial position at Klarovel takes seriously. A more potent mechanism on paper does not outrank a better human safety record in practice. Readers who want to map the broader cognitive peptide landscape can use the peptide calculator to compare candidates by route, evidence weight, and protocol complexity, and the how-it-works overview explains how Klarovel curates the protocol layer while partner suppliers handle fulfilment.

What the community protocols look like (and why they are not guidance)#

The published animal dosing was modest. Behavioral effects in rats appeared at oral 2 mg/kg in the McCoy 2013 study, and the Sun 2021 work used 1.44 and 2.88 mg/kg/day intragastric in mice. Extrapolating across species without human pharmacokinetic data is exactly the kind of guesswork that makes safety claims unreliable.

Community-reported protocols circulate in the 2 to 5 mg subcutaneous or 8 to 45 mg oral daily range with cycling, but these numbers are not derived from controlled trials and are not endorsed here. Without published human pharmacokinetics, there is no defensible dose. This is a structural problem, not an editorial preference.

## Where this leaves the protocol layer

Dihexa is a fascinating molecule with a coherent preclinical mechanism, an independent rodent replication, a retracted foundational paper, and no human data. That combination warrants serious scientific interest and serious caution in equal measure. The editorial position here is that overstated benefits from any cognitive compound, dihexa included, deserve to be pushed back against, even when the underlying biology is genuinely interesting.

Klarovel does not sell or stock peptides. Klarovel curates the protocol layer, the evidence-weighting, and the comparison logic that lets readers decide what is worth pursuing and what is worth waiting on. For dihexa specifically, the honest answer is wait: wait for human pharmacokinetics, wait for a Phase I, wait for the safety questions around c-Met activation to be addressed by data rather than speculation. Create an account to track the literature as it develops, and read the disclosures page for the full editorial framework that underpins this assessment.