Dihexa

1. Overview

Dihexa (N-hexanoic-Tyr-Ile-(6) aminohexanoic amide; developmental code PNB-0408) is a synthetic, metabolically stabilized analog of angiotensin IV developed by Joseph W. Harding, PhD, and John W. (Jay) Wright, PhD, at Washington State University in Pullman, Washington [4][6]. It emerged from more than two decades of research into the unexpected procognitive effects of angiotensin IV acting through the AT4 receptor subtype in the brain [1][2][3]. Unlike angiotensin IV itself -- a hexapeptide fragment (Val-Tyr-Ile-His-Pro-Phe) of angiotensin II with a serum half-life of roughly two minutes -- dihexa was designed with structural modifications that confer enzymatic stability, blood-brain barrier (BBB) permeability, and oral bioavailability [4].

Chemically, dihexa is a modified tripeptide-like structure with a molecular formula of C27H44N4O5 and a molecular weight of 504.66 g/mol (CAS 1401708-83-5). Its design replaces the N-terminal valine of Nle1-angiotensin IV with a hexanoyl (N-hexanoic acid) cap, retains the core Tyr-Ile pharmacophore, and substitutes the C-terminal His-Pro-Phe with a 6-aminohexanoic amide moiety [4][6]. These modifications eliminate peptidase-sensitive bonds while preserving the structural elements required for biological activity, resulting in an extraordinarily long serum half-life of approximately 12.7 days following intravenous administration in rats, compared with the sub-minute half-life of the parent compound [4].

The compound gained wide attention for reports that it promotes dendritic spine formation and synaptogenesis at picomolar concentrations in hippocampal neuron cultures -- seven orders of magnitude (approximately 10 million-fold) more potent than brain-derived neurotrophic factor (BDNF) in this specific in vitro assay [4]. However, this potency comparison reflects a narrow endpoint (spine induction in cultured neurons) and does not translate to a general statement about overall cognitive superiority. No human clinical trials have been completed for dihexa itself, though a prodrug derivative called fosgonimeton (ATH-1017) entered Phase 2/3 clinical trials for Alzheimer disease before key foundational papers were affected by data integrity concerns [5][13][20].

Molecular Weight

504.66 g/mol

Chemical Formula

C27H44N4O5

CAS Number

1401708-83-5

Mechanism

Allosteric potentiation of HGF at the c-Met receptor tyrosine kinase; promotes synaptogenesis via PI3K/AKT/mTOR signaling

Routes Studied

Intracerebroventricular, intraperitoneal, oral (rodents)

FDA Status

Not approved for any therapeutic use

WADA Status

Prohibited under S0 (Non-Approved Substances) at all times

2. Scientific Context: From Angiotensin IV to Dihexa

2.1 The Brain Renin-Angiotensin System and AT4 Receptor

The discovery path to dihexa began in the late 1980s and 1990s when Harding, Wright, and colleagues at Washington State University observed that angiotensin IV (AngIV), a metabolite of the vasoconstrictor angiotensin II, had unexpected effects on learning and memory in rodents [1][3]. Unlike angiotensin II, which acts through AT1 and AT2 receptors to regulate blood pressure, AngIV was found to bind a distinct receptor subtype designated AT4, which is heavily distributed in brain regions critical for cognitive processing -- particularly the CA1-CA3 fields of the hippocampus, the neocortex, the thalamus, and the cerebellum [1][3][10].

In a landmark 1999 study, Wright et al. demonstrated that chronic intracerebroventricular infusion of the AT4 agonist Nle1-AngIV (norleucine-substituted angiotensin IV) via osmotic pump facilitated spatial learning acquisition in the Morris water maze, while the AT4 antagonist Divalinal significantly impaired it [1]. Follow-up studies showed that Nle1-AngIV could overcome scopolamine-induced memory deficits at picomolar doses [2].

2.2 The AT4/IRAP Identity Debate

A significant controversy emerged regarding the molecular identity of the AT4 receptor. In 2001, Albiston and colleagues presented compelling evidence that the AT4 binding site corresponds to insulin-regulated aminopeptidase (IRAP, also known as oxytocinase or placental leucine aminopeptidase), a zinc-dependent metallopeptidase [9][10][15]. According to this model, angiotensin IV and its analogs enhance cognition by inhibiting IRAP enzymatic activity, thereby prolonging the action of pro-memory neuropeptides such as oxytocin and vasopressin in the brain [9][10].

However, the Harding and Wright laboratory proposed an alternative mechanism: that the procognitive effects of angiotensin IV analogs operate through the hepatocyte growth factor (HGF)/c-Met receptor tyrosine kinase system rather than, or in addition to, IRAP inhibition [5][6]. This debate remains unresolved in the broader field, though the HGF/c-Met hypothesis has been the dominant framework presented in dihexa-related publications.

2.3 Development of Dihexa

The primary limitation of angiotensin IV and Nle1-AngIV as drug candidates was their extremely rapid enzymatic degradation in serum (half-life of roughly two minutes) and inability to cross the blood-brain barrier, necessitating direct intracerebroventricular administration [4]. The Harding laboratory systematically modified the Nle1-AngIV structure to overcome these limitations:

• N-terminal: Replaced norleucine with a hexanoyl cap to resist aminopeptidase degradation
• Core dipeptide: Retained the Tyr-Ile motif identified as the minimal pharmacophore
• C-terminal: Replaced His-Pro-Phe with 6-aminohexanoic amide to resist carboxypeptidase degradation and increase lipophilicity

The resulting compound, dihexa, proved remarkably stable in serum (intrinsic clearance of 2.72 uL/min/mg; half-life of approximately 509 minutes in vitro) and demonstrated predicted oral bioavailability based on physicochemical modeling, with an effective human jejunal permeability (Peff) value of 1.78, intermediate between two established orally bioavailable drugs (enalapril at 1.25 and piroxicam at 2.14) [4].

3. Mechanism of Action

3.1 HGF/c-Met Pathway Potentiation

The primary mechanistic claim for dihexa is that it functions as an allosteric potentiator of hepatocyte growth factor (HGF) signaling through its receptor tyrosine kinase c-Met [5][6][8]. HGF is a pleiotropic growth factor that, upon binding and activating c-Met, triggers intracellular signaling cascades including PI3K/AKT, MAPK/ERK, and mTOR pathways -- all of which play established roles in neuronal survival, axonal growth, dendritic morphogenesis, and synaptogenesis [17][21].

According to the Benoist et al. 2014 study (subsequently retracted), dihexa binds to HGF with high affinity (Kd = 65 pM) and facilitates the dimerization and activation of HGF at subthreshold concentrations [5]. Rather than acting as a direct c-Met agonist, dihexa was proposed to form a functional heterodimer with endogenous HGF, amplifying the signal at c-Met receptors that would otherwise be insufficient to trigger downstream cascades [5][11]. This allosteric mechanism would explain the compound's extraordinary potency at picomolar concentrations.

Key evidence supporting this mechanism (noting the caveats regarding data integrity discussed in Section 7):

• Dihexa augmented c-Met phosphorylation in HEK-293 cells at subthreshold HGF concentrations [5]
• The HGF antagonist "Hinge" blocked dihexa-induced spinogenesis in hippocampal cultures [5]
• Oral dihexa-mediated cognitive rescue in rats was blocked by intracerebroventricular administration of an HGF antagonist [5]
• c-Met knockdown via shRNA abolished dihexa-induced spine formation [5]
• The PI3K inhibitor wortmannin reversed dihexa's cognitive and neuroprotective effects in APP/PS1 mice [12]

3.2 Synaptogenesis and Spinogenesis

The most widely cited property of dihexa is its ability to promote formation of new dendritic spines and functional synapses. In the McCoy et al. 2013 study, dihexa treatment produced a near 3-fold increase in dendritic spine number in cultured hippocampal neurons at picomolar concentrations [4]. This effect was observed at concentrations seven orders of magnitude lower than those required for BDNF to produce comparable spinogenesis, giving rise to the frequently cited "10 million times more potent than BDNF" claim [4].

It is important to contextualize this comparison: the potency difference is specific to in vitro dendritic spine induction and reflects the different mechanisms of action (allosteric amplification of endogenous HGF signaling vs. direct TrkB receptor activation by BDNF). It does not imply 10 million-fold greater cognitive enhancement in a living organism.

3.3 Downstream Signaling

The intracellular signaling pathways activated by dihexa through HGF/c-Met include:

• PI3K/AKT/mTOR: Confirmed as essential for dihexa's cognitive effects in APP/PS1 mice; the PI3K inhibitor wortmannin abolished dihexa's benefits [12]
• MAPK/ERK: Implicated in dendritic growth and neuronal survival downstream of c-Met [17]
• Akt/TOR/MEK: Required for dihexa-mediated hair cell protection from aminoglycoside toxicity in zebrafish [7]

3.4 Anti-Inflammatory Effects

In the APP/PS1 mouse model, dihexa reduced neuroinflammation markers including decreased activation of astrocytes (GFAP) and microglia (Iba-1), reduced pro-inflammatory cytokines IL-1beta and TNF-alpha, and increased anti-inflammatory IL-10 [12]. These effects suggest dihexa may modulate neuroinflammation in addition to promoting synaptogenesis, though the relative contributions of these mechanisms to cognitive rescue remain unclear.

4. Researched Applications

Cognitive Enhancement and Dementia Models

Evidence level: Preclinical (animal studies only)

The cognitive effects of dihexa have been tested in three principal rodent paradigms:

Scopolamine-induced amnesia model: In the McCoy et al. 2013 study, scopolamine (a muscarinic antagonist that mimics early cholinergic deficits of Alzheimer disease) impaired rat performance in the Morris water maze. Dihexa administered via ICV injection (0.1-1.0 nmol), intraperitoneal injection (0.05-0.50 mg/kg), or oral gavage (1.25-2.0 mg/kg/day) reversed these deficits in a dose-dependent manner [4]. At the highest oral dose (2.0 mg/kg/day), dihexa-treated rats performed indistinguishably from healthy, non-scopolamine-treated controls [4].

Age-related cognitive decline: Aged rats (22-26 months) treated with oral dihexa at 2 mg/kg/day showed significantly improved spatial learning in the Morris water maze compared to untreated age-matched controls [4].

APP/PS1 transgenic mouse model: Sun et al. (2021) -- an independent group from the original WSU laboratory -- 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 [12]. This study provided important independent replication of dihexa's procognitive effects and identified PI3K/AKT as the mediating signaling pathway [12].

Otoprotection

Evidence level: Preclinical (zebrafish)

Uribe et al. (2015) demonstrated that dihexa protects sensory hair cells from aminoglycoside antibiotic toxicity in the larval zebrafish lateral line model [7]. Hair cells of the lateral line are structurally and functionally homologous to mammalian inner ear hair cells. Dihexa provided dose-dependent protection against both neomycin and gentamicin without blocking aminoglycoside entry into hair cells, instead activating pro-survival Akt/TOR/MEK signaling [7]. A patent (US9475854B2) was filed for dihexa as an otoprotective agent.

Parkinson Disease

Evidence level: Theoretical/review only

Wright, Kawas, and Harding (2014) discussed the theoretical potential of HGF/c-Met-activating compounds for Parkinson disease based on HGF's known neurotrophic effects on dopaminergic neurons [6]. However, no published preclinical studies have directly tested dihexa in Parkinson disease animal models.

5. Clinical Evidence Summary

6. Pharmacokinetics and Blood-Brain Barrier Penetration

Metabolic Stability

Dihexa demonstrates remarkable metabolic stability for a peptide-derived compound. Phase I metabolism was found to be very low, with an average intrinsic clearance (Clint) of 2.72 uL/min/mg and an average in vitro half-life of 509.4 minutes [4]. In rats, the serum half-life was approximately 12.7 days following intravenous administration and 8.8 days following intraperitoneal injection [4]. This unusually long persistence stands in stark contrast to the parent compound Nle1-AngIV, which has a half-life of roughly two minutes [4].

Blood-Brain Barrier Penetration

Radiolabeled dihexa studies demonstrated that the compound crossed the BBB and accumulated in multiple brain regions following peripheral administration [4]. The structural modifications that enhance BBB penetration include increased lipophilicity from the hexanoyl cap and aminohexanoic amide tail, and reduced molecular weight compared to the parent hexapeptide.

Oral Bioavailability

Physicochemical modeling predicted oral bioavailability for dihexa based on its predicted effective human jejunal permeability and favorable LogP. Behavioral data confirmed that oral administration was effective: oral dihexa at 2 mg/kg/day completely reversed scopolamine-induced cognitive deficits in rats, indicating that sufficient compound reached the brain after gut absorption [4].

7. Dosing in Research

The following table summarizes doses used in published preclinical research studies. These are not therapeutic recommendations. Dihexa is not approved for human use by any regulatory authority.

8. Safety and Side Effects

Preclinical Safety

No systematic toxicology studies for dihexa have been published. In the available behavioral studies, no overt adverse effects were noted at the doses tested in rats and mice, but these studies were not designed to assess safety endpoints [4][12].

The HGF/c-Met Oncogenesis Concern

The most significant theoretical safety concern with dihexa is the well-established role of HGF/c-Met signaling in cancer biology. The HGF/c-Met pathway is frequently overexpressed or constitutively activated in many human cancers, where it promotes tumor cell proliferation, survival, invasion, and metastasis [17]. This oncogenic role is so well established that multiple pharmaceutical c-Met inhibitors have been developed and approved as cancer therapeutics, including capmatinib, tepotinib, and crizotinib.

Dihexa, by design, potentiates this same pathway. While promoting HGF/c-Met signaling may be beneficial for neuronal health, the systemic consequences of chronically activating a pathway that is simultaneously a driver of cancer progression remain completely unstudied. No long-term carcinogenicity, tumorigenicity, or tumor promotion studies have been conducted with dihexa in any species.

Unknown Risk Profile

Additional areas of uncharacterized risk include:

• Long-term neurotrophic effects: The consequences of sustained, exogenous stimulation of synaptogenesis are unknown. Excessive or disordered synaptic connectivity could theoretically be maladaptive.
• Reproductive and developmental safety: No data exist.
• Drug interactions: No interaction studies have been performed.
• Dose-response safety: The extraordinarily long serum half-life (12.7 days in rats) raises concerns about compound accumulation with repeated dosing.

Anecdotal Reports (Non-Peer-Reviewed)

Some users and clinics that have administered dihexa have informally reported headaches (possibly from excessive synaptic activity), anxiety or overstimulation, sleep disruption, and occasional gastrointestinal discomfort. These reports are not from controlled studies and cannot be verified.

9. Data Integrity Concerns

A significant caveat applies to the dihexa literature. In 2021, an investigation by Washington State University found that Leen H. Kawas -- a co-author on several key dihexa publications and former PhD student of Harding who went on to co-found and serve as CEO of Athira Pharma -- had altered images in her doctoral dissertation and at least four co-authored research papers published between 2011 and 2014 [20]. Research integrity consultants examined 30 images from her dissertation and found problems in 19.

The manipulations included copying and pasting data between experiments, digitally altering western blot band intensities, and reusing identical images to represent different experimental conditions. Most critically, the Benoist et al. 2014 paper [5] -- 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" [20].

The McCoy et al. 2013 paper [4], 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.

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 [12] -- but they cast significant doubt on the specific HGF binding kinetics and mechanistic details reported in the retracted paper.

10. Fosgonimeton (ATH-1017): The Clinical-Stage Prodrug

Athira Pharma (initially M3 Biotechnology) was founded to commercialize dihexa-related compounds. After determining that dihexa itself lacked optimal drug-like characteristics for clinical development, the company developed fosgonimeton (ATH-1017), a prodrug that is administered via subcutaneous injection and rapidly converted in plasma to its active metabolite (the dihexa tyrosine metabolite) [13].

Fosgonimeton entered Phase 1 clinical trials in 2017, testing single and multiple ascending doses from 2 to 90 mg subcutaneously in 88 healthy volunteers and Alzheimer disease patients [13]. The compound was generally well tolerated, with injection site reactions as the most common adverse event [13]. Athira subsequently advanced fosgonimeton into Phase 2 trials (ACT-AD for mild-to-moderate Alzheimer disease and LIFT-AD as a Phase 2/3 study) and a Phase 2 trial for Parkinson disease dementia and dementia with Lewy bodies. However, the data integrity investigation into Kawas's foundational research led to her departure as CEO in 2021, and the clinical development program has been significantly impacted by these controversies.

LIFT-AD Phase 2/3 results (2024): The pivotal LIFT-AD trial announced topline results showing that neither the primary endpoint (Global Statistical Test) nor the key secondary endpoints (ADAS-Cog11 and ADCS-ADL23) reached statistical significance compared to placebo at 26 weeks. Pre-specified subgroup analyses showed numerically greater treatment effects in patients with moderate Alzheimer's disease and in APOE4 carriers. Biomarkers associated with AD pathology showed changes consistent with the neuroprotective mechanism of HGF modulation. Athira has subsequently pivoted to developing next-generation, orally delivered HGF modulators with improved pharmacological properties for neurodegenerative diseases including ALS.

11. Comparison with Related Compounds

| Compound | Structure | Half-Life | BBB Penetration | Oral Activity | Primary Target | |---|---|---|---|---|---| | Angiotensin IV | Val-Tyr-Ile-His-Pro-Phe | ~1-2 min | No | No | AT4/IRAP | | Nle1-AngIV | Nle-Tyr-Ile-His-Pro-Phe | ~2 min | No | No | AT4/IRAP, HGF | | Norleual | Modified Nle1-AngIV analog | Short | Limited | No | HGF antagonist | | Dihexa | Hexanoyl-Tyr-Ile-AHA-NH2 | ~12.7 days (rat IV) | Yes | Yes | HGF/c-Met potentiator | | Fosgonimeton | Dihexa prodrug | Prodrug converted rapidly | Via metabolite | No (subcutaneous) | HGF/c-Met (via active metabolite) |

Norleual is notable as an HGF/c-Met antagonist -- structurally related to dihexa but with opposing pharmacological effects. It has been used as a tool compound to demonstrate that the cognitive effects of angiotensin IV analogs require HGF/c-Met signaling rather than IRAP inhibition alone [5][6].

12. Regulatory Status

United States (FDA): Dihexa is not approved by the FDA for any therapeutic indication. It is commercially available as a research chemical under "for research purposes only" designations. Some clinics prescribe it off-label, operating in a regulatory gray area. Its prodrug derivative fosgonimeton (ATH-1017) has been tested in FDA-regulated clinical trials under an IND.

WADA (World Anti-Doping Agency): As a non-approved pharmacological substance, dihexa falls under WADA's S0 classification (Non-Approved Substances), prohibited at all times both in and out of competition.

ClinicalTrials.gov: No clinical trials for dihexa itself are registered. Multiple trials are registered for fosgonimeton (ATH-1017).

13. Extended Pharmacokinetics

13.1 Extraordinary Metabolic Stability

Dihexa's pharmacokinetic profile is among the most remarkable of any peptide-derived compound. The structural modifications that replace peptidase-sensitive bonds with non-natural linkages produce an in vitro intrinsic clearance (Clint) of only 2.72 uL/min/mg -- indicating minimal Phase I metabolism -- and an in vitro microsomal half-life of 509.4 minutes [4]. For comparison, the parent compound Nle1-AngIV has a serum half-life of approximately 2 minutes, meaning dihexa's stability improvement represents a factor of approximately 250 in vitro and vastly more in vivo [4].

13.2 In Vivo Half-Life: The 12.7-Day Persistence

In rat pharmacokinetic studies, dihexa demonstrated a serum half-life of approximately 12.7 days following intravenous administration and 8.8 days following intraperitoneal injection [4]. These extraordinarily long half-lives for a molecule of only 504.66 Da are unusual and raise both opportunities and concerns:

• Accumulation potential: With a half-life of 12.7 days and daily dosing, steady-state concentrations would not be reached for approximately 6-8 weeks (4-5 half-lives), and compound accumulation could produce significantly higher steady-state levels than the initial dose suggests
• Washout period: Following cessation of dosing, approximately 50-60 days would be required for near-complete elimination (4-5 half-lives), meaning adverse effects, if they occurred, could persist for months
• Protein binding: The long half-life suggests extensive plasma protein binding, which would create a reservoir of bound drug that slowly releases active compound. The specific binding proteins and binding fraction have not been characterized

13.3 Blood-Brain Barrier Penetration

Radiolabeled dihexa studies confirmed BBB penetration following peripheral (IP) administration, with the compound detected in multiple brain regions [4]. The structural features contributing to BBB permeability include the hexanoyl cap (increasing lipophilicity), the reduced number of hydrogen bond donors and acceptors compared to the parent hexapeptide, and the moderate molecular weight (504.66 Da, below the BBB size cutoff of approximately 600 Da for lipophilic molecules). However, the brain-to-plasma ratio and time course of brain accumulation have not been quantified.

13.4 Oral Bioavailability

Dihexa's oral bioavailability was predicted through physicochemical modeling rather than directly measured through formal oral bioavailability studies (AUC oral / AUC IV). The predicted effective human jejunal permeability (Peff) of 1.78 falls between enalapril (1.25, known oral bioavailability approximately 60%) and piroxicam (2.14, known oral bioavailability approximately 100%) [4]. Behavioral studies confirmed functional CNS activity following oral administration (2 mg/kg/day reversed scopolamine-induced cognitive deficits), providing indirect evidence of adequate oral absorption and brain delivery [4]. However, the absolute oral bioavailability fraction has not been formally determined.

13.5 Fosgonimeton Human Pharmacokinetics

The Phase 1 study of fosgonimeton (ATH-1017), dihexa's subcutaneous prodrug, provides the only human pharmacokinetic data for the dihexa chemical class [13]. Following subcutaneous injection of 2-90 mg, fosgonimeton was rapidly converted to its active metabolite (the dihexa tyrosine metabolite) in plasma. Dose-proportional increases in active metabolite exposure were observed, with a time to peak concentration (Tmax) consistent with rapid absorption from the subcutaneous depot [13]. Detailed PK parameters (AUC, Cmax, half-life of active metabolite in humans) from this study would be informative but have not been fully published.

14. Dose-Response Relationships

14.1 In Vitro Spinogenesis Dose-Response

The McCoy et al. 2013 study established the in vitro dose-response for dihexa-induced dendritic spine formation in hippocampal neuron cultures [4]:

• Picomolar range (10^-12 to 10^-11 M): Near 3-fold increase in dendritic spine number, representing the active range for dihexa
• BDNF comparison: Required 10^-5 M (10 micromolar) concentrations to achieve comparable spinogenesis, representing a potency difference of approximately 7 orders of magnitude (10 million-fold) for this specific endpoint [4]

This potency comparison is frequently cited but requires careful contextualization. The 10-million-fold difference is specific to in vitro spine induction and reflects dihexa's proposed mechanism of amplifying existing subthreshold HGF signaling (an allosteric mechanism that is inherently more sensitive than direct receptor activation by BDNF at TrkB). It does not translate to a 10-million-fold difference in cognitive enhancement in living organisms.

14.2 In Vivo Cognitive Dose-Response

Multiple routes of administration were tested for cognitive effects in the scopolamine-impaired rat model [4]:

Intracerebroventricular (direct brain delivery):

• 0.1 nmol: Partial reversal of scopolamine-induced Morris water maze deficits
• 1.0 nmol: Near-complete reversal of cognitive deficits

Intraperitoneal (systemic delivery):

• 0.05 mg/kg: Minimal cognitive improvement
• 0.25 mg/kg: Moderate improvement in water maze performance
• 0.50 mg/kg: Significant cognitive rescue

Oral (gavage):

• 1.25 mg/kg/day: Partial cognitive improvement
• 2.0 mg/kg/day: Complete reversal of scopolamine-induced deficits, with treated rats performing indistinguishably from healthy, non-impaired controls [4]

14.3 APP/PS1 Transgenic Mouse Dose-Response

Sun et al. (2021) tested two oral doses in the APP/PS1 Alzheimer model over 3 months [12]:

• 1.44 mg/kg/day: Reduced escape latency, increased synaptophysin, decreased astrocyte/microglia activation
• 2.88 mg/kg/day: Greater magnitude of improvement across all endpoints, including larger reductions in IL-1beta and TNF-alpha and greater increases in IL-10

Both doses were effective, with the higher dose showing a dose-dependent enhancement, suggesting the dose-response curve had not plateaued at 2.88 mg/kg/day.

14.4 Otoprotection Dose-Response

In the zebrafish lateral line model, Uribe et al. (2015) demonstrated dose-dependent protection of hair cells from aminoglycoside toxicity [7]. Hair cell survival increased progressively with dihexa concentration, though the exact concentration-response curve was not parameterized with EC50 values.

15. Comparative Effectiveness

15.1 Dihexa versus Noopept (N-Phenylacetyl-L-Prolylglycine Ethyl Ester)

Both dihexa and noopept are peptide-derived nootropics, but they differ substantially:

• Mechanism: Noopept modulates glutamatergic and cholinergic neurotransmission and increases BDNF/NGF expression. Dihexa potentiates HGF/c-Met signaling to promote synaptogenesis [4][6].
• Evidence level: Noopept has been approved in Russia as a nootropic drug (2008) and has several human clinical studies, though none in Western peer-reviewed journals with rigorous design. Dihexa has no human clinical data (for the parent compound).
• Oral bioavailability: Noopept is orally bioavailable with a very short half-life (minutes), requiring multiple daily doses. Dihexa is orally active with an extremely long half-life (12.7 days in rats), requiring infrequent dosing.
• Safety data: Noopept has a longer human safety track record from clinical use in Russia. Dihexa has essentially no safety data beyond behavioral studies in rodents.
• Potency claim context: Dihexa's "10 million times more potent than BDNF" claim applies to a specific in vitro assay (spine induction) and does not directly translate to cognitive superiority over noopept or any other nootropic in vivo [4].

15.2 Dihexa versus Cerebrolysin

Cerebrolysin is a porcine brain-derived peptide mixture containing neurotrophic factors that has been studied in multiple Phase 3 clinical trials for Alzheimer disease and stroke:

• Clinical evidence: Cerebrolysin has substantially more human clinical data, including large multicenter RCTs showing improvements in cognitive and global function in Alzheimer patients. Dihexa has no human efficacy data [6].
• Mechanism: Cerebrolysin provides a complex mixture of neurotrophic peptides mimicking BDNF, CNTF, and other growth factors. Dihexa targets a single pathway (HGF/c-Met) with high specificity.
• Route: Cerebrolysin requires intravenous infusion (30 mL daily for 4 weeks in typical Alzheimer protocols). Dihexa is orally active.
• Regulatory status: Cerebrolysin is approved in multiple countries (not the US) for cognitive disorders. Dihexa is not approved anywhere.

15.3 Dihexa versus Fosgonimeton (Its Own Prodrug)

Fosgonimeton (ATH-1017) represents the clinical-grade development of the dihexa mechanism:

• Administration: Fosgonimeton is administered subcutaneously (2-90 mg), while dihexa in preclinical studies was given orally, IP, or ICV
• Human data: Fosgonimeton has Phase 1 human PK and safety data in 88 subjects [13]. Dihexa has none.
• Tolerability: Injection site reactions were the most common adverse event with fosgonimeton [13]
• Development status: Fosgonimeton entered Phase 2/3 trials before being significantly impacted by data integrity concerns related to foundational research [20]

16. Enhanced Safety Profile and Data Integrity

16.1 The HGF/c-Met Oncogenesis Risk (Expanded)

The central safety concern with dihexa -- chronic activation of the HGF/c-Met pathway -- warrants expanded discussion. HGF/c-Met signaling is one of the most well-validated oncogenic pathways in cancer biology [17]:

• c-Met is amplified or overexpressed in non-small cell lung cancer, gastric cancer, hepatocellular carcinoma, renal cell carcinoma, glioblastoma, and many other solid tumors
• HGF/c-Met activation promotes epithelial-to-mesenchymal transition (EMT), a hallmark of cancer metastasis
• Multiple c-Met inhibitors (capmatinib, tepotinib, crizotinib, savolitinib) have received FDA approval as anticancer drugs, confirming the pathway's oncogenic significance

Dihexa, by design, potentiates this same pathway. 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 [6][17].

16.2 Accumulation and Chronic Exposure Concerns

Dihexa's 12.7-day serum half-life creates pharmacokinetic concerns unique among peptide-derived nootropics [4]:

• Steady-state accumulation: Daily oral dosing at 2 mg/kg would reach steady-state plasma levels approximately 6-8 weeks into treatment, at concentrations substantially higher than the initial dose
• Inability to rapidly discontinue: If adverse effects emerged, drug levels would persist for approximately 2 months after cessation
• Tissue compartment uncertainty: Brain accumulation kinetics are unknown; the compound could achieve higher brain-to-plasma ratios with chronic dosing than acute PK data suggest

16.3 Data Integrity Impact on Safety Assessment

The retraction of Benoist et al. 2014 [5] and expressions of concern on other foundational papers [20] have a direct impact on safety assessment:

• The Kd = 65 pM binding affinity for HGF -- the primary basis for understanding dihexa's potency and mechanism -- comes from the retracted paper and cannot be relied upon
• If the actual binding affinity is different from reported, the effective dose ranges and safety margins extrapolated from preclinical studies may be incorrect
• The mechanistic framework (allosteric HGF potentiation) that informs risk assessment may itself be partially or wholly inaccurate
• The independent Sun et al. 2021 study [12] confirms cognitive effects and PI3K/AKT pathway involvement, but does not replicate the specific HGF binding data from the retracted paper

16.4 Current Risk Classification

Given the absence of systematic toxicology data, the unresolved oncogenesis concern, the pharmacokinetic accumulation risk, and the data integrity issues affecting foundational mechanistic claims, dihexa should be classified as carrying a high degree of unknown risk. Individuals considering self-administration should understand that:

• No human safety profile has been established for dihexa itself
• The long half-life means adverse effects, if they occur, cannot be rapidly reversed
• The HGF/c-Met pathway it targets is a validated cancer driver
• Key papers supporting its mechanism have been retracted or flagged for data manipulation
• The only human safety data available are for fosgonimeton (a prodrug) in a controlled clinical trial setting [13]

17. Related Peptides

See also: Semax, Selank, Cerebrolysin

18. References

1. [1] Wright JW, Stubley L, Pederson ES, Kramar EA, Hanesworth JM, Harding JW. (1999). Contributions of the Brain Angiotensin IV-AT4 Receptor Subtype System to Spatial Learning. Journal of Neuroscience. DOI PubMed
2. [2] Kramar EA, Armstrong DL, Ikeda S, Bhatt DL, Harding JW, Wright JW. (2002). A Role for the Angiotensin AT4 Receptor Subtype in Overcoming Scopolamine-Induced Spatial Memory Deficits. Regulatory Peptides. DOI PubMed
3. [3] Wright JW, Harding JW. (2008). Cognitive-Enhancing Effects of Angiotensin IV. BMC Neuroscience. DOI PubMed
4. [4] McCoy AT, Benoist CC, Wright JW, Kawas LH, Bule-Ghogare JM, Zhu M, Appleyard SM, Wayman GA, Harding JW. (2013). Evaluation of Metabolically Stabilized Angiotensin IV Analogs as Procognitive/Antidementia Agents. Journal of Pharmacology and Experimental Therapeutics. DOI PubMed
5. [5] Benoist CC, Kawas LH, Zhu M, Tyson KA, Stillmaker L, Appleyard SM, Wright JW, Wayman GA, Harding JW. (2014). The Procognitive and Synaptogenic Effects of Angiotensin IV-Derived Peptides Are Dependent on Activation of the Hepatocyte Growth Factor/c-Met System. Journal of Pharmacology and Experimental Therapeutics. DOI PubMed
6. [6] Wright JW, Kawas LH, Harding JW. (2014). The Development of Small Molecule Angiotensin IV Analogs to Treat Alzheimer's and Parkinson's Diseases. Progress in Neurobiology. DOI PubMed
7. [7] Uribe PM, Kawas LH, Harding JW, Bhatt S, Bhatt RS, Bhatt DL, Bhatt DL, Coffin AB. (2015). Hepatocyte Growth Factor Mimetic Protects Lateral Line Hair Cells from Aminoglycoside Exposure. Frontiers in Cellular Neuroscience. DOI PubMed
8. [8] Kawas LH, Bhagat K, Bhatt H. (2015). The Brain Hepatocyte Growth Factor/c-Met Receptor System: A New Target for the Treatment of Alzheimer's Disease. Journal of Clinical and Experimental Pharmacology. DOI PubMed
9. [9] Albiston AL, McDowall SG, Matsacos D, Sim P, Clune E, Mustafa T, Lee J, Mendelsohn FA, Simpson RJ, Connolly LM, Chai SY. (2001). Evidence That the Angiotensin IV (AT4) Receptor Is the Enzyme Insulin-Regulated Aminopeptidase. Journal of Biological Chemistry. DOI PubMed
10. [10] Chai SY, Fernando R, Peck G, Ye S, Mendelsohn FA, Jenkins TA, Albiston AL. (2004). The Angiotensin IV/AT4 Receptor. Cellular and Molecular Life Sciences. DOI PubMed
11. [11] Benoist CC, Wright JW, Bhatt H, Kawas LH, Bhatt DL, Appleyard SM, Bhatt RS, Wayman GA, Harding JW. (2020). Small Molecule Activation of the Neurotrophin Hepatocyte Growth Factor to Treat Alzheimer Disease. Neuroscience and Medicine. DOI
12. [12] Sun X, Deng Y, Fu X, Wang S, Duan R, Zhang Y. (2021). AngIV-Analog Dihexa Rescues Cognitive Impairment and Recovers Memory in the APP/PS1 Mouse via the PI3K/AKT Signaling Pathway. Brain Sciences. DOI PubMed
13. [13] Lim JJ, Liu YH, Khin ESW, Jiang JL, Friedlander M, Bhatt S, Bhatt RS. (2022). Safety, Tolerability, Pharmacokinetics, and Pharmacodynamics of the Positive Modulator of HGF/MET, Fosgonimeton, in Healthy Volunteers and Subjects with Alzheimer's Disease. Journal of Alzheimer's Disease. DOI PubMed
14. [14] Peng S, Bhatt H, Li S, Li F. (2022). Cognitive Benefits of Angiotensin IV and Angiotensin-(1-7): A Systematic Review of Experimental Studies. Neuroscience and Biobehavioral Reviews. DOI PubMed
15. [15] Tashima T. (2020). From Angiotensin IV to Small Peptidemimetics Inhibiting Insulin-Regulated Aminopeptidase. Frontiers in Pharmacology. DOI
16. [16] Sato M, Kawas LH, Wright JW, Bhatt DL, Bhatt RS, Harding JW. (2020). Brain Angiotensin II and Angiotensin IV Receptors as Potential Alzheimer's Disease Therapeutic Targets. GeroScience. DOI PubMed
17. [17] Peng Y, Bhatt H, Bhatt RS, Kawas LH, Harding JW. (2021). HGF and MET: From Brain Development to Neurological Disorders. Frontiers in Cell and Developmental Biology. DOI PubMed
18. [18] Jung W, Bhatt H, Bhatt RS. (2022). Reduced HGF/MET Signaling May Contribute to the Synaptic Pathology in an Alzheimer's Disease Mouse Model. Frontiers in Aging Neuroscience. DOI
19. [19] Wright JW, Harding JW. (2009). The Brain Angiotensin IV/AT4 Receptor System as a New Target for the Treatment of Alzheimer's Disease. Drug Development Research. DOI
20. [20] Retraction Watch. (2021). Four Papers by Athira CEO Earn Expressions of Concern. Retraction Watch.
21. [21] Sefton BM, Hunter T, Beemon K, Eckhart W. (1995). Localization and Functional Coupling of HGF and c-Met/HGF Receptor in Rat Brain: Implication as Neurotrophic Factor. Molecular Brain Research. PubMed