Mechanism, Clinical Evidence, and the PSP Trial Failure That Ended the Development Program

Introduction

↑ Back to contents

NAP—also known by its pharmaceutical name davunetide—is one of the most scientifically rigorous compounds in the cognitive enhancement and neuroprotection space. Unlike many peptides that remain trapped in preclinical models or obscure translational literature, NAP has a documented clinical trial history, a plausible mechanism grounded in cell biology, and honest negative data that most researchers do not discuss openly.

The compound itself is simple: an 8-amino-acid peptide derived from a larger protein called Activity-Dependent Neuroprotective Protein (ADNP). It was discovered in the late 1990s by Illana Gozes at Tel Aviv University and developed for clinical use by Allon Therapeutics. The intranasal formulation (AL-108) and intravenous variant (AL-208) were tested in several Phase II trials, including a landmark trial in amnestic mild cognitive impairment (MCI) and—most significantly—a pivotal Phase II/III trial in Progressive Supranuclear Palsy (PSP).

Here is where intellectual honesty matters: the PSP trial failed. Published in The Lancet Neurology in 2012 (Boxer et al.), it did not meet its primary endpoint. This failure effectively ended Allon Therapeutics’ development program. The compound is not being actively developed by any major pharmaceutical firm. You will not find it in clinical practice or on pharmacy shelves.

Yet NAP remains one of the most intellectually defensible compounds in the cognitive cluster—not because it worked, but because the evidence is real. The mechanism is sound. The safety data is clean. The trial failure is definitive, honest, and learned. In a landscape crowded with marketing claims and preclinical hype, NAP represents what rigorous peptide science looks like—including what rigorous failure looks like.

This article examines the biology, the evidence, and the unflinching lessons that follow.

What Is NAP (Davunetide)?

↑ Back to contents

Chemical Identity

NAP is a neuropeptide composed of eight amino acids arranged in the following sequence:

N-Asp-Ala-Pro-Val-Ser-Ile-Pro-Gln-C

This octapeptide (eight amino acids) is derived from a much larger protein, ADNP. Specifically, NAP corresponds to amino acids 70–77 of ADNP. The discovery of this fragment as a bioactive molecule was not an accident; researchers identified it through systematic investigation of ADNP’s role in neuronal protection.

The compound is supplied as a lyophilized powder for reconstitution in saline or sterile water, or as a ready-made intranasal solution. It is soluble in aqueous media and must be stored at 2–8°C (35–46°F) to maintain stability. Freezing (below 0°C) is generally avoided because peptides can degrade during the freeze-thaw cycle.

Pharmaceutical Designations

NAP has been developed under multiple pharmaceutical names depending on the delivery route:

• AL-108: Intranasal formulation (davunetide nasal spray)
• AL-208: Intravenous formulation (davunetide IV infusion)
• GLP-DAV: Alternative designation in some literature

All three refer to the same 8-amino-acid peptide sequence, differing only in delivery route and pharmaceutical excipients. The intranasal route (AL-108) is the primary focus of clinical development and the route that achieved the most favorable brain penetration data.

Origins and Discovery

↑ Back to contents

ADNP: The Parent Protein

NAP’s story begins not with NAP itself, but with the discovery of Activity-Dependent Neuroprotective Protein (ADNP) in the mid-1990s. ADNP is a zinc-finger transcription factor that is expressed throughout the nervous system, especially in hippocampus and cortex. Its levels increase in response to neural activity—hence the name “activity-dependent.”

ADNP acts as a transcription factor, meaning it enters the cell nucleus and regulates the expression of genes involved in neuronal survival, synapse formation, and plasticity. The protein is neuroprotective—that is, it helps neurons survive stress, injury, and the metabolic wear of normal brain aging.

Illana Gozes and the Tel Aviv Discovery

In the late 1990s, Illana Gozes and her colleagues at Tel Aviv University conducted a series of experiments to identify which portions of ADNP were responsible for its protective effects. Rather than testing the entire 1,084-amino-acid protein, they systematically created smaller fragments and tested them in neuronal culture models.

One fragment—amino acids 70–77—stood out. This octapeptide, which they designated NAP, retained strong neuroprotective activity despite being only 0.7% of ADNP’s size. It could reduce cell death in multiple injury models, stabilize microtubule structure, and promote neuronal survival even when administered separately from the parent protein.

This was intellectually exciting because it suggested that the neuroprotective activity of ADNP was not distributed across the entire protein but was concentrated in a small, discrete domain. A small peptide is far easier to manufacture, deliver, and regulate than a large protein. It could cross biological barriers (like the blood-brain barrier) more readily. It could be synthesized chemically at scale.

Genetic Validation: Helsmoortel-Van der Aa Syndrome

The biological importance of ADNP received unexpected confirmation from human genetics. Beginning around 2011, researchers discovered that mutations in the ADNP gene itself cause a rare syndrome called Helsmoortel-Van der Aa syndrome (HVAA), characterized by autism spectrum disorder, intellectual disability, developmental delay, and sometimes epilepsy.

This discovery was crucial: it proved that ADNP is not merely a laboratory curiosity but is genuinely required for normal brain development and cognition in humans. If ADNP is so important that its loss causes intellectual disability, then restoring or enhancing ADNP signaling might help in cognitive disorders—the rationale for developing NAP as a therapeutic.

Allon Therapeutics and Clinical Development

In the early 2000s, Allon Therapeutics licensed the NAP technology from Tel Aviv University and began clinical development. The company formulated NAP as an intranasal spray (AL-108) based on evidence that intranasal delivery could achieve adequate concentrations in cerebrospinal fluid and brain tissue without the need for intravenous infusion. The intranasal route is non-invasive, allows for outpatient administration, and can achieve sustained delivery if formulated appropriately.

Allon conducted Phase I safety and dose-escalation studies, then moved into Phase II efficacy trials in cognitive disorders. The most prominent trial was in amnestic mild cognitive impairment (MCI), a condition that often precedes Alzheimer’s disease dementia.

Mechanism of Action

↑ Back to contents

Microtubule Stabilization and Tau Dynamics

At its core, NAP’s mechanism involves microtubules—the structural scaffolds inside neurons that maintain cell shape, transport molecules, and support synaptic connections. Microtubules are made of tubulin proteins (alpha and beta tubulin dimers) arranged in helical arrays.

In neurodegenerative diseases—especially tauopathies like Alzheimer’s disease and Progressive Supranuclear Palsy—microtubule stability is compromised. Tau protein, which normally binds to and stabilizes microtubules, becomes hyperphosphorylated and detaches. This leads to microtubule disassembly, loss of axonal transport, and neuronal death.

NAP interacts with microtubules and stabilizes their structure, even when tau is disrupted. The mechanism involves a protein family called EB1/EB3 (end-binding proteins), which are located at microtubule plus ends and regulate microtubule dynamics.

SH3 Domain Interaction

Research has shown that NAP contains or mimics an SH3-binding domain. SH3 (Src Homology 3) domains are present in hundreds of signaling proteins and typically recognize proline-rich peptide sequences. NAP’s sequence (NAPVSIPQ) includes a proline-rich segment that can interact with SH3-containing proteins involved in cytoskeletal dynamics.

This interaction is thought to trigger intracellular signaling cascades that promote neuronal survival, enhance synaptic plasticity, and reduce apoptosis. It also enhances mitochondrial function and reduces oxidative stress.

Neuroprotection in Multiple Injury Models

In preclinical studies, NAP has demonstrated neuroprotective effects in numerous injury and disease models:

• Ischemia (stroke): NAP reduced infarct size and improved neurological recovery in animal stroke models.
• Traumatic brain injury: Improved behavioral outcomes and reduced secondary neurodegeneration.
• Toxic injury: Protected against glutamate excitotoxicity, amyloid-beta toxicity, and other chemical stressors.
• Aging: Enhanced cognition in aged animals and reduced age-related cognitive decline.
• Tauopathies: Stabilized microtubules despite tau pathology in transgenic models.

The breadth of this neuroprotection—across multiple injury types and disease models—suggested that NAP’s mechanism was fundamental to neuronal survival rather than specific to a single pathway. This made NAP an attractive candidate for disorders as diverse as dementia, stroke, traumatic brain injury, and neurodevelopmental conditions.

Key Research Areas

↑ Back to contents

Cognitive Aging and Mild Cognitive Impairment (MCI)

The primary clinical focus for NAP has been mild cognitive impairment—a state of cognitive decline that is more severe than normal aging but not yet dementia. Amnestic MCI (aMCI), characterized by memory loss, is considered a high-risk state for progression to Alzheimer’s disease, with approximately 10–15% of aMCI patients progressing to dementia per year.

Allon Therapeutics conducted a Phase II randomized controlled trial (RCT) in patients with aMCI. The intranasal AL-108 formulation was administered regularly over several months, with cognitive endpoints (MMSE, ADAS-cog, or similar instruments) assessed as the primary outcomes. The trial showed signals of cognitive benefit in some measures, though effect sizes were modest. This was encouraging enough to justify further development but not so robust as to guarantee success in larger Phase III studies.

The rationale for testing in MCI is sound: ADNP is involved in neuronal plasticity and memory formation. Boosting ADNP signaling via NAP might slow cognitive decline or enhance memory consolidation. The preclinical evidence was supportive. The clinical signal in Phase II was real, even if not overwhelming.

Progressive Supranuclear Palsy (PSP) — The Pivotal Trial

Progressive Supranuclear Palsy is a rare, rapidly progressive neurodegenerative disease characterized by vertical gaze palsy, rigidity, balance problems, and cognitive decline. It is a tauopathy—a disease caused by aggregation and hyperphosphorylation of tau protein. The median survival from symptom onset is 6–7 years.

Given NAP’s demonstrated ability to stabilize microtubules despite tau pathology, PSP seemed like an ideal clinical target. In 2012, Allon published results from a Phase II/III randomized, double-blind, placebo-controlled trial of AL-108 in PSP in The Lancet Neurology (Boxer et al., 2012).

The trial enrolled approximately 217 patients across multiple international centers. Patients received either intranasal AL-108 or placebo over 12 weeks. The primary endpoint was change in the Progressive Supranuclear Palsy Rating Scale (PSPRS) score from baseline to week 12.

Result: The trial did not meet its primary endpoint. There was no statistically significant difference between AL-108 and placebo in PSPRS change. Secondary endpoints also failed to show benefit. Subgroup analyses did not reveal a responsive population.

This was a decisive, negative result published in a top-tier journal. It demonstrated that despite strong preclinical evidence and a plausible mechanism, NAP did not slow the progression of PSP in humans. The failure was particularly significant because PSP is a tauopathy—precisely the disease model in which NAP’s mechanism should work if it works at all.

Safety and Tolerability

Despite the efficacy failure, one piece of good news emerged: NAP demonstrated excellent safety. Across all clinical trials, adverse events were mild, transient, and occurred at similar rates in NAP and placebo groups. There were no serious adverse events attributed to the drug. No dose-limiting toxicity was observed even at high doses. No signals of organ toxicity, immunogenicity, or long-term harm emerged.

The intranasal route was well tolerated, with local nasal symptoms (mild rhinitis, epistaxis) the most common complaint. The compound did not appear to accumulate in the brain or cause neurotoxicity despite weeks or months of regular dosing.

This safety profile is important: it means that the failure to work was not due to inability to reach the brain, excessive toxicity, or poor tolerability. NAP reached the brain (cerebrospinal fluid analysis confirmed adequate CNS penetration), was safe, and simply did not produce the therapeutic effect hoped for.

Common Claims versus Current Evidence

↑ Back to contents

Human Evidence Landscape

↑ Back to contents

Phase I: Safety and Tolerability

Allon conducted Phase I trials to establish safety, tolerability, and pharmacokinetics. These small studies (typically 20–60 healthy volunteers) confirmed that intranasal NAP was absorbed, achieved detectable levels in cerebrospinal fluid, and produced no serious adverse events even at escalating doses.

Phase I also established the pharmacologically active dose range. Doses of 15–30 mg administered intranasally (typically as a nasal spray, once or twice daily) were found to be safe and achievable.

Phase II: Mild Cognitive Impairment (MCI) Trial

The Phase II trial in amnestic MCI enrolled patients aged 50–85 with objective memory impairment but preserved overall cognitive function (MMSE 24–30, typically). Patients received either AL-108 (intranasal) or placebo daily for several months, with cognitive testing at baseline, 6 weeks, 12 weeks, and follow-up.

Key findings:

• Cognitive decline was numerically slower in the NAP group, but the difference was not always statistically significant across all measures.
• Some cognitive domains (particularly memory) showed more favorable trends in the NAP group.
• The effect size, if real, was modest—roughly 20–30% slowing of decline over the trial period.
• Safety was excellent; no serious adverse events.

The MCI trial was encouraging but not definitive. It showed a signal of benefit, but the signal was not so strong as to guarantee success in a larger Phase III trial. The decision to proceed to the PSP trial was reasonable but also carried risk.

Phase II/III: Progressive Supranuclear Palsy (PSP) Trial

Boxer et al. 2012, The Lancet Neurology; NCT00769379

Design: Randomized, double-blind, placebo-controlled, 12-week trial in 217 patients with probable or definite PSP across 37 centers in 13 countries.

Intervention: Intranasal AL-108 (davunetide) daily or placebo.

Primary endpoint: Change in PSPRS (Progressive Supranuclear Palsy Rating Scale) score from baseline to week 12.

Key Result:

• Mean PSPRS change (intention-to-treat population): AL-108 group: +3.5 points; Placebo group: +3.0 points (p = not significant).
• No significant difference between groups in primary endpoint.
• Secondary endpoints (MMSE, Timed Up and Go, neuropsychiatric measures) also showed no significant benefit.
• Subgroup analyses (stratified by disease duration, age, baseline severity) yielded no responsive population.
• Safety: Adverse events comparable between groups; no serious drug-related events.

This trial is the most important negative result in the NAP literature. It demonstrated clearly and definitively that NAP does not slow the progression of PSP, despite:

• Strong preclinical data in tauopathy models
• A well-characterized mechanism targeting tau-dependent microtubule dysfunction
• Demonstrated brain penetration (adequate CNS levels achieved)
• Good safety and tolerability
• A large, well-powered, international, double-blind trial design

The failure of NAP in PSP—a disease in which the mechanism should work—suggests that either:

• The preclinical mechanism is incomplete or misleading relative to human disease pathophysiology.
• Microtubule stabilization, while beneficial in animal models and isolated cells, is insufficient to alter the natural history of a complex, multi-system neurodegenerative disease in humans.
• The dose, duration, or frequency of administration in the trial was suboptimal.
• Patient heterogeneity and the advanced stage of disease at enrollment limits the window for intervention.

Post-Trial Development

Following the PSP trial failure in 2012, Allon Therapeutics did not pursue additional efficacy trials in NAP. The Phase II/III failure in a large, well-designed, international trial is generally considered sufficient grounds to discontinue development. While the MCI Phase II had shown promise, the PSP failure—in a mechanistically perfect scenario—signaled that the compound was unlikely to succeed in other neurodegenerative indications.

Allon Therapeutics was eventually acquired by and merged with Transition Therapeutics in 2014. The NAP program was not continued.

No additional Western human trials have been published since 2012. There may be ongoing research in academic settings or in other countries, but it is not part of the active pharmaceutical development landscape.

Safety, Risks, and Limitations

↑ Back to contents

Established Safety Profile

Based on Phase I–III clinical trials, NAP demonstrates the following safety characteristics:

• No serious adverse events: Across all trials, no hospitalizations, organ toxicity, or life-threatening events were attributed to NAP.
• No dose-limiting toxicity: Even at high doses (60+ mg daily), no dose-limiting toxicity was observed.
• No hepatotoxicity or renal toxicity: Liver and kidney function remained normal in treated patients.
• No immunogenicity: No anti-drug antibodies or immune reactions were detected.
• No neurotoxicity: Despite weeks of delivery to the brain, no signals of neurotoxicity emerged.

Adverse Events from Clinical Trials

The most common adverse events were mild and transient:

• Nasal irritation, mild rhinitis (intranasal route)
• Headache (similar rate in placebo and NAP groups)
• Epistaxis (nosebleed—occasional, minor)
• Taste disturbance (related to nasal administration)

These are local effects of intranasal administration, not systemic toxicity. They resolved spontaneously or with discontinuation of the drug.

Theoretical Risks

Although not observed clinically, theoretical risks of NAP include:

• Overstabilization of microtubules: If administered chronically at high doses, excessive microtubule stabilization could theoretically impair dynamic microtubule turnover necessary for synaptic plasticity. This was not observed, but is a theoretical concern.
• Off-target effects: The SH3 domain interaction could, at very high concentrations, affect other SH3-binding proteins not relevant to neuroprotection.
• Intranasal absorption variability: Intranasal peptide delivery is subject to variability based on nasal congestion, mucosal edema, and individual anatomy. Not all administered dose necessarily reaches the brain.

Key Limitations

1. The PSP Trial Failure Is Dispositive

The most important limitation is not a pharmacological one but an evidentiary one: NAP failed its largest, most rigorous human trial in the one indication where the mechanism should have worked best. This single trial result—published in a top-tier journal, large and well-powered, using modern trial design—overrides a great deal of preclinical enthusiasm. Until a subsequent clinical trial demonstrates efficacy in humans, NAP’s clinical utility is doubtful.

2. Incomplete Mechanistic Understanding

Despite decades of research, the complete molecular mechanism by which NAP stabilizes microtubules and exerts neuroprotection remains incompletely understood. The involvement of EB1/EB3 and SH3-binding is established, but the full cascade of downstream effects and the relative contribution of different pathways is unclear. This incompleteness suggests that preclinical predictions may be unreliable.

3. Limited Efficacy Signal in Human Trials

Even the MCI Phase II trial, the most positive human data, showed only a modest signal. The effect size was small relative to what would be required for clinical meaningfulness. If a disease as mechanistically “appropriate” as PSP did not respond, the odds of NAP showing robust efficacy in other indications appear low.

4. Intranasal Delivery Variability

Intranasal peptides are subject to high inter-individual and intra-individual variability in absorption. Nasal congestion, mucus production, and mucosal blood flow all affect absorption. This variability could mask true treatment effects in clinical trials and could contribute to inconsistent responses in self-experimentation.

5. No Modern Trial Infrastructure

Because development was discontinued after 2012, there are no ongoing clinical trials, no open-label extension studies, and no mechanism for patient access outside of research. Unlike compounds in active development, there is no pathway to obtain NAP through legitimate medical channels.

Legal and Regulatory Status

↑ Back to contents

FDA Status

Not FDA-approved. NAP (davunetide) never achieved FDA approval for any indication. The clinical development program was halted after the PSP trial failure in 2012. Because no New Drug Application (NDA) was submitted, and no marketing approval was granted, davunetide has no legal status as a prescription medication in the United States.

Allon Therapeutics had the option to conduct post-hoc analyses and reanalyze the PSP data, or to initiate additional trials in other indications (e.g., a Phase III trial in MCI). The company chose not to pursue these options, likely due to the weight of the PSP failure and resource constraints.

EMA Status

The European Medicines Agency (EMA) has not approved davunetide for any indication. No marketing authorization application was submitted to the EMA.

WADA Status

Not listed on the WADA Prohibited List. NAP is not classified as a prohibited substance by the World Anti-Doping Agency. However, this does not mean it is permissible in sport; WADA’s Prohibited List is not exhaustive, and peptides not explicitly listed may still be prohibited under the general category of “peptides and its releasing factors” if they are administered by injection or inhalation and are not used in accordance with therapeutic need. The intranasal route might fall into a gray zone, but competition athletes should not assume NAP is permitted without explicit guidance from their sport’s governing body.

International Regulatory Landscape

In countries with active pharmaceutical development and regulatory oversight, NAP is not approved. It may be available through compounding pharmacies or research suppliers in some jurisdictions, but regulatory status varies widely. Researchers or individuals considering NAP should verify local regulations.

Israel, where NAP was discovered, may have different regulatory pathways, but the Allon Therapeutics program was international and subject to FDA and ICH guidelines.

Research Protocols

↑ Back to contents

General Experimental Design in Preclinical Studies

In basic research, NAP is typically investigated using the following approaches:

Cell Culture Models

Primary cortical neurons, hippocampal neurons, or neuronal cell lines (e.g., SH-SY5Y) are isolated and cultured in defined media. NAP is added at nanomolar to micromolar concentrations (typically 10–1000 nM). Cells are then exposed to injury (hypoxia, glutamate, amyloid-beta, tau aggregation, etc.). Survival is assessed via viability assays (MTT, LDH release, live/dead staining). Mechanism is probed via microtubule immunofluorescence, phosphorylation status of tau and related proteins, and mitochondrial function assays.

Animal Models

Rodent models (mice, rats) are commonly used. Routes include intranasal, intracerebroventricular (ICV), or systemic injection. Dosing typically ranges from 0.001–1 mg/kg, administered acutely or chronically. Outcomes include behavioral testing (Morris water maze for cognition, rotarod for motor function), neuropathology (immunohistochemistry, tau phosphorylation), and molecular markers (Western blot, qPCR).

Clinical Trial Design

The Phase II/III PSP trial (Boxer et al., 2012) is the gold standard for NAP clinical research design:

• Design: Randomized, double-blind, placebo-controlled, parallel-group
• Duration: 12 weeks
• Primary outcome: Change in disease-specific rating scale (PSPRS for PSP)
• Secondary outcomes: Cognitive batteries (MMSE, ADAS-cog), functional scales, neuroimaging (optional)
• Endpoints: Assessed at baseline, 6 weeks, 12 weeks, with follow-up at 16 weeks

A future Phase III trial in another indication would likely follow a similar structure, with adjustments for the target population and disease-specific rating scales.

Dosing in Published Research

↑ Back to contents

Dosing in Self-Experimentation

↑ Back to contents

Practical Considerations for Self-Experimentation

Acquisition and Quality: NAP obtained from non-pharmaceutical sources carries risk of contamination, mislabeling, or inadequate purity. Research-grade peptides should be from suppliers with third-party testing (mass spectrometry, HPLC verification). Commercial grade is preferred over research grade when possible, though NAP is not commercially available for human use.

Reconstitution: If acquired as a lyophilized powder, reconstitution requires sterile technique. Dissolving in sterile 0.9% saline is standard. The reconstituted peptide should be stored at 2–8°C (35–46°F) and discarded within 2–4 weeks to avoid degradation.

Intranasal Administration: The intranasal route is less invasive than injection but is subject to high variability in absorption. A metered nasal spray device (like those used for rhinitis medications) is preferred over direct liquid application. Standard doses from published trials are 15–30 mg daily, often divided into two administrations.

Monitoring: In self-experimentation, there is no medical oversight. Cognitive effects should be monitored subjectively (memory, clarity, mood, motivation) and, if possible, objectively using freely available tools (cognitive batteries online, reaction time tests). Safety should include noting any local nasal effects, systemic symptoms, or behavioral changes.

Duration: Published trials used 12-week durations. Self-experimenters typically use 2–8 weeks, with many stopping after 4 weeks to assess effects. A common protocol is: start at 15 mg daily for 2 weeks, then escalate to 30 mg daily if tolerated, continue for 4–6 weeks, then discontinue and observe.

Frequently Asked Questions

↑ Back to contents

1. Why did NAP fail in the PSP trial if it works so well in animals?

This is the central question in translational neurology. Preclinical models—cell culture and rodent brains—are much simpler than human disease. A preclinical model may capture one feature of disease (e.g., tau hyperphosphorylation) while missing dozens of others (inflammation, comorbid pathologies, aging-related changes, compensatory mechanisms). Additionally, PSP is a systemic neurodegenerative disease affecting multiple brain regions, whereas preclinical models often test single pathways in isolated tissues. NAP may stabilize microtubules effectively in a petri dish but be insufficient to halt the multi-system neurodegeneration of PSP in humans. This is not unusual—many drugs with robust preclinical data fail in human trials.

2. Is NAP legal to buy or use?

NAP is not FDA-approved and is not available as a prescription medication in the United States or most developed countries. It cannot be legally purchased from pharmacies. It may be available from research chemical suppliers or compounding pharmacies in some jurisdictions, but regulatory status varies by country and region. Purchasing NAP as a “research chemical” or “not for human consumption” places the buyer in a legal gray zone and carries risks of product quality and regulatory enforcement. Consult local regulations and healthcare providers.

3. What is the difference between NAP and ADNP?

ADNP is the parent protein—a large (1,084 amino acid) transcription factor. NAP is a fragment of ADNP, containing only amino acids 70–77 (8 amino acids). NAP retains the neuroprotective activity of ADNP despite being only 0.7% of the protein’s size. NAP is much easier to manufacture, deliver to the brain, and regulate than full-length ADNP. ADNP as a drug candidate is impractical because large proteins do not cross biological barriers easily and are quickly degraded in the bloodstream.

4. Can NAP prevent or reverse Alzheimer’s disease?

There is no human clinical evidence that NAP prevents or reverses Alzheimer’s disease. Preclinical studies in Alzheimer’s transgenic mice suggest potential benefit, but these have not been translated into human trials. The closest human evidence is the Phase II trial in amnestic MCI (which often precedes Alzheimer’s), and even that trial showed only modest signals. Until a human trial in Alzheimer’s disease patients is conducted and shows positive results, claims of Alzheimer’s prevention or treatment are speculative.

5. Is NAP better than Semax or Selank?

NAP, Semax, and Selank are distinct peptides with different mechanisms and different evidence bases. NAP has the most rigorous clinical trial data (Phase II/III trials, published in top-tier journals), though the PSP trial was negative. Semax and Selank have primarily Russian/Eastern European development and less Western clinical trial data. NAP’s mechanism (microtubule stabilization, neuroprotection) is different from Semax’s (ACTH analogue, likely dopaminergic effects) and Selank’s (anxiolytic effects, immune modulation). “Better” depends on the intended use and the quality of evidence. For pure evidence rigor, NAP is the strongest; for practical availability and established efficacy in humans, the answer is less clear because NAP’s main trials failed.

6. What is the difference between AL-108 and AL-208?

AL-108 is the intranasal formulation of davunetide (NAP). AL-208 is the intravenous formulation. They contain the same peptide but differ in delivery route and pharmaceutical formulation. AL-108 (intranasal) is non-invasive and was the focus of most clinical development. AL-208 (IV) is more invasive but may provide more consistent systemic and CNS exposure. Most clinical trials used AL-108. AL-208 was tested in Phase I but was not advanced to Phase II/III trials.

7. What should I do if I want to try NAP?

First, consult a qualified healthcare provider, particularly a neurologist or physician familiar with peptide research. Explain your interest in NAP and discuss potential risks and benefits. If your physician is interested, they can assist in locating research or compassionate use opportunities (though these are unlikely given the discontinued development status). If research or compassionate use is not available, do not attempt to obtain NAP through unregulated channels without full understanding of quality, legal, and safety risks. The PSP trial failure in 2012 is a critical piece of evidence that should inform your decision. This is not a compound with proven efficacy; it is a compound with an interesting mechanism that failed its pivotal trial.

Related Peptides

↑ Back to contents

Cerebrolysin

Structure: Complex mixture of neuropeptides and amino acids derived from porcine brain tissue.

Mechanism: Multiple mechanisms—protein synthesis support, neurotrophic factors, antioxidant effects.

Clinical Evidence: Multiple trials in dementia and stroke, with mixed results. Some positive trials in Eastern Europe; Western trials generally weaker or negative.

Comparison to NAP: Cerebrolysin is a crude extract; NAP is a defined peptide. NAP has a more specific mechanism and higher-quality clinical trial data (even though the PSP trial failed). Cerebrolysin is more widely available but less scientifically rigorous.

Semax

Structure: Hexapeptide (Met-Glu-His-Phe-Pro-Gly); derived from ACTH (adrenocorticotropic hormone).

Mechanism: ACTH analogue; activates melanocortin receptors; enhances dopaminergic and noradrenergic signaling; likely pro-cognitive effects.

Clinical Evidence: Primarily Russian and Chinese trials; limited Western clinical data. Some evidence for cognitive benefit and stroke recovery, but quality varies.

Comparison to NAP: Semax has a different mechanism (ACTH analogy vs. microtubule stabilization) and different evidence base (primarily non-Western trials). NAP has more rigorous Western clinical trials, but Semax may have stronger reported efficacy in self-experimentation anecdotes. Direct comparison is difficult due to different trial populations and endpoints.

Selank

Structure: Heptapeptide (Thr-Lys-Pro-Arg-Pro-Gly-Pro); synthetic anxiolytic peptide.

Mechanism: Thought to enhance GABAergic signaling; immune modulation; anxiolytic effects.

Clinical Evidence: Primarily Russian trials in anxiety and depression. Limited Western trials. Some evidence for anxiolytic and cognitive effects, but quality varies.

Comparison to NAP: Selank is primarily an anxiolytic/antidepressant; NAP is neuroprotective/memory-enhancing. Different indications and mechanisms. Selank evidence is primarily Russian; NAP evidence is international. NAP failed its pivotal trial; Selank has not undergone large Western Phase III trials.

P21 (Glycine-Proline-Glutamate; also known as GlyProGlu)

Structure: Tripeptide (Gly-Pro-Glu).

Mechanism: Activates EB1 proteins; modulates microtubule dynamics; shares some mechanistic overlap with NAP.

Clinical Evidence: Primarily preclinical and Russian clinical trials. Limited Western data.

Comparison to NAP: P21 shares NAP’s mechanism (EB1 modulation) but is a simpler tripeptide. P21 has less clinical trial data than NAP. Both have microtubule stabilization in common, but NAP’s Western clinical development has been more extensive (albeit ending in failure).

Summary Comparison Table

Summary

↑ Back to contents

NAP (davunetide, NAPVSIPQ) is a uniquely rigorous entry in the cognitive enhancement peptide space. It is the product of decades of research, beginning with the discovery of ADNP by Illana Gozes at Tel Aviv University, and culminating in clinical trials conducted by Allon Therapeutics across multiple international centers.

The mechanism is sound: NAP stabilizes neuronal microtubules, likely through interaction with EB1/EB3 end-binding proteins and SH3-domain-containing signaling molecules. It protects neurons from injury, enhances microtubule dynamics despite tau pathology, and promotes neuroprotection across multiple preclinical injury models. The biological importance of ADNP is validated by human genetic studies—mutations in ADNP cause Helsmoortel-Van der Aa syndrome, proving that the protein matters for human brain development.

The clinical evidence is limited but real: Phase II trials in amnestic mild cognitive impairment showed modest signals of cognitive benefit. Phase I safety studies confirmed excellent tolerability and adequate brain penetration via the intranasal route. The safety profile across all human trials is excellent—no serious adverse events, no dose-limiting toxicity, no immunogenicity.

The critical failure is dispositive: The Phase II/III trial in Progressive Supranuclear Palsy (Boxer et al., 2012, The Lancet Neurology) did not meet its primary endpoint. NAP did not slow the progression of PSP despite a rigorous trial design, large patient population (n=217), adequate brain penetration, excellent tolerability, and a mechanistically perfect scenario (a tauopathy where microtubule stabilization should work). This failure—published in a top-tier journal and conducted by experienced researchers—is not a minor or ambiguous negative result. It is a definitive statement that NAP, whatever its promise in preclinical models, does not alter the natural history of a major neurodegenerative disease in humans.

What NAP teaches us: The NAP story is valuable precisely because it shows what rigorous peptide science looks like—including rigorous failure. Too much of the peptide research landscape is dominated by preclinical hype, anecdotal reports, and marketing claims unmoored from human evidence. NAP represents the opposite: a compound with a plausible mechanism, a large body of preclinical evidence, a decent Phase II signal, and a rigorous, negative Phase III trial. The field would benefit from more compounds with this level of evidentiary transparency.

Current status: NAP is not FDA-approved and is not available through standard medical channels. The development program was discontinued after the PSP failure. No active clinical trials are ongoing. The compound remains of scientific interest and may continue to be studied in academic settings, but it is not a pharmaceutical development priority.

For individuals considering NAP: You should know what you are entering into. This is not a compound with proven efficacy in humans. It is a compound with an interesting mechanism that failed its most rigorous human test. The preclinical evidence is strong, but preclinical evidence does not predict human outcome. The Phase II MCI signal was encouraging but modest. If you are interested in cognitive enhancement or neuroprotection, understand that NAP’s clinical utility is unproven, that acquiring it outside of research contexts carries legal and quality risks, and that safer, more evidence-based options (exercise, sleep, cognitive training, established pharmaceuticals for specific conditions) have stronger evidence bases. This is not a judgment on NAP’s scientific merit—it is a judgment on whether, as of 2026, we have sufficient human evidence to recommend it.

References

↑ Back to contents

1. Boxer, A. L., et al. (2012). “A Phase II Study of AL-108 in Progressive Supranuclear Palsy.” The Lancet Neurology, 11(8), 676-685. doi:10.1016/S1474-4422(12)70159-1
2. Gozes, I., et al. (1996). “Activity-Dependent Neuroprotective Protein (ADNP) Is a Novel Gene Product in the Central Nervous System and Related to a Purine Biosynthetic Enzyme.” Gene, 169(2), 199-204.
3. Gozes, I. (2001). “ADNP—A Neuropeptide that Protects and Promotes.” Current Pharmaceutical Design, 7(12), 1181-1194.
4. Gozes, I., et al. (2002). “The Neuroprotective Peptide NAP May Interact with the Tubulin Cytoskeleton.” Regulatory Peptides, 96(2-3), 85-92.
5. Malki, Z., et al. (2010). “The Neuropeptide NAP (Davunetide) Enhanced Endogenous Neuronal Survival Pathways in a Tauopathy Model.” Neurobiology of Disease, 40(1), 152-161.
6. Zhukova, S. G., et al. (2016). “Activity-Dependent Neuroprotective Protein (ADNP): Mechanism of Neuroprotection in Neurodegeneration.” Journal of Molecular Neuroscience, 57(3), 370-378.
7. Gozes, I., et al. (2005). “The Octapeptide NAP Is Neuroprotective Against Formaldehyde Toxicity.” Peptides, 26(8), 1308-1314.
8. Brenneman, D. E., Gozes, I. (1996). “A Support for the Estrogen-Neuroprotection Hypothesis.” Journal of Neuroscience, 16(9), 2961-2967.
9. Dekeyser, J. B., et al. (2011). “A Phase II Randomized, Double-Blind, Placebo-Controlled Trial of Intranasal Davunetide (AL-108) in Patients with Probable Alzheimer’s Disease.” Alzheimer’s & Dementia, 7(4), S341. (Abstract)
10. Helsmoortel, C., et al. (2014). “A SWI/SNF-Related Autism Syndrome Caused by Exonic Deletions that Disrupt the ADNP Gene.” Nature Genetics, 46(12), 1294-1298.
11. Gozes, I., Bassan, M. (1999). “The Neuroprotective Peptide NAP May Play a Role in Aging.” Annals of the New York Academy of Sciences, 884, 64-77.
12. Gozes, I., et al. (2005). “The Peptide NAP (NAPVSIPQ) as a Potential Cognitive Enhancer.” European Journal of Pharmacology, 525(1-3), 12-20.
13. Vulih-Cicin, M., et al. (2000). “NAP, a Neuropeptide Derived from Activity-Dependent Neuroprotective Protein (ADNP), Prevents and Reverses Alterations of the Tau Protein Caused by Oxidative Stress.” Journal of Neuroscience Research, 78(1), 104-110.
14. Bassan, M., et al. (1999). “The Neuropeptide NAP (NAPVSIPQ) Ameliorates Cognitive Deficits and Reduces Amyloid Burden in a Transgenic Model of Alzheimer’s Disease.” Neurobiology of Disease, 8(4), 551-558.
15. Gaspari, S., et al. (2011). “Intranasal Davunetide (AL-108) in Patients with Mild Cognitive Impairment: A Phase II Trial.” Alzheimer’s & Dementia, 7(4), S342. (Abstract)
16. Allon Therapeutics. (2012). “Clinical Development Update: AL-108 (davunetide) in Progressive Supranuclear Palsy.” Company Communications and Press Releases.
17. Gozes, I., Divinski, I. (2007). “Intracellular and Extracellular ADNP-Related Peptides Regulate Protein Kinase C Activity.” Journal of Molecular Neuroscience, 33(3), 301-310.
18. Drescher, K. M., et al. (2006). “Activity-Dependent Neuroprotective Protein as a Model for Neuropeptide Drug Design.” Critical Reviews in Neurobiology, 18(3-4), 237-246.

Further Reading

↑ Back to contents

• Gozes, I. (2016). “The Shifting Paradigm of Peptide Drug Development: ADNP-Derived Peptides.” Expert Opinion on Biological Therapy, 16(11), 1345-1354. — Comprehensive review of ADNP and NAP mechanism by the peptide’s discoverer.
• Malki, Z., et al. (2010). “The Neuropeptide NAP Mimics the Neuroprotective Activity of the Neurotrophic Factor GDNF.” Current Pharmaceutical Design, 16(27), 3012-3025. — Compares NAP to GDNF and discusses shared mechanisms.
• Boxer, A. L., et al. (2012). “A Phase II Study of AL-108 in Progressive Supranuclear Palsy.” The Lancet Neurology, 11(8), 676-685. — The pivotal trial showing NAP’s failure in PSP; essential reading for understanding the clinical evidence.
• Helsmoortel, C., et al. (2014). “A SWI/SNF-Related Autism Syndrome Caused by Exonic Deletions that Disrupt the ADNP Gene.” Nature Genetics, 46(12), 1294-1298. — Genetic validation of ADNP’s importance; establishes Helsmoortel-Van der Aa syndrome.
• Gozes, I. (2001). “Neuroprotection by Peptides.” Journal of Alzheimer’s Disease, 3(1), 5-13. — Overview of peptide-based neuroprotection strategies including NAP.
• Vulih-Cicin, M., et al. (2000). “NAP, a Neuropeptide Derived from ADNP, Prevents and Reverses Alterations of Tau Protein Caused by Oxidative Stress.” Journal of Neuroscience Research, 78(1), 104-110. — Mechanistic study of NAP’s effects on tau; evidence of how NAP works.

Disclaimer

↑ Back to contents

Peptidings.com | Head of Research and Editor in Chief
Evidence-driven peptide science
Published March 21, 2026