Dihexa for TBI, Concussion & Stroke Recovery: The Evidence in 2026

Brain injury — whether from a single concussion, repeated sub-concussive impacts, a severe traumatic brain injury (TBI), or an ischaemic stroke — leaves the brain in a state of disrupted connectivity that often recovers partially, slowly, and unevenly. Dihexa is a synaptogenic peptide whose mechanism (HGF/c-Met activation) overlaps with several of the endogenous repair pathways activated after brain injury. This has led to sustained interest in whether Dihexa could support recovery from TBI, concussion, post-concussion syndrome, or stroke. This 2026 review summarises what is actually known, what remains speculation, and where the evidence is genuinely absent.

Why Brain Injury Is a Biologically Interesting Question for Dihexa

To understand why Dihexa keeps reappearing in discussions about brain injury recovery, it helps to look at what the brain actually does after injury — and what Dihexa was designed to do.

Traumatic brain injury, concussion, and stroke share overlapping biology. In all three, an initial insult (mechanical, vascular, or both) triggers a cascade of secondary injury: excitotoxicity, mitochondrial dysfunction, oxidative stress, neuroinflammation, disrupted axonal transport, and — at the circuit level — the loss of synapses and the retraction of dendritic spines. Neurons may survive the initial event but lose their connectivity with neighbours, which is what translates into functional deficits: memory impairment, slowed processing, mood disturbance, word-finding problems, motor weakness.

Recovery, when it happens, is driven by a predictable set of endogenous repair mechanisms. Surviving neurons sprout new axon terminals and grow new dendritic spines. Silent synapses reactivate. Surviving networks reorganise to take over functions previously handled by damaged tissue. Neurogenesis (in restricted brain regions like the hippocampus) contributes modestly. Growth factors — brain-derived neurotrophic factor (BDNF), nerve growth factor (NGF), insulin-like growth factor 1 (IGF-1), and hepatocyte growth factor (HGF) — are upregulated at the injury site to orchestrate this repair.

Dihexa is a small-molecule mimetic of this last pathway. It activates the HGF/c-Met receptor system — the same system the brain upregulates itself after injury. The mechanistic logic is simple: if the brain is already trying to repair itself by turning on HGF signalling, what happens if you add a pharmacological agonist of that same pathway?

The full molecular detail is covered in the mechanism of action guide. For this article, what matters is that Dihexa is not a random nootropic layered on top of an injured brain. It is specifically targeting a growth-factor pathway that is biologically relevant to injury recovery. Whether that mechanistic relevance translates into real benefit is the question the rest of this article examines.

The HGF/c-Met Pathway After Brain Injury

Hepatocyte growth factor was originally named for its role in liver regeneration, but it is now recognised as a multifunctional trophic factor expressed throughout the brain. Its receptor, c-Met, is present on neurons, astrocytes, microglia, and endothelial cells. In the uninjured adult brain, HGF/c-Met signalling contributes to synaptic plasticity, spine stability, and hippocampal-dependent memory.

After injury, the picture changes substantially:

• Ischaemic stroke: HGF and c-Met are both rapidly upregulated in the penumbra — the tissue around the infarct core that is functionally silenced but not yet dead. Animal studies show that exogenous HGF reduces infarct size, promotes angiogenesis, and improves behavioural recovery.
• Traumatic brain injury: HGF expression increases around the lesion site within hours. Genetic or pharmacological enhancement of the pathway in rodent TBI models has been associated with reduced cell death, better preservation of cognitive function, and enhanced neurite outgrowth.
• Neuroinflammation: HGF modulates microglial activation, tending to shift microglia toward reparative phenotypes rather than chronic pro-inflammatory ones. Chronic microglial activation is a proposed driver of post-concussion and CTE pathology.
• Angiogenesis: HGF is a potent angiogenic factor. The formation of new capillary networks in recovering brain tissue is partly HGF-dependent — relevant because adequate blood supply underpins tissue recovery.

This converging biology is why Dihexa as a positive modulator of the same system has attracted attention from researchers and — separately — from the self-experimenting community. The academic case is compelling enough that a pharmaceutical company (Athira) developed an HGF enhancer through to Phase 3 clinical trials, though in Alzheimer's rather than injury indications. That clinical programme is discussed in the dedicated Fosgonimeton & Athira article.

Preclinical Evidence: What the Animal Studies Actually Show

No published study has tested Dihexa specifically in a rodent TBI or stroke model at the time of writing. This is the single most important evidential gap to understand. Claims about Dihexa and brain injury are built on two evidence chains: (1) Dihexa's performance in non-injury cognitive models and (2) HGF/c-Met or angiotensin IV activity in injury models using other compounds. Neither is the same as direct evidence.

Dihexa in Cognitive-Deficit Models

Dihexa's strongest preclinical data come from models of cognitive impairment — notably scopolamine-induced amnesia in rats and rodent models of aged cognitive decline. In these paradigms, Dihexa restored performance on spatial memory tasks such as the Morris water maze and object recognition tests. Synaptogenesis in the hippocampus was a reported downstream finding, consistent with its proposed mechanism.

These studies are reviewed in full in the research and studies page. The important caveat: chemically induced amnesia is a functional model of memory deficit, not a structural model of brain injury. The neurons are intact; acetylcholine signalling is transiently blocked. This differs fundamentally from TBI or stroke, where neurons die, axons are sheared, and glial responses persist for months.

HGF/c-Met Activation in Injury Models (Other Compounds)

A larger literature exists on HGF itself, or on other HGF pathway activators, in rodent models of brain injury:

• Ischaemic stroke (MCAO models): Intracerebral or intravenous HGF administered before or shortly after middle cerebral artery occlusion reduces infarct volume, attenuates brain oedema, and improves neurological scores in rats and mice. Effect sizes vary substantially with dose and timing.
• Controlled cortical impact TBI: Gene-therapy approaches expressing HGF at the contusion site have been associated with reduced lesion volume and preserved motor function in some studies.
• Spinal cord injury: HGF has been shown in rodent contusion models to promote axonal regeneration and functional recovery — relevant by analogy because the underlying biology (growth-cone dynamics, myelin inhibition, glial scar) overlaps.

Dihexa is not HGF. It is an HGF mimetic that signals through the same receptor. Whether it recapitulates the full spectrum of HGF's actions after injury — particularly those involving matrix binding, proteolytic activation, and paracrine glial effects — is unknown. Treating Dihexa as equivalent to HGF is a common but unsupported assumption.

The Angiotensin IV Family in Brain Injury

Dihexa was developed from angiotensin IV (Ang IV), a six-amino-acid fragment of angiotensin II with pro-cognitive properties. Ang IV itself, and its more stable analogues (including Dihexa's immediate precursors), have been tested in several brain injury paradigms:

• Intracerebroventricular Ang IV in rats has been reported to reduce ischaemic damage and improve cognitive recovery post-MCAO.
• Norleucine-1 angiotensin IV (Nle1-AngIV), a metabolically stabilised analogue, demonstrated pro-cognitive effects in rodent models broadly consistent with HGF/c-Met activation.
• The specific angiotensin-IV-derived peptide PNB-0408 (the formal designation of Dihexa) was engineered for oral bioavailability and blood-brain barrier penetration, making it the most promising systemic agent in this class.

A detailed breakdown of how Dihexa compares mechanistically to BDNF and other synaptogenic signals is available in the Dihexa vs BDNF article, which clarifies the famous "10 million times more potent" claim and what it actually measures.

The Fosgonimeton Clinical Context: What It Does and Does Not Tell Us

Dihexa has a close clinical cousin that has been tested in human trials: fosgonimeton (ATH-1017), developed by Athira Pharma. Fosgonimeton is a small-molecule HGF enhancer from the same pharmacological lineage as Dihexa. Although the two molecules are not identical, the clinical fosgonimeton programme is the closest thing to human data for Dihexa's mechanism.

The crucial point for this article: fosgonimeton was developed for Alzheimer's disease, not for TBI, concussion, or stroke. The Phase 2 ACT-AD trial and the Phase 3 LIFT-AD trial enrolled patients with mild-to-moderate Alzheimer's disease. Topline LIFT-AD results, reported in 2024, did not meet the primary cognitive endpoint. Secondary analyses in patient subgroups produced signals that the sponsor interpreted positively, but regulators and independent commentators have not treated the programme as a clear success.

For Dihexa-and-brain-injury claims, this clinical context cuts two ways:

• Against: The closest clinical test of this mechanism (in a different but not unrelated neurological indication) did not produce decisive benefit. Cognitive symptoms in late Alzheimer's are driven by protein pathology and massive neuronal loss — distinct from TBI — but the mechanism's inability to robustly rescue cognition in that population is a relevant negative data point.
• In favour: Fosgonimeton was not tested in TBI, concussion, or stroke. The cellular environment of post-injury synaptic remodelling differs from the Alzheimer's brain (active astrogliosis, different tau burden, younger patient population, functional preservation of many circuits). A mechanism could theoretically be more useful for injury recovery than for late-stage neurodegeneration — or it could be equally ineffective. The extrapolation is uncertain.

The controversy around Athira's early research integrity — addressed in detail in the Dihexa vs BDNF article — also affects the interpretability of some upstream preclinical data, which is relevant when assessing whether Dihexa "should" work in injury models.

Concussion and Post-Concussion Syndrome: A Specific Biological Case

Concussion is the most common form of TBI and the condition most frequently discussed in online communities considering Dihexa. Unlike severe TBI, most concussions involve no macroscopic brain damage visible on standard imaging. The injury is diffuse and functional rather than focal and structural — and this has specific implications.

The Underlying Pathology

Concussion pathophysiology is best characterised as a temporary "neurometabolic crisis": rapid release of excitatory neurotransmitters, depolarisation of neuronal membranes, massive potassium efflux, calcium influx, mitochondrial dysfunction, and transient energy crisis. Structurally, diffuse axonal stretching can occur in white matter tracts even when imaging is unremarkable. Most of this resolves within days to weeks in most individuals.

In a minority — often estimated at 10–20% — symptoms persist beyond the expected recovery window (typically two to four weeks in adults; longer in adolescents), producing post-concussion syndrome (PCS). Symptoms include headache, cognitive fog, sleep disturbance, mood dysregulation, and vestibular disturbance. The biological basis of persistence is incompletely understood but likely involves chronic neuroinflammation, persistent autonomic dysfunction, and disrupted thalamocortical connectivity.

Where Dihexa's Mechanism Might Fit (And Where It Does Not)

Dihexa's synaptogenic activity could theoretically address the disrupted connectivity component of PCS — if synaptic loss or dysfunction is a driver of chronic symptoms. HGF's anti-inflammatory shift in microglial phenotype is another theoretical match for persistent neuroinflammation.

But Dihexa has no plausible role in the acute phase. The neurometabolic crisis of the first 72 hours is managed by rest, hydration, and avoidance of further impact — not by synaptogenic agents. Introducing a growth factor agonist into an acutely injured brain has unknown risks, including the possibility of aberrant circuit formation.

For concussion specifically, the evidence-based approach is well established: brief symptom-limited cognitive rest, graded return to aerobic activity beginning within days (the older "dark room and complete rest" advice has been superseded by active protocols), and individualised rehabilitation for persistent symptoms. Nothing in the Dihexa literature justifies replacing or delaying these interventions.

Stroke Recovery: A Different Biological Problem

Ischaemic stroke — the most common form — involves blood-flow interruption producing a core of dead tissue surrounded by a penumbra of functionally compromised but potentially salvageable neurons. Haemorrhagic stroke involves direct tissue damage from bleeding. The recovery trajectories differ from TBI in several important ways.

Stroke recovery is strongly driven by reorganisation of surviving networks, not regeneration of lost tissue. The most important intervention window is the first few weeks to months, when neuroplasticity is maximally responsive to rehabilitation input. This is the period in which evidence-based therapies — constraint-induced movement therapy, intensive aphasia therapy, high-repetition task-specific practice, aerobic exercise — produce their gains.

A pro-synaptogenic compound like Dihexa could in principle augment this plasticity window. Animal data for angiotensin-IV-family compounds in ischaemic models supports this possibility in rodents. But:

• No stroke patient has been formally tested on Dihexa.
• Many candidate neuroplasticity enhancers (including fluoxetine in the large FLAME/FOCUS trials) have failed to show functional benefit despite plausible mechanisms.
• HGF/c-Met activation during the subacute phase of stroke has theoretical risks, including amplification of vascular remodelling in ways that might affect stroke recurrence or haemorrhagic transformation — none of which has been studied in humans for Dihexa.

The bottom line for stroke: the mechanism is among the more interesting candidates for adjunctive therapy. The evidence is absent. Stroke patients considering any unlicensed compound should have the conversation with their stroke physician, not with an internet forum.

CTE and the Chronic-Impact Population

Chronic traumatic encephalopathy (CTE) is a neuropathological condition defined by the perivascular accumulation of hyperphosphorylated tau at the depths of cortical sulci. It has been identified at autopsy in individuals with histories of repetitive head impacts — boxers, American footballers, rugby players, military personnel exposed to blast injury, and some domestic-violence victims. The clinical picture associated with suspected CTE includes mood instability, impulsivity, cognitive decline, and motor disturbance, often emerging years after the last impact.

Several points are relevant for Dihexa:

• No in-vivo diagnosis. CTE cannot currently be diagnosed definitively in the living. Emerging PET tracers for tau have limitations and are not clinical standard of care for CTE. Individuals describing themselves as having CTE are usually inferring the diagnosis from their symptom pattern and exposure history.
• Tau pathology is not a Dihexa target. Dihexa's mechanism is synaptogenic and HGF-mediated; it does not directly clear tau, modify tau phosphorylation, or address the tauopathy that defines CTE neuropathologically.
• The symptoms may overlap with treatable conditions. Sleep apnoea, depression, medication side effects, alcohol use, and chronic pain all produce cognitive and mood symptoms that can be confused with CTE. Addressing these is far more likely to help than an unlicensed peptide.

Any marketing or forum claim that Dihexa "treats CTE" should be treated as unsupported. The same applies to post-service military cognitive concerns from blast injury — a population deserving of proper research, not experimental self-dosing.

Theoretical Risks When Used After Brain Injury

The general Dihexa risk profile is covered in depth in the side effects and risks guide. Several concerns take on heightened relevance in the post-injury setting.

Blood-Brain Barrier Permeability

The blood-brain barrier (BBB) is physically disrupted in TBI and stroke, sometimes for days to weeks (even in mild injury models). A compound whose CNS exposure has been studied in healthy-BBB conditions may achieve substantially higher (or lower) brain concentrations when the BBB is leaky. Dihexa has no published human pharmacokinetic study even in the intact-BBB condition, which makes post-injury dosing a complete pharmacokinetic unknown.

Oncogenic Risk Is Unchanged

Sustained c-Met activation is implicated in tumour growth, angiogenesis, and metastasis across multiple cancer types. The theoretical oncogenic concern associated with Dihexa use is not diminished in the post-injury setting. If anything, the repeated use patterns sometimes advocated by those hoping for "faster recovery" increase cumulative pathway activation.

Seizure Risk

Post-traumatic epilepsy and post-stroke seizures are recognised complications. How HGF/c-Met activation interacts with post-injury cortical excitability is not characterised. Anecdotal Dihexa side-effect reports include mood and arousal changes; none specifically describe seizure activity, but the injured brain has a lower seizure threshold by definition and drug effects in this population cannot be extrapolated from healthy users.

Interaction With Standard-of-Care Medications

Stroke survivors are commonly on antiplatelet or anticoagulant therapy, statins, antihypertensives, and sometimes antidepressants. TBI survivors may be on antiepileptics, neuromodulators, or sleep agents. Dihexa has no published interaction data. The safest assumption is that interactions are unknown. The appropriate response to unknown interactions in a medically managed patient is not self-experimentation.

Community Reports and Anecdotes

Self-reports from users who have tried Dihexa during recovery from concussion, TBI, or stroke exist across Reddit, Longecity, and independent forums. Common themes include:

• Gradual improvement in verbal fluency and word-finding over 6–12 weeks.
• Subjective reduction in "mental fog" during sustained tasks.
• Vivid dreams (a near-universal Dihexa report, covered in detail in the Dihexa Review 2026).
• A minority report worsening headaches or heightened irritability.
• A substantial proportion report no perceptible effect.

These reports are not evidence of efficacy in brain injury for several reasons: recovery from most concussions and many mild TBIs proceeds spontaneously over the same timeframe that users attribute improvement to Dihexa; reports are unblinded and self-selected; survivorship bias favours positive reports (people who had bad experiences may not post); and there is no controlled baseline against which to measure change.

Anecdotal data has value as hypothesis-generating signal but cannot substitute for controlled evidence. The community context for Dihexa use is further discussed in the stacking guide and the general cognitive enhancement overview.

Evidence-Based Interventions With Genuine Human Data

The gap between what Dihexa might theoretically offer and what is actually known stands in sharp contrast to interventions with robust human evidence for brain injury recovery. None are exciting. All are effective.

• Graded aerobic exercise: For concussion, sub-symptom-threshold aerobic exercise initiated within days of injury is now recommended over strict rest; the Buffalo Concussion Treadmill Test protocol has strong evidence. For stroke and TBI, aerobic fitness strongly predicts recovery trajectory.
• Task-specific rehabilitation: Constraint-induced movement therapy (stroke), vestibular rehabilitation (concussion with dizziness), cognitive rehabilitation (TBI), and high-intensity speech therapy (aphasia) all have robust randomised evidence.
• Sleep optimisation: Sleep disturbance is nearly universal after brain injury and independently slows recovery. Treating sleep apnoea, restoring circadian regularity, and evidence-based insomnia interventions (CBT-I) all improve outcomes.
• Mood treatment: Depression is common after TBI and stroke and predicts worse functional outcome. Treating it — pharmacologically or with psychotherapy — produces measurable gains.
• Nutrition: Adequate protein, omega-3 fatty acids (DHA/EPA), and micronutrient sufficiency have supportive human data. Creatine monohydrate has small but consistent trial evidence in TBI cognitive recovery.
• Vascular risk factor management (stroke): Blood pressure control, statin therapy, diabetes management, and smoking cessation produce the largest long-term benefits in stroke populations.

For anyone comparing Dihexa to other nootropic peptides as an adjunct, the Dihexa vs other nootropics guide covers the broader peptide landscape.

The Bottom Line in 2026

Dihexa is a mechanistically interesting compound for brain injury recovery. Its HGF/c-Met activity touches a pathway that is genuinely relevant to the endogenous repair response after TBI, concussion, and stroke. The preclinical evidence for related compounds is real. The synaptogenic mechanism aligns with the biological problem of disrupted connectivity after injury.

It is also an entirely unproven intervention in humans, for any of these indications, in 2026. There are no trials. The closest clinical data come from a related compound (fosgonimeton) tested in a different condition (Alzheimer's) with mixed-to-negative results. The theoretical risks — oncogenicity, pharmacokinetic unknowns in the injured brain, interaction with standard-of-care medications — are not trivial.

For someone recovering from a concussion, stroke, or TBI, the highest-value decisions are the ones supported by strong human evidence: engaging with rehabilitation, managing sleep and mood, rebuilding aerobic fitness, optimising vascular risk factors, and working with a neurologist or rehabilitation physician. These produce the largest, most reliable gains. A research-chemical peptide with no human trials cannot reasonably be recommended to supplement this — not because the mechanism is implausible, but because the knowledge base required to use it responsibly simply does not exist.

Anyone who proceeds anyway should understand what they are doing: they are self-experimenting with an unproven compound on an injured brain, without clinical data to guide timing, dose, or interaction safety. For a detailed overview of the compound itself and its broader evidence base, see What Is Dihexa?, the research summary, and the dosage guide. For the overall safety profile, the side effects page is required reading.

Frequently Asked Questions