# Cerebrolysin: what the evidence actually shows for a brain peptide complex approved in 50 countries but not the US

> Source: https://peptidewellnessusa.com/peptides/cerebrolysin/
> Field guide · Peptide deep-dive
> By PeptideWellness Editorial Team · Medically reviewed by Ksenia Petrushkina, PA-C, MPAS · Updated August 20, 2026 · 14 min read

Cerebrolysin sits in a strange place. It's approved in Austria, Germany, China, Russia, South Korea, and roughly forty more countries for stroke, dementia, and traumatic brain injury [1]. In the United States, it has no approval at all. Its regulatory footprint here is essentially nonexistent. The drug has been studied for forty years. Cochrane has reviewed it multiple times. And the honest answer to "does it work" depends so heavily on which trial you read, who funded it, and how severe the pat

This guide walks through what the evidence shows, what it doesn't, and why a drug used routinely in Vienna hospitals isn't available in Boston ones.

## What is Cerebrolysin, actually

Cerebrolysin is a parenteral drug, meaning injected rather than swallowed, manufactured from purified porcine brain tissue. The starting material is pig brain. The endpoint is a clear solution containing low-molecular-weight peptides (roughly 25% of the active fraction) and free amino acids (the other 75%) [2]. Each milliliter holds 215.2 mg of concentrate, suspended in water with sodium hydroxide as the only excipient [3].

The manufacturer, EVER Neuro Pharma, is based in Unterach, Austria, and has been producing the drug for decades. The peptide fragments are generated by controlled enzymatic breakdown of brain proteins, then standardized batch-to-batch by chromatography and immunoassay. That standardization is the whole game with biologics like this. You can't synthesize Cerebrolysin from a structural formula because it doesn't have one. It's a defined mixture, not a single molecule, and the FDA's substance registry classifies it as "structurally diverse" for exactly that reason [4].

The closest analogy is something like a vaccine adjuvant or a heparin preparation: a biological product whose identity is defined by its manufacturing process rather than its molecular structure. That's also why generic versions don't exist in any meaningful sense. Either you have access to EVER's manufacturing line, or you don't.

## What's actually in it

The peptide fraction is where the proposed activity lives. ELISA assays of Cerebrolysin batches detect fragments with measurable similarity to several endogenous neurotrophic factors, including nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), ciliary neurotrophic factor (CNTF), and glial-derived neurotrophic factor (GDNF).

The peptides are short. None of them are the full neurotrophic proteins themselves; those are too large to survive enzymatic processing or to cross the blood-brain barrier on their own. What's in the vial is fragments that appear to mimic the binding behavior of these factors at their receptors, particularly the TrkB receptor that mediates BDNF signaling.

The amino acid fraction is mostly free amino acids in proportions roughly mirroring brain tissue composition. Whether that fraction contributes meaningfully to clinical effect, or is essentially nutritional padding, is one of the open mechanistic questions.

Standardization matters here in a way that doesn't get enough attention. Two vials from two manufacturing batches need to behave the same way in a patient. EVER's quality-control documentation describes batch-release testing against reference standards, but independent verification of batch-to-batch consistency is limited in the public literature. For a drug that has been on the market for decades, that's a real gap.

## The proposed mechanism

Cerebrolysin is positioned by its manufacturer and by most researchers as a multimodal neurotrophic agent, acting on several pathways at once rather than hitting a single receptor. This is both a strength of the narrative and a problem for cleanly evaluating it.

The best-characterized pathway is TrkB activation. BDNF normally binds the full-length TrkB receptor (gp145TrkB), triggering downstream cascades through MEK/ERK, PI3K/AKT/mTOR, and CAMK [5]. Those cascades do two things: they suppress pro-apoptotic machinery (BAD, BAX, caspase-3) and they upregulate pro-survival proteins like Bcl-2. The peptide fragments in Cerebrolysin appear to activate this same pathway in vitro, at least at the level of phosphorylation events downstream of TrkB.

The second mechanism is what researchers call "endogenous induction." Cerebrolysin appears to stimulate the patient's own neurons, glia, and endothelial cells to produce native neurotrophic factors. This includes catalyzing the conversion of proNGF (which is pro-apoptotic in certain contexts) into mature, neuroprotective NGF. It also shifts the balance between proBDNF/p75NTR signaling (associated with neuronal pruning and apoptosis) toward mature BDNF/TrkB signaling (associated with survival and synaptic strengthening).

Then there's the vascular and structural side. The drug induces vascular endothelial growth factor (VEGF) and angiopoietin-1, both of which support angiogenesis, the formation of new blood vessels in damaged tissue. The Sonic Hedgehog pathway, involved in neurogenesis and myelination, also shows modulation in animal models.

Finally: blood-brain barrier integrity. Tissue plasminogen activator, used to break up stroke-causing clots, has a known side effect of degrading the basement membrane that holds the BBB together. Cerebrolysin appears to attenuate that effect in animal stroke models, which is part of why most stroke trials administer it alongside or after thrombolysis rather than as a standalone.

That's a lot of mechanisms. The honest read of the data: any one of them, in isolation, would be hard to attribute to a peptide mixture rather than to confounders. Taken together, they describe a plausible biological story. Whether that story translates to consistent clinical benefit is the actual question.

## Stroke trials - the most-studied indication

Stroke is where Cerebrolysin has the largest evidence base and the most active controversy.

CASTA, published in 2012, is the trial most often cited. It enrolled 1,070 patients across multiple sites in Asia with acute ischemic stroke, randomizing them to 30 mL of IV Cerebrolysin daily for 10 days versus placebo, starting within 12 hours of stroke onset. The primary endpoint was the modified Rankin Scale at 90 days [6]. CASTA missed its primary endpoint [7]. But subgroup analyses, and this is where it gets contentious, suggested benefit in patients with moderate-to-severe baseline strokes (NIHSS ≥12) [8]. Whether you find that compelling depends on your priors about subgroup analyses in trials that miss their primary outcome.

The Cochrane reviews are less generous. Ziganshina and colleagues have now published seven systematic reviews of Cerebrolysin in acute ischemic stroke, most recently in 2023 [9]. The conclusions have been consistent across updates: evidence for functional benefit is uncertain, the certainty of evidence by GRADE criteria is low, and no clear harm signal has emerged. Low certainty is Cochrane language for "the trials exist, but we don't trust them enough to draw conclusions." Heterogeneity across trials, risk of bias, and large industry involvement in trial design are the main reasons.

A 2021 safety meta-analysis by Strilciuc and colleagues pooled 12 RCTs covering 2,202 patients [10]. It found no statistically real differences between Cerebrolysin and placebo for deaths, serious adverse events, or any-grade adverse events (p>0.05 throughout). There was a numerical trend toward reduced all-cause mortality in the Cerebrolysin arms, which didn't reach significance. The safety picture is reassuring; the efficacy picture remains contested.

The Brainin 2018 expert review made a specific claim worth flagging: effect size appears to scale with stroke severity [11]. Patients with mild strokes recover well regardless of treatment, so there's no room to show benefit. Patients with severe strokes have more deficit to recover, so an active drug has more signal to generate. If that's true, it explains why pooled analyses look weak. They dilute responders with a large bloc of patients who were going to do fine anyway.

A roughly 36-point improvement on the Western Aphasia Battery in one trial's Cerebrolysin arm, compared with a smaller gain on placebo, is the kind of effect size that, if reproducible in larger independent trials, would matter clinically. Reproducibility is the question.

## Alzheimer's and vascular dementia

The dementia evidence is older, smaller, and on balance more positive than the stroke evidence, though with its own quality concerns.

Plosker and Gauthier's 2009 systematic review in Drugs & Aging covered RCTs of up to 28 weeks [12] and concluded Cerebrolysin was superior to placebo for both global outcomes and cognition in mild-to-moderate Alzheimer's and in vascular dementia. The most common adverse event was dizziness, generally mild and transient. Gauthier returned to the topic in a 2015 meta-analysis with similar conclusions [13].

Alvarez and colleagues ran one of the more interesting comparison trials in 2011: a head-to-head of Cerebrolysin versus donepezil [14] versus the combination. Cerebrolysin was as effective as donepezil. The combination was superior to either alone on cognitive endpoints. And, this is the mechanistically interesting part, the combination but not donepezil alone increased measurable BDNF levels in patients with mild-to-moderate AD. That's consistent with the proposed mechanism, though it's a small finding from a single trial.

The vascular dementia evidence is thinner. NCT00947531, a Phase 4 trial of 242 patients, used two 4-week courses of 20 mL IV Cerebrolysin separated by a 2-month gap, with cognitive and global clinical outcomes as primary endpoints [15]. Cui and colleagues' 2019 Cochrane review of vascular dementia trials [16] reached similar conclusions to the stroke review: some signal, low certainty, heterogeneous methodology.

What the dementia evidence doesn't show, and this is worth being direct about: there's no demonstrated effect on disease progression, on biomarker trajectories, or on the underlying neuropathology of Alzheimer's. The effect, to the extent it's real, is symptomatic and modest. Nothing here changes the trajectory of the disease the way an effective disease-modifying therapy would.

## Traumatic brain injury

Jarosz and colleagues published a 2023 systematic review and meta-analysis pooling multiple clinical studies of Cerebrolysin in TBI, totaling thousands of patients. The pooled analysis found statistically real improvements in Glasgow Coma Scale and Glasgow Outcome Scale scores. It found no real effect on mortality or length of stay.

That's a specific kind of finding. Functional recovery improved; survival did not. For a drug being evaluated as neuroprotective rather than as life-saving, that's not necessarily a failure, but it's a narrower claim than the marketing sometimes implies.

CAPTAIN II (Muresanu et al., 2020) was a single-center, double-blind, placebo-controlled RCT of patients with moderate-to-severe TBI. Cerebrolysin produced real reductions in post-injury depression and improvements in memory and concentration on multidimensional outcome measures. A single-center trial is a starting point, not a definitive answer.

There's also an ongoing Phase 3 trial, NCT06052787, evaluating combined Cerebrolysin plus amantadine sulfate in ICU TBI patients. It's enrolling 150 patients with functional and neurological outcomes as primary measures [17]. That trial, when it reads out, will be one of the larger and more methodologically current data points in TBI.

## What the evidence doesn't show

This is the section that matters most for a researcher trying to take an honest position.

The trials, taken together, don't establish Cerebrolysin as a category-defining therapy. They establish a plausible mechanism, a reasonable safety profile, and an inconsistent efficacy signal that's strongest in the most severely affected patients. That's not nothing. It's also not what its advocates sometimes claim.

The methodological problems are real. Most of the positive trials have been conducted in Eastern Europe, Russia, China, and parts of Asia, jurisdictions where regulatory oversight of clinical trial conduct varies and where the manufacturer has had large influence on study design. Many trials were sponsored or co-sponsored by EVER Neuro Pharma. That's not disqualifying; pharmaceutical companies sponsor trials of their own products everywhere. But it does mean independent replication in jurisdictions without manufacturer involvement has been limited.

Heterogeneity across trials is the other issue. Dosing varies (10 mL to 60 mL daily). Treatment duration varies (10 days to 6 months). Timing of initiation relative to the acute event varies. Outcome measures vary. When Cochrane reviewers say "heterogeneity prevents firm conclusions," what they mean is: the trials aren't asking the same question in the same way, so pooling them is statistically awkward even when the individual results look favorable.

What nobody has yet established: long-term outcomes past one year, best dosing schedules, which patient subgroups derive the largest benefit (beyond the severity-scaling observation), and whether the drug adds anything to current standard-of-care thrombolysis and thrombectomy protocols for stroke.

Short answer on quality: the evidence is suggestive, not conclusive, and the gap between those two words is where the entire FDA-versus-rest-of-world divergence lives.

## Regulatory status

Cerebrolysin is approved in Austria, Germany, Russia, China, South Korea, and roughly forty more countries. The list includes most of Eastern Europe, much of Central Asia, and parts of Latin America. EMA classifies it in the "SAFE" category for its approved indications in member states where it's licensed.

In the United States, the drug is not FDA-approved for any indication. It has an FDA Orphan Drug Designation (code 513315) for frontotemporal dementia [18], which provides development incentives for the manufacturer but doesn't constitute approval or permit marketing. Cerebrolysin is registered in the FDA's Global Substance Registration System under UNII 37KZM6S21G and CAS 12656-61-0 [19], classified as a structurally diverse mammalian-origin substance.

It does not appear to be on the FDA's 503A bulks list, nor on the 503B nominated substances list. US compounding pharmacies are not sanctioned to prepare it. That's a clean regulatory line: there's no legal pathway for Cerebrolysin to be dispensed by a US pharmacy as currently constituted.

Why hasn't EVER pursued FDA approval? The honest answer is some mixture of: the cost of running a Phase 3 trial that meets current FDA standards (likely $200M+), the regulatory difficulty of demonstrating efficacy for a structurally undefined biological mixture against modern controls, and the commercial calculation that the existing approved markets generate sufficient revenue without the US burden. None of those are reasons to dismiss the drug. They're reasons it sits where it sits.

For the EMA, the situation is technically different. Cerebrolysin is approved at the national level in several EU member states but has not gone through the centralized EMA approval pathway that would harmonize its status across the bloc.

## Administration and dosing in trials

Cerebrolysin is administered intravenously in virtually all clinical contexts. The standard preparation involves dilution in normal saline and infusion over a period of typically less than an hour.

Across the major stroke trials, the dose is 20-50 mL daily, with 30 mL being the most common single-day dose [20]. Treatment duration runs 10-21 days, started as soon as possible after onset, within 12 hours in CASTA, up to 48 hours in others. The dementia protocols typically run 4 to 6 weeks, sometimes repeated in cycles separated by 2-month gaps. The vascular dementia Phase 4 trial used 20 mL daily, five days per week, for two 4-week courses.

Pediatric use in the autism trials has used intramuscular administration on varying dosing schedules, none of which has been standardized across studies. The pediatric evidence base is small and not strong enough to support broad conclusions.

What's not well-characterized in the public literature: the drug's pharmacokinetics. Half-life isn't cleanly defined because what's being measured isn't a single molecule. The proposed mechanism is downstream induction of endogenous neurotrophic activity rather than direct prolonged peptide presence, which makes traditional PK modeling awkward. That's a genuine gap in the regulatory dossier from a US perspective.

Contraindications are straightforward: hypersensitivity to any component, epilepsy or status epilepticus (the drug is proconvulsant in some animal models), and severe renal impairment. Drug interactions of note: additive effects with antidepressants and MAO inhibitors, with prescribing recommendations to reduce co-medication doses.

## Safety profile

This is the area where the evidence is most consistent. Cerebrolysin's safety record across decades of use in approved markets is solid.

The Strilciuc 2021 meta-analysis of 2,202 patients across 12 RCTs found no statistically real difference between Cerebrolysin and placebo on any safety endpoint. SAE rates were numerically lower in the Cerebrolysin arms in the higher-dose and moderate-to-severe stroke subgroups.

The most common adverse events across trials are mild and transient: dizziness or vertigo (the single most-reported AE), headache, flushing, transient anxiety or agitation, nausea, and injection-site discomfort. In dementia trials, one cohort showed depression in 12.5% of the 10 mL Cerebrolysin arm versus 25% of placebo. Interpret with the usual caution about small-cohort cross-comparisons.

FDA FAERS data, which captures real-world reports rather than trial data, shows the largest signal for chronic kidney disease (736 reports) and "drug ineffective" (577 reports). Causality is not established for FAERS signals. They reflect reports filed by clinicians and patients, not adjudicated adverse events. The CKD signal is consistent with the existing contraindication for severe renal impairment rather than evidence of new harm.

The theoretical safety concern that doesn't show up in trial data but is worth naming: Cerebrolysin is derived from porcine brain tissue. The manufacturer's quality controls test for bacterial, viral, and prion contamination, and no contamination-related adverse event has been reported in the trial literature. But for a US regulatory body evaluating a biological product from porcine CNS tissue against current standards, the theoretical risk is part of the calculus even when the empirical record is clean.

## Where it fits among neurotrophic alternatives

The comparison set is small because the field is small. Synthetic BDNF mimetics, small molecules designed to activate TrkB without being BDNF itself, have been pursued by several drug-discovery programs over two decades. None have reached approval. The mechanism is biologically attractive; the clinical translation has been brutal.

Cerebrolysin's proposition is essentially: rather than try to design a single synthetic molecule that mimics one neurotrophic factor cleanly, use a biologically derived mixture that nudges the system across multiple factors simultaneously. That's either a feature or a bug depending on your regulatory philosophy. The FDA's drug-development paradigm prefers single-molecule clarity. The Cerebrolysin approach is harder to characterize, harder to standardize, and harder to test against the modern trial benchmark, which is part of why it's been more readily approved in regulatory systems that emerged from different traditions.

Among individual peptide alternatives sometimes discussed in research and self-experimentation contexts, Cerebrolysin's closest comparators in proposed mechanism are [Sermorelin](https://peptidewellnessusa.com/peptides/sermorelin/), [Selank](https://peptidewellnessusa.com/peptides/selank/), [Semax](https://peptidewellnessusa.com/peptides/semax/), and [Dihexa](https://peptidewellnessusa.com/peptides/dihexa/). None of those have stroke or TBI approvals anywhere. None have the trial volume Cerebrolysin has. They occupy different regulatory categories with different risk profiles.

The honest framing: Cerebrolysin is the most clinically-studied multimodal neurotrophic agent in the world. That doesn't make it definitively effective. It makes it the most data-rich entry in a category where data is otherwise extremely thin.

## Who's researching it and why

Active clinical research continues primarily in Europe and Asia. The CAPTAIN II investigators, the NCT06052787 team studying combination therapy in ICU TBI, multiple stroke neurorehabilitation groups in Romania and Austria, and ongoing vascular dementia work in China are the main research clusters. Most of this work has some degree of manufacturer involvement, which is normal for proprietary biologics but worth knowing.

Outside formal clinical research, Cerebrolysin shows up in two distinct communities. The first is European and Asian clinical practice. Neurologists in Vienna, Bucharest, Moscow, and Beijing routinely use it as adjunctive therapy for stroke and dementia patients, and have done so for decades. That's medicine as it actually exists, not research.

The second is the self-experimentation community in countries where the drug isn't approved, including a US-based cohort that imports it through gray-market channels. The dosing protocols in those communities often differ substantially from clinical protocols (shorter courses, intramuscular rather than IV administration, sometimes intranasal) and there's essentially no quality control on the supply chain. The combination of a porcine-derived biological, no batch testing, and self-administration is the kind of risk profile that produces real adverse events that don't get captured anywhere. We wouldn't personally use this route without a clinician involved.

For a researcher trying to understand the drug, the most productive frame is probably this: Cerebrolysin is what happens when a plausible biological product, developed before modern regulatory science fully crystallized, ages into a global market split between regulatory traditions that treat it very differently. It's neither vindicated nor debunked by that split. It's a drug with forty years of clinical use, a real but contested evidence base, a clean safety record, and an open question about whether the efficacy signal is strong enough to matter when measured against current standards.

What would change the picture: an independently sponsored, multi-center, methodologically rigorous Phase 3 trial in moderate-to-severe stroke patients, run outside the manufacturer's typical orbit, with pre-specified primary endpoints and modern adjudication. That trial doesn't currently exist. Until it does, the disagreement between Vienna and Washington isn't going to resolve.

## Frequently asked

### What is Cerebrolysin?

Cerebrolysin is a peptide-and-amino-acid mixture made from purified pig brain tissue, given intravenously, and used in 40+ countries for stroke, dementia, and traumatic brain injury. It has no FDA approval in the US.

### Does Cerebrolysin work for stroke or dementia?

The evidence is suggestive but contested. The large CASTA stroke trial missed its primary endpoint (with a subgroup signal in more severe strokes), Cochrane rates the certainty of benefit as low, and the dementia data are modestly positive but largely manufacturer-involved. Effect appears to scale with severity.

### Is Cerebrolysin safe?

Its safety record is the most consistent part of the file — pooled trials show no significant difference from placebo on serious adverse events, and the common effects are mild dizziness, headache, and flushing. It's contraindicated in epilepsy/status epilepticus and severe renal impairment.

### How is Cerebrolysin dosed?

In trials, 20–50 mL IV daily (most commonly 30 mL) for 10–21 days in stroke, or 4–6 week courses in dementia, sometimes repeated. It's always parenteral (IV); there is no oral form.

### Why isn't Cerebrolysin FDA-approved?

It's a structurally undefined biological mixture derived from porcine brain, expensive to run through a modern Phase 3, with most positive trials manufacturer-involved. The US has never approved it (it holds only an orphan designation for frontotemporal dementia), and there's no US compounding pathway.

## References

1. [Cerebrolysin approved in ~40+ countries including Austria, Germany, China, Russia, South Korea. ClinicalTrials.gov](https://clinicaltrials.gov/study/NCT02541227)
2. [Cerebrolysin composition is approximately 25% peptides and 75% free amino acids. MDPI](https://www.mdpi.com/2227-9059/13/7/1661)
3. [Cerebrolysin contains 215.2 mg/mL of concentrate](https://www.everpharma.com/products/cerebrolysin)
4. [FDA substance registry classifies Cerebrolysin as structurally diverse](https://gsrs.ncats.nih.gov/ginas/app/beta/substances/37KZM6S21G)
5. [BDNF binds full-length TrkB receptor (gp145TrkB) triggering MEK/ERK, PI3K/AKT/mTOR, CAMK cascades](https://pmc.ncbi.nlm.nih.gov/articles/PMC5250655)
6. [CASTA trial 2012, n=1070, 30 mL IV daily x10 days, within 12 hours](https://pubmed.ncbi.nlm.nih.gov/22282884/)
7. [CASTA primary endpoint mRS at 90 days, missed](https://pubmed.ncbi.nlm.nih.gov/22282884)
8. [CASTA subgroup benefit NIHSS ≥12](https://www.ahajournals.org/doi/10.1161/strokeaha.111.628537)
9. [Ziganshina Cochrane review Cerebrolysin acute ischemic stroke 2023](https://pubmed.ncbi.nlm.nih.gov/37818733/)
10. [Strilciuc 2021 meta-analysis 12 RCTs, 2202 patients](https://pubmed.ncbi.nlm.nih.gov/34959697/)
11. [Brainin 2018 expert review claims Cerebrolysin effect size scales with stroke severity](https://pubmed.ncbi.nlm.nih.gov/30004268/)
12. [Plosker Gauthier 2009 Drugs & Aging review Cerebrolysin](https://pubmed.ncbi.nlm.nih.gov/19848437/)
13. [Gauthier 2015 meta-analysis Cerebrolysin dementia](https://pubmed.ncbi.nlm.nih.gov/25832905/)
14. [Alvarez 2011 trial Cerebrolysin vs donepezil vs combination](https://pubmed.ncbi.nlm.nih.gov/21679156/)
15. [NCT00947531 Phase 4 vascular dementia, n=242, 20 mL IV. ClinicalTrials.gov](https://clinicaltrials.gov/study/NCT00947531)
16. [Cui 2019 Cochrane review Cerebrolysin vascular dementia](https://pubmed.ncbi.nlm.nih.gov/31710397/)
17. [NCT06052787 Phase 3 Cerebrolysin + amantadine TBI, n=150. ClinicalTrials.gov](https://clinicaltrials.gov/study/NCT06052787)
18. [FDA Orphan Drug Designation code 513315 Cerebrolysin FTD](https://gsrs.ncats.nih.gov/ginas/app/beta/substances/37KZM6S21G)
19. [Cerebrolysin UNII 37KZM6S21G, CAS 12656-61-0. FDA](https://precision.fda.gov/uniisearch/srs/unii/37kzm6s21g)
20. [Cerebrolysin stroke trial doses range 20-50 mL daily, 30 mL most common](https://www.ahajournals.org/doi/10.1161/strokeaha.115.009416)

> Editorial note: Informational only — not medical advice. Decisions about peptide therapy should be made with a licensed healthcare provider familiar with your medical history. See our [methodology](https://peptidewellnessusa.com/methodology/). Last reviewed August 2026.