Cerebrolysin

Cerebrolysin occupies an unusual position in the peptide world. Most compounds discussed in research peptide communities have animal data, a handful of small human studies, and a great deal of extrapolation. Cerebrolysin has close to the opposite problem: it has been given to many thousands of patients in randomized controlled trials across stroke, traumatic brain injury and dementia, and the scientific community still cannot agree on whether it works.

The product is a parenteral (injectable or infusion) preparation manufactured by EVER Neuro Pharma, an Austrian pharmaceutical company. It is produced from purified porcine (pig) brain protein that has been delipidated and broken down by a standardized enzymatic process. The result is a sterile aqueous solution containing short peptides and free amino acids rather than a defined single molecule. This matters for how the evidence should be read: Cerebrolysin is a biological preparation closer in character to a standardized extract than to a synthetic peptide such as Semax.

Regulatory status varies sharply by jurisdiction. Cerebrolysin holds national marketing authorizations in a number of countries, with the heaviest clinical use historically in Russia, the CIS states, Eastern and Central Europe, China and other parts of Asia. Registered indications in those markets have typically covered ischemic stroke, traumatic brain injury and dementia syndromes, though the exact wording of each license differs from country to country. Readers should verify the current status and approved indications with their own national drug regulator rather than relying on secondary sources, including this one.

In the United States, Cerebrolysin has never been approved by the FDA for any indication. There is no centralized European Medicines Agency marketing authorization either, which is why the product is familiar to clinicians in Bucharest or Moscow and almost unknown to clinicians in Boston or Lyon. That asymmetry is one of the most important things to understand about Cerebrolysin: its clinical reputation is strongly regional, and so is much of the literature that supports it.

This guide is written for people who want to understand what the published human trials and systematic reviews actually report, including the results that did not favor the product. It is educational material only. Cerebrolysin is a hospital-administered prescription medicine in the countries where it is licensed, and nothing in this article is a recommendation to obtain or use it.

Cerebrolysin solution is described by the manufacturer as containing roughly 215.2 mg per millilitre of a low-molecular-weight fraction obtained from porcine brain protein. The peptide component is filtered so that its constituents fall below approximately 10,000 Da, and the preparation also contains a large proportion of free amino acids. Product documentation and review literature generally describe the mixture as roughly one quarter low-molecular-weight peptides and three quarters free amino acids when expressed in terms of nitrogen content, although the precise proportions quoted vary between sources, and batches are standardized by assay rather than by listing individual molecules.

That composition has an important consequence: there is no sequence to quote, no molecular formula, and no single active ingredient that can be isolated and tested on its own. Pharmacological reviews of the preparation, such as the dementia-focused review by Plosker and Gauthier, describe it as a peptidergic mixture whose activity is attributed to the ensemble rather than to one identified fragment.

Several candidate bioactive fragments have been proposed over the years, including peptides reported to show activity resembling that of endogenous neurotrophic factors. Analytical work comparing Cerebrolysin with other brain-derived peptide preparations has also been published, partly because similar products of bovine origin have circulated in some markets. This distinction matters for reading the trial literature: the 2023 Cochrane review of acute ischaemic stroke pooled trials of Cerebrolysin together with a comparable peptide mixture derived from cattle brain, on the grounds that the preparations were closely similar.

Because the product is animal-derived, it carries the regulatory and manufacturing considerations that apply to any biological preparation: batch-to-batch consistency, control of source material, and viral safety assurance throughout the production chain. These are routine in licensed pharmaceutical manufacturing, and they are precisely what cannot be assumed for material sold outside licensed channels.

One practical point follows from all of this. Unlike a synthetic peptide, which can in principle be checked by mass spectrometry against a known sequence, a brain-derived mixture cannot easily be authenticated by a buyer. There is no simple analytical test that a non-specialist could use to confirm that an unlabeled vial contains what a licensed ampoule contains.

The hypothesis behind Cerebrolysin is neurotrophic mimicry. Endogenous neurotrophic factors such as nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF) and glial cell line-derived neurotrophic factor support neuronal survival, dendritic growth and synaptic plasticity. These proteins are too large to cross the blood-brain barrier in useful amounts when given systemically, which has frustrated decades of attempts to use them as drugs. The proposition for Cerebrolysin is that small peptide fragments can reach brain tissue and reproduce part of that signaling.

Preclinical work attributed to the preparation spans several mechanisms: promotion of neuronal survival in culture and in injury models, modulation of the inflammatory response after ischemia, effects on amyloid precursor protein processing and tau phosphorylation in transgenic models, reduced apoptosis, and increased neurogenesis or synaptic density in injured tissue. Pharmacology reviews written by investigators associated with the compound have grouped these under the heading of a dual action: acute neuroprotection followed by longer-term support of neurorecovery and plasticity.

Two caveats deserve emphasis. First, almost all of this mechanistic work is animal or in vitro research, and the step from a rodent lesion model to a human stroke ward has ended a very long list of neuroprotective candidates. Second, because the preparation is a mixture, mechanistic claims cannot be anchored to a specific molecule acting at a specific receptor. The pharmacodynamic story is therefore descriptive rather than mechanistic in the strict sense, which makes it difficult to test and difficult to falsify.

It is also worth noting what the proposed mechanism does not predict. A neurotrophic, plasticity-supporting agent would be expected to help functional recovery in damaged tissue rather than to produce a noticeable acute cognitive effect in a healthy brain. The mechanistic rationale, taken on its own terms, is a rationale for rehabilitation support after injury, not for cognitive enhancement in intact tissue. That gap between the proposed biology and the way the product is sometimes discussed online is one of the recurring themes of this guide.

No mechanism, however plausible, substitutes for clinical outcomes. The next sections look at what the randomized trials found.

Acute ischemic stroke is where Cerebrolysin has been tested most extensively, and where the disagreement is sharpest. The largest single placebo-controlled trial is CASTA, published in Stroke in 2012 by Heiss and colleagues. It enrolled 1,070 patients with acute ischemic hemispheric stroke, randomized within 12 hours of symptom onset, who received either 30 mL of Cerebrolysin daily or saline placebo as an intravenous infusion for 10 days in addition to aspirin. CASTA did not demonstrate a significant advantage for Cerebrolysin on its primary multidimensional endpoint at day 90. Later analyses explored whether more severely affected patients fared differently, but subgroup findings generated after a neutral primary result are hypothesis-generating, not confirmatory.

A second line of evidence comes from the CARS program (Cerebrolysin and Recovery After Stroke), which asked a different question: motor recovery in the subacute phase rather than acute neuroprotection. CARS, reported in Stroke in 2016, randomized 208 patients to Cerebrolysin or placebo for 21 days beginning 24 to 72 hours after stroke onset, alongside standardized rehabilitation, and reported superiority on the Action Research Arm Test at day 90. Supporters read CARS as evidence that the preparation belongs in rehabilitation rather than in the emergency department. Skeptics note that CARS was far smaller than CASTA and was conducted with manufacturer involvement.

The most important counterweight is the Cochrane review. The 2023 update by Ziganshina and colleagues, the seventh version of that review, pooled seven randomized trials with 1,773 participants comparing Cerebrolysin or a closely similar cattle brain-derived peptide mixture against placebo added to standard acute stroke care. Its conclusions were negative on benefit and cautionary on harm: moderate-certainty evidence indicated no beneficial effect on all-cause death, and moderate-certainty evidence indicated a possible increase in non-fatal serious adverse events, with a reported risk ratio of about 2.39 that was more pronounced in the subgroup receiving the higher cumulative exposure. The reviewers also recorded that the drug and the methodological support for most included trials came from the manufacturer, creating a likely conflict of interest.

Other meta-analyses reach friendlier conclusions, and understanding why is instructive. A 2021 safety-focused meta-analysis in Pharmaceuticals by Strilciuc and colleagues pooled 2,202 patients from twelve randomized trials and found no statistically significant difference in serious adverse events between Cerebrolysin and placebo (risk ratio 1.08, 95% confidence interval 0.84 to 1.40). A 2025 systematic review and meta-analysis in Cureus covering 14 randomized trials and 2,884 patients reported a small improvement in NIHSS change favoring Cerebrolysin (mean difference 1.39, 95% confidence interval 0.53 to 2.25), a non-significant trend toward functional independence (modified Rankin Scale 0 to 2; risk ratio 1.31, 95% confidence interval 0.90 to 1.91), and no significant difference in mortality.

The honest summary is therefore this: the largest acute trial was neutral, Cochrane reads the pooled randomized evidence as showing no benefit plus a possible harm signal, and several non-Cochrane meta-analyses with different inclusion criteria and different author affiliations report modest benefit on neurological severity scores without a clear effect on the outcomes that matter most to patients. The regimens described above are what was administered under supervision inside hospital trials. They appear here to describe the evidence, not as guidance for use.

Traumatic brain injury is the second major indication studied. The flagship program was CAPTAIN (Cerebrolysin Asian Pacific Trial in Acute Brain Injury and Neurorecovery), designed as a randomized, double-blind, placebo-controlled multicentre trial in moderate to severe TBI using a multidimensional ensemble of outcome scales rather than a single score. CAPTAIN I was conducted in the Asia-Pacific region and was terminated early because recruitment was too slow, a problem that has affected many acute TBI trials.

Results from the program were reported as a prospective meta-analysis of the CAPTAIN trial series in Neurological Sciences in 2021 by Vester and colleagues. Pooling the two phase IIIb/IV randomized, placebo-controlled trials gave a total of 185 patients for analysis. The authors reported a small-to-medium sized effect on the multidimensional primary endpoint favoring Cerebrolysin, reaching statistical significance at day 30 and at day 90. A result in 185 patients is a signal rather than a settled conclusion, and the early termination of the parent trial limits what can be inferred from it.

Broader syntheses have tried to fill the gap with observational data. A 2023 systematic review and meta-analysis in Brain Sciences by Jarosz and colleagues included ten clinical studies, both retrospective and prospective, covering 8,749 patients. It reported statistically significant changes in Glasgow Coma Scale and Glasgow Outcome Scale scores associated with Cerebrolysin treatment, but no effect on all-cause mortality and no effect on length of stay. The large headline patient number comes mostly from retrospective cohorts, where treatment allocation is not random and where sicker or healthier patients may be selected for treatment in ways that no statistical adjustment fully removes.

Unlike stroke and vascular dementia, there is no Cochrane review dedicated to Cerebrolysin in traumatic brain injury, so the rigorous independent appraisal available in stroke has no direct counterpart here. Readers should treat the TBI evidence base as smaller and more fragile than the stroke evidence base, not as stronger simply because the published conclusions are more favorable.

It is also worth separating severity categories. Studies in mild TBI, in moderate-to-severe TBI, and in combined-therapy settings such as Cerebrolysin alongside repetitive transcranial magnetic stimulation ask different questions in different populations. Pooling them, or citing a mild TBI result as though it applied to severe injury, is a common error in secondary summaries of this literature.

In vascular dementia, the reference appraisal is again a Cochrane review. The original 2013 review by Chen and colleagues identified six randomized controlled trials with 597 participants and reported a beneficial effect on general cognitive function in meta-analysis, with no serious adverse events attributable to treatment. The 2019 update by Cui and colleagues reached a more guarded verdict. It found no new eligible studies since the previous version, graded the evidence for cognitive benefit as very low quality, and concluded that the analyses were limited by heterogeneity and by high risk of bias in the included papers. Critically, the reviewers added that if a benefit exists, the effect may be too small to be clinically meaningful. Their closing observation is quoted often and deserves to be: Cerebrolysin continues to be used and promoted for vascular dementia while the supporting evidence base remains weak.

In Alzheimer's disease, the picture is older and thinner. Several double-blind, placebo-controlled trials were run between the mid-1990s and the mid-2000s, typically using four-week courses of intravenous infusion and measuring cognition alongside a clinician's global impression of change. A 2007 meta-analysis in the Journal of Neural Transmission by Wei and colleagues combined six randomized double-blind placebo-controlled trials and found that a four-week course (30 mL daily on five consecutive days of each week, as used in those trials) produced a statistically significant improvement in clinical global impression, with a log odds ratio of 1.1799 (95% confidence interval 0.7463 to 1.6135). The same analysis concluded that more convincing evidence was needed for effects on cognitive performance and activities of daily living. That qualification is notable: the global impression scale moved, while the cognitive and functional measures did not move convincingly.

A later meta-analysis by Gauthier and colleagues, published in Dementia and Geriatric Cognitive Disorders in 2015, pooled randomized controlled trials in mild-to-moderate Alzheimer's disease and reported results favoring Cerebrolysin on cognitive and global outcome measures. That analysis was conducted with involvement from authors affiliated with the manufacturer and with its biostatistics contractor, which does not invalidate it but is material context when weighing it against independent appraisals.

No Cochrane review of Cerebrolysin in Alzheimer's disease exists to parallel the vascular dementia review, and no trial of the size and rigor now expected of a dementia therapeutic has been completed. Set against the modern standard for Alzheimer's drug development, with its multi-thousand-patient trials, long follow-up and biomarker endpoints, the Cerebrolysin dementia literature is small, mostly short-term and largely of an earlier methodological generation.

For readers comparing options in the cognitive space, our overview of peptides studied for brain and cognitive outcomes places this literature alongside other candidates. None of them is an established treatment for dementia, and cognitive decline that appears in real life requires medical assessment rather than experimentation.

In the licensed-market literature, Cerebrolysin is generally described as well tolerated. Commonly reported non-serious effects in trials and product information include a sensation of heat or flushing, sweating, dizziness, headache, nausea, and local reactions at the infusion or injection site. Rapid infusion has been associated with flushing and dizziness, which is one reason the product is given as a slow infusion in supervised settings. As a porcine protein-derived preparation given parenterally, it also carries a theoretical and occasionally reported risk of hypersensitivity and allergic reactions, including rare severe reactions.

The serious-event picture is where published syntheses diverge, and readers should see both sides. The 2023 Cochrane review reported moderate-certainty evidence of a possible increase in the number of participants with non-fatal serious adverse events among those receiving Cerebrolysin or a similar cattle brain-derived mixture, with a risk ratio of approximately 2.39 and a stronger signal in the higher cumulative exposure subgroup. Earlier versions of the same review reported a comparable direction of effect on smaller datasets. Cochrane's framing is that a harm signal of this size, at moderate certainty, is a reason for caution in a population already at high risk of complications.

Against that, the 2021 safety meta-analysis by Strilciuc and colleagues, pooling 2,202 patients from twelve randomized trials, found no statistically significant difference in serious adverse events (risk ratio 1.08, 95% confidence interval 0.84 to 1.40), and the 2025 meta-analysis of 14 randomized trials likewise reported no significant difference in serious adverse events or mortality. The discrepancy is driven by which trials are included, how events are classified as fatal or non-fatal and serious or non-serious, and which comparisons are pooled. It has not been resolved, and a reader who cites only one side of it is not describing the literature.

Beyond the trial setting there is a separate and more immediate safety question: material obtained outside licensed pharmacy channels. Cerebrolysin is an injectable biological preparation, which means sterility, correct storage, correct labeling and authenticity all matter. Non-sterile injection carries risks of abscess, bloodstream infection and worse, and no laboratory report supplied by a vendor substitutes for a regulated manufacturing chain. Our general discussion of peptide safety considerations covers these risks in more detail.

Medical disclaimer: this section describes adverse effects reported in the scientific literature for educational purposes. It is not a safety assessment for any individual. Cerebrolysin is a prescription medicine where it is licensed, it is not approved in the United States or centrally in the European Union, and anyone considering it in a clinical context should consult a qualified healthcare professional. Do not use this article to self-treat a stroke, a head injury or a memory problem. Those conditions require urgent or specialist medical care.

Readers who compare a Cochrane review with a manufacturer-affiliated meta-analysis of the same product can be forgiven for assuming that one of them must be wrong. Usually neither is fabricating anything. They are answering slightly different questions with slightly different datasets and very different thresholds for calling evidence adequate.

Four factors explain most of the divergence in the Cerebrolysin literature. First, endpoint choice: trials that measure a multidimensional ensemble of scales, or a clinician's global impression, tend to produce more favorable results than trials that measure death, disability on the modified Rankin Scale, or validated cognitive batteries. Second, inclusion criteria: Cochrane restricted its stroke analysis to placebo-controlled randomized trials added to standard care, while broader reviews include open-label and retrospective studies that are far more prone to bias. Third, risk-of-bias grading: Cochrane downgraded the vascular dementia evidence to very low quality on the basis of bias and heterogeneity, whereas other syntheses pooled the same studies without applying that discount.

Fourth, and most uncomfortably, sponsorship. The Cochrane reviewers explicitly noted that the medication and the methodological support for most included stroke trials were provided by the manufacturer, which they described as creating a likely conflict of interest. Several influential meta-analyses and pooled analyses include authors employed by or contracted to EVER Neuro Pharma. Industry funding is normal in drug development and does not by itself make a result false, but the empirical literature on sponsorship effects consistently shows that industry-funded trials report favorable conclusions more often than independently funded ones. When nearly the entire positive evidence base for a product shares a single sponsor, that is a structural feature of the evidence rather than an incidental detail.

What would settle the question is straightforward in principle and expensive in practice: large, independently funded, placebo-controlled randomized trials with pre-registered patient-centered primary endpoints, conducted in populations and care systems outside the product's traditional markets. The Cochrane reviewers have called for exactly this in successive updates. Until such trials exist, the accurate statement about Cerebrolysin is that it has a large but contested evidence base. It is neither proven nor formally disproven.

This is also why readers should be skeptical of any source, promotional or dismissive, that presents the Cerebrolysin literature as settled. The number of published trials is impressive. The convergence of their findings is not.

Cerebrolysin has an unusual profile among compounds discussed in cognitive enhancement communities, and that profile explains its persistence there. It is a real, licensed pharmaceutical product in some countries, with an ampoule, a manufacturer, a package insert and decades of clinical use behind it. That lends it a credibility purely experimental research peptides do not have. At the same time it is not approved in the United States or centrally in Europe, so it sits in a grey zone where online enthusiasm is not constrained by a regulator's labeling.

Three further factors feed the interest. The indications studied, stroke recovery, brain injury and dementia, are read by some readers as proof of a general brain-repair capability, even though recovery after injury and enhancement in a healthy brain are different biological questions. The proposed neurotrophic mechanism is intuitively attractive and maps neatly onto a popular narrative about BDNF and neuroplasticity. And the product's Central and Eastern European clinical pedigree places it alongside other peptides from adjacent research traditions, notably Semax and Selank, which have their own online followings.

What is missing from that picture is any meaningful body of evidence in healthy people. The randomized trials discussed above enrolled patients with acute stroke, traumatic brain injury or diagnosed dementia. There is no comparable randomized literature establishing that the preparation improves memory, focus or productivity in neurologically healthy adults, and the mechanistic rationale, read strictly, does not predict that it would. Extrapolating from a stroke rehabilitation trial to a healthy user is a leap across populations, endpoints and timeframes simultaneously.

There is also the delivery problem. In the trials that generated the evidence, Cerebrolysin was administered parenterally, typically as an intravenous infusion over a course of days in a clinical setting with monitoring. That is not a format that translates safely to unsupervised use, and it is a meaningful practical barrier even for those inclined to ignore the regulatory one.

We do not provide dosing guidance, administration protocols or sourcing information for Cerebrolysin, and this site does not link to vendors for it. The combination of an injectable animal-derived biological, an unresolved serious-adverse-event signal in the most rigorous available review, and no evidence at all in healthy users does not support self-experimentation.

Cerebrolysin is frequently discussed alongside three other compounds with neurological claims. They differ fundamentally in chemistry, evidence base and regulatory position, and conflating them produces a great deal of bad information.

The practical hierarchy of evidence here runs in one direction. Cerebrolysin has by far the largest human trial literature of the four, which is exactly why its negative and neutral results matter so much: a product tested this extensively that still cannot demonstrate a consistent benefit on patient-centered endpoints is telling us something. Semax and Selank sit in a middle position, with clinical use and published studies concentrated in one language and research tradition and limited independent verification. A detailed side-by-side treatment is available in our comparison of Selank and Semax.

Dihexa is at the opposite extreme. It is an interesting molecule in preclinical neuroscience, with reported effects on synapse formation in animal models, and it has no human trial evidence at all. The fact that it circulates in the same online conversations as a licensed hospital medicine is a good illustration of how flattened these discussions can become.

A fair summary across all four: none is an approved treatment for cognitive decline in the United States or the European Union, none has evidence supporting use for enhancement in healthy people, and the one with the most human data is also the one whose most rigorous independent review raised a safety concern. That is not a promising set of options for self-experimentation, however appealing the underlying neuroscience may be.