Cerebrolysin
porcine brain-derived peptide preparation, brain-derived peptide complex, cerebroprotein hydrolysate (related preparations)
Cerebrolysin is not a peptide but a peptide preparation: an enzymatic hydrolysate of purified porcine brain protein, containing a mixture of low-molecular-weight peptides and free amino acids. It is a licensed medicine in Austria, Russia, China, South Korea and dozens of other countries, and has the largest human trial database of any compound in this class by a very wide margin. It is also the one where large, well-run trials have repeatedly failed to confirm the effects that smaller studies suggested.
Mechanism
Because Cerebrolysin is a mixture rather than a molecule, its mechanism cannot be described the way a single agent's can, and this is a genuine scientific limitation rather than a presentational one. Batch composition is defined by a manufacturing process rather than by a structure, no single active constituent has been identified, and no receptor or target has been established. The manufacturer's position, supported by a substantial preclinical literature, is that the preparation exerts pleiotropic neurotrophic activity resembling that of endogenous neurotrophic factors. The phrase most often used is that it mimics the action of neurotrophic factors such as BDNF, GDNF, NGF and CNTF, without containing those proteins, which are far too large to survive the hydrolysis.
Preclinical work reports several convergent effects: reduced excitotoxicity and free-radical damage after ischaemia, inhibition of calpain activation, reduced apoptosis, promotion of neurogenesis and dendritic arborisation, and modulation of amyloid precursor protein processing. In transgenic Alzheimer models it has been reported to ameliorate cerebrovascular amyloidosis. The practical difficulty is that all of these findings are attributed to an undefined mixture, so they cannot be linked to a structure, dose-normalised across batches, or tested by the usual pharmacological methods of selective antagonism. Any mechanistic account of Cerebrolysin is therefore descriptive rather than causal.
What the research shows
Cerebrolysin has been studied in tens of thousands of patients, and the pattern is instructive. In acute ischaemic stroke, the CASTA trial randomised 1,070 patients in Asia (529 to Cerebrolysin and 541 to placebo, given as a daily 30 mL intravenous infusion for 10 days alongside aspirin) and found no significant difference on the confirmatory global endpoint; a favourable trend in patients with NIHSS greater than 12 emerged only in post hoc analysis, which is hypothesis-generating and nothing more. The current Cochrane review of Cerebrolysin for acute ischaemic stroke, updated in 2023 and covering seven RCTs in 1,773 participants, concluded from moderate-certainty evidence that Cerebrolysin or similar peptide mixtures probably have no beneficial effect on preventing all-cause death (RR 0.96, 95% CI 0.65 to 1.41; 6 trials, 1,689 participants) and probably no effect on the total number of people with serious adverse events, while also indicating a potential increase in non-fatal serious adverse events (RR 2.39, 95% CI 1.10 to 5.23; 3 trials, 1,335 participants), more prominent at the cumulative dose of 300 mL. That is about as clear a negative signal, coupled with a possible harm signal, as this literature produces. A pilot trial in aneurysmal subarachnoid haemorrhage in 50 patients was likewise neutral, its authors concluding that Cerebrolysin does not improve six-month global functional performance.
The picture in recovery and rehabilitation is more favourable but much less secure. The CARS trial, a randomised, double-blind, placebo-controlled multicentre study in early rehabilitation after stroke, reported a large effect on the Action Research Arm Test at day 90 and a smaller effect on global status across twelve scales. Its own authors described the study as exploratory with a relatively small sample size and called for confirmation in a large-scale trial; no such confirmatory trial has been published. In vascular dementia, the current Cochrane review (updated 2019, 6 trials, 597 participants) found beneficial effects on cognition and global function but rated the evidence very low quality, noted that every trial with disclosed funding was industry-supported, and concluded that if there are benefits the effects may be too small to be clinically meaningful.
In Alzheimer's disease, several placebo-controlled trials from 2001 to 2006 reported benefit, but read individually they are more equivocal than the summary suggests. The 2001 Ruether study (n=149) reported a 3.2-point ADAS-cog advantage and a CGI responder rate of 63.5% versus 41.4%. The 2002 Panisset trial (n=192) found a significant difference on the global CIBIC+ measure at week 12 but did not report a significant cognitive benefit, and had baseline imbalances in age, age of dementia onset and hallucinations. The 2006 Alvarez dose-ranging study (n=279 across four arms) found cognitive improvement only at the lowest 10 mL dose; the 30 mL and 60 mL doses improved global outcome but failed on cognition, which the authors interpreted as a reversed U-shaped dose-response, an interpretation that is also consistent with a fragile signal.
The honest summary is that Cerebrolysin occupies an unusual position: it is not an unstudied grey-market compound, it is a licensed medicine with real trials behind it, and the trials have mostly delivered small, inconsistent or null results. Its widest approvals are in jurisdictions whose regulatory thresholds differ from those of the FDA and the EMA's centralised procedure, and after four decades neither of those authorities has approved it.
Evidence assessment
Mixed evidence
Large, well-conducted human RCTs exist, but the biggest acute stroke trial was neutral, the current Cochrane review finds probably no effect on mortality alongside a statistically significant increase in non-fatal serious adverse events, the vascular dementia evidence is rated very low quality, and the positive findings in rehabilitation and dementia come from smaller, industry-supported or unreplicated studies.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
Cerebrolysin in patients with acute ischemic stroke in Asia: results of a double-blind, placebo-controlled randomized trial (CASTA) Mixed evidence
The confirmatory endpoint showed no significant difference between treatment groups. A favourable trend in patients with NIHSS greater than 12 appeared only in post hoc analysis, and the authors themselves called for it to be confirmed in a further trial.
Cerebrolysin and Recovery After Stroke (CARS): A Randomized, Placebo-Controlled, Double-Blind, Multicenter Trial Mixed evidence
A large superiority over placebo on the Action Research Arm Test at day 90 (Mann-Whitney estimator 0.71, 95% CI 0.63-0.79) and a small-to-medium superiority on global status across twelve scales. The authors describe the study as exploratory with a relatively small sample size and state that the results should be confirmed in a large-scale trial.
Cerebrolysin for acute ischaemic stroke High-quality evidence
Moderate-certainty evidence indicates Cerebrolysin or Cerebrolysin-like peptide mixtures probably have no beneficial effect on preventing all-cause death (RR 0.96, 95% CI 0.65-1.41) and probably no effect on the total number of people with serious adverse events, with a potential increase in non-fatal serious adverse events (RR 2.39, 95% CI 1.10-5.23; 3 trials, 1,335 participants), more prominent at a cumulative dose of 300 mL.
Cerebrolysin for vascular dementia Mixed evidence
Pooled cognition data (3 studies, 420 people) favoured Cerebrolysin (SMD 0.36, 95% CI 0.13-0.58) and global function response favoured it too (2 studies, 379 participants, RR 2.69, 95% CI 1.82-3.98), but both were graded very low quality. The authors conclude the data are not definitive, that risk of bias was high, and that if there are benefits the effects may be too small to be clinically meaningful.
A 24-week, double-blind, placebo-controlled study of three dosages of Cerebrolysin in patients with mild to moderate Alzheimer's disease Mixed evidence
At week 24 only the 10 mL dose improved cognition on the ADAS-cog (P=0.038); the 30 mL and 60 mL doses improved global function but failed to improve cognition. The authors interpreted this as a reversed U-shaped dose-response.
A 28-week, double-blind, placebo-controlled study with Cerebrolysin in patients with mild to moderate Alzheimer's disease Mixed evidence
At week 16 the CGI responder rate was 63.5% with Cerebrolysin versus 41.4% with placebo (P<0.004), with a 3.2-point ADAS-cog difference in favour of treatment (P<0.0001); improvements were largely maintained to week 28. Adverse events occurred in 43% versus 38%.
Cerebrolysin in Alzheimer's disease: a randomized, double-blind, placebo-controlled trial with a neurotrophic agent Mixed evidence
A significant difference on the global CIBIC+ at week 12 (P=0.033), with 76% responders on Cerebrolysin versus 57% on placebo (P=0.007). No significant cognitive benefit is reported in the abstract, and there were baseline imbalances between groups in age, age of dementia onset and number of patients with hallucinations.
Safety profile of Cerebrolysin: clinical experience from dementia and stroke trials Mixed evidence
Described a favourable overall safety profile with predominantly mild adverse effects across the trial programme.
Randomized, placebo-controlled, double-blind, pilot trial to investigate safety and efficacy of Cerebrolysin in patients with aneurysmal subarachnoid hemorrhage Mixed evidence
Neutral. No significant difference in favourable six-month GOSE (OR 1.49, 95% CI 0.43-5.17), in modified Rankin outcome, in MOCA neurocognitive performance or in delayed cerebral ischaemia. The authors concluded that Cerebrolysin is safe and feasible but does not improve six-month global functional performance.
Safety
Cerebrolysin has the most substantial safety record in this class, drawn from dementia and stroke trials involving thousands of patients, and a 2012 review of that pooled experience described a favourable profile. Reported adverse effects are typically mild: dizziness, sensations of heat, agitation, headache, weight loss and injection-site reactions. Two qualifications matter. First, the current Cochrane review of acute ischaemic stroke found a statistically significant increase in non-fatal serious adverse events (RR 2.39, 95% CI 1.10 to 5.23), more prominent at the 300 mL cumulative dose. That is not a theoretical concern and cannot be dismissed. Second, and specific to this preparation, it is derived from porcine brain tissue by enzymatic hydrolysis. That origin raises theoretical concerns about immunogenicity and about transmissible agents that are managed by the manufacturer's sourcing and processing controls but that do not apply to synthetic peptides at all. Product obtained outside a regulated supply chain has no such controls, and for a biologically derived, batch-defined preparation that gap is materially more serious than it would be for a synthetic molecule.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | No UK marketing authorisation from the MHRA. It is licensed in a number of European countries (notably Austria, where the manufacturer is based, and in several central and eastern European states) under national authorisations rather than through the EMA's centralised procedure. Supplying it for human use in the UK without a licence falls foul of the Human Medicines Regulations 2012. |
| United States | Not FDA-approved for any indication and never granted a US marketing authorisation. It has been the subject of investigational interest but has not completed a regulatory pathway for any indication in the United States. |
| WADA (sport) | Not named on the WADA Prohibited List. Athletes in jurisdictions where it is unlicensed should note that category S0 covers pharmacological substances without current regulatory approval for human therapeutic use in the relevant jurisdiction. |
Questions
In many countries, yes: it holds national marketing authorisations in Austria, Russia, China, South Korea and dozens of others. It is not approved by the FDA, has no UK marketing authorisation, and has never gone through the EMA's centralised procedure. Approval in some jurisdictions and not others usually reflects differing evidentiary thresholds rather than differing evidence.
The largest trial, CASTA, randomised 1,070 patients and found no significant difference on its confirmatory endpoint. The current Cochrane review (2023, 7 trials, 1,773 participants) concluded from moderate-certainty evidence that it probably has no beneficial effect on preventing death after acute ischaemic stroke, and found a statistically significant increase in non-fatal serious adverse events. Results in the separate setting of post-stroke rehabilitation, notably CARS, are more favourable but the CARS authors themselves called their study exploratory and asked for confirmation that has never come.
Weaker than the headline claims. In vascular dementia the current Cochrane review rates the evidence very low quality and says that if there are benefits they may be too small to matter clinically. In Alzheimer's, the 2002 Panisset trial found a global but not a cognitive effect, and the 2006 dose-ranging study found cognitive benefit only at its lowest of three doses. Neither pattern is what a robust drug effect looks like.
An enzymatic hydrolysate of purified porcine brain protein, a mixture of low-molecular-weight peptides and free amino acids, defined by its manufacturing process rather than by a chemical structure. No single active constituent has been identified and no receptor target has been established.
The theoretical concerns are immunogenicity and transmissible agents, both managed in the licensed product by controlled sourcing and processing. Those controls are the whole safeguard, which means product obtained outside a regulated supply chain carries a materially different risk profile from a synthetic peptide obtained the same way. There is no structure to verify, only a process to trust.