C-REGS2—A multinational,high-quality comparative effectiveness study of Cerebrolysin in moderate acute ischemic stroke

Abstract

Background:

The main objective of the Cerebrolysin REGistry Study in Stroke 2 (C-REGS2) was to systematically record the routine clinical use of Cerebrolysin in patients with moderate ischemic stroke (IS) following the principles of a prospective controlled effectiveness study (CES) to compare its effectiveness in terms of functional recovery to patients treated with standard therapy alone.

Methods:

C-REGS2 used an open-label, prospective controlled comparative effectiveness design aligning with the Target Trial Emulation Framework (TTEF) and the GRACE principles for high-quality observational studies based on the principles of high-quality comparative effectiveness research (HQCER) to capture treatment effects in clinical practice. The study was conducted in 16 countries worldwide between April 2018 and April 2024. Moderate IS was defined as baseline NIH Stroke Scale (NIHSS) score 8–15. Treatment modalities and concomitant medications were according to local standards. The methodology included rigorous pre-specified analysis and tight risk-based centralized monitoring, to ensure minimal enrollment bias, maximize data quality and overall reliability of trial results. The compared patient groups were standardized using a restricted cohort design and non-parametric multilevel stratification following the Good Research for Comparative Effectiveness (GRACE) principles. The primary endpoint was ordinal analysis of the modified Rankin Scale (mRS) at 90 days after stroke onset. Secondary endpoints were the ordinal NIH Stroke Scale (NIHSS) at day 21 and 90 after stroke onset, the ordinal mRS at 21 days after IS, the proportion of patients with excellent recovery (mRS 0–1) as well as the proportion of patients with functional independence (mRS 0–2) at 90 days after stroke onset and the ordinal analysis of Montreal Cognitive Assessment (MoCA) scale at 90 days after IS.

Results:

Out of 1865 enrolled patients, the target population (TP) comprised 1769 patients (1021 Cerebrolysin-treated and 748 controls). The median NIHSS at baseline was 10.0. Median Cerebrolysin dose was 30 ml, median treatment duration was 10 days. Cerebrolysin was superior to standard therapy in the primary endpoint independently to prior thrombolysis (MW 0.6157; confidence interval (CI) 0.5910–0.6404; P < 0.0001) as well as in all secondary endpoints: mRS at day 21 (MW 0.6065, 95% CI 0.5811–0.6319, P < 0.0001), NIHSS at day 21 (MW 0.5792; 95% CI 0.5576–0.6008; P < 0.0001) and NIHSS at day 90 (MW 0.5781; CI 0.5561–0.6002; P < 0.0001). Additional pre-specified secondary endpoints (proportion of patients with excellent recovery and functional independence) showed moderate superiority for Cerebrolysin. The ordinal MoCA showed superiority for Cerebrolysin in the TP (MW 0.5530; CI 0.5282–0.5778; P < 0.0001) with more pronounced effects in the subgroup with cognitive impairments at baseline (Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE) ⩾ 3.3). No differences in safety measures were recorded. The study is notable for its robust data integrity, with valid entries of 90.9% for the primary 90 day mRS assessment with multilevel case mix standardization and an overall dropout rate to the final visit of only 5.7%.

Conclusion:

The results of the C-REGS2 study showed the effectiveness and safety of Cerebrolysin treatment for moderate acute IS in real-world clinical practice.

Graphical abstract

Keywords

high quality comparative effectiveness research acute ischemic stroke Cerebrolysin functional outcome HQCER GRACE

Introduction

Stroke remains a leading cause of disability and mortality worldwide, meaning new treatments and rehabilitation are needed.¹ While randomized controlled trials (RCTs) are considered the gold standard in clinical research, their design is optimized for internal validity, hence including narrowly defined patient populations. Consequently, further research on the effectiveness and safety of therapeutic interventions under routine clinical conditions is needed.

Post-approval phase 4 and non-interventional studies have increased in relevance due to real-world evidence’s growing value, complementing pre-approval studies^2,3 by providing important data on safety and performance across diverse patient populations in routine clinical settings.^4
–6

These data are increasingly used by regulatory agencies, health technology assessment bodies, and payers to facilitate informed decision-making.^7,8 Comparative Effectiveness Research (CER) studies are particularly valuable in therapeutic areas such as stroke, where patient heterogeneity and the complexity of care necessitate evidence from routine clinical contexts.⁹

To ensure the methodological robustness of such studies, several frameworks have been established. The GRACE and Target Trial Emulation Framework (TTEF) principles provide guidance on study design, transparency, and analytical rigor in CER.^10,11 They are used to steer the design of high-quality studies by emphasizing pre-defined protocols, selection of appropriate comparator groups, implementation of bias mitigation strategies, and sensitivity analyses.^12,13 With the principles applied consistently CER can offer reliable data sets that complement and extend those from RCTs.¹⁴

Cerebrolysin, a neuropeptide preparation with multimodal activity, was demonstrated to mimic the activity of endogenous neurotrophic factors. Thereby it promotes cerebroprotection, neuroplasticity and neurogenesis by modulating relevant signal transduction pathways such as PI3K/Akt, Sonic Hedgehog and MAPK/ERK. It improves the blood brain barrier integrity, reduces inflammation and has shown efficacy in improving functional as well as cognitive outcomes in stroke models.^15
–18 The Cerebrolysin and Recovery After Stroke (CARS) trials demonstrated a significantly better improvement in motor function and global outcome in patients receiving Cerebrolysin and motor rehabilitation compared to placebo with a number needed to treat of 7.1 for clinically relevant changes in early neurological recovery.^19,20 In patients suffering from nonfluent aphasia after stroke (ESCAS trial), treatment with Cerebrolysin resulted in significantly improved speech recovery and reduced neurological deficits.²¹ Meta-analyses confirmed the beneficial effects of Cerebrolysin on early neurological recovery and safety of the drug.^22
–24 Several guidelines recommend the use of Cerebrolysin for neurorehabilitation after stroke.^25,26 Following the previous results, we aimed to evaluate the effectiveness of Cerebrolysin treatment in a real-world setting in acute IS patients with moderate stroke severity.

Methods

C-REGS2 was a prospective, non-randomized, open label observational CER study adhering to principles of HQCER²⁷ and the GRACE principles¹⁰ and additionally covering key conceptual components of a target trial¹¹ (Grace Checklist and Target Trial Emulation Framework components see supplemental material Tables S1 and S2). The study combined a structured methodological approach with pre-defined analysis procedures and risk-based centralized monitoring, as well as blinded implementation of the Statistical Analysis Plan and analysis database by an independent central statistical institution.

All enrolled patients signed an Informed Consent Form. IS treatment and use of Cerebrolysin were according to local standards and not amended or influenced by the study protocol. Concomitant medication was not restricted.

To minimize bias and confounder effects, patient groups were standardized using non-parametric multilevel stratification procedures, as recommended by the GRACE guideline.¹⁰ Top-level stratification was based on NIHSS including baseline values from 8–15 formally combined by means of the robust nonparametric Wei-Lachin procedure of stochastic ordering.^28,29 Given the recommended lower inclusion limit of 8,³⁰ the upper limit had technically to be set to 15 to preserve sufficient substrata sample sizes.

Second-level stratification was performed to adjust for age, diabetes, and small vessel disease using Cochran-Mantel-Haenszel pooling²⁸ (Figure 1). In accordance with CER principles, a restricted cohort design was used to further minimize confounding by excluding patients with prior stroke or baseline disability (mRS > 1).²⁷ Furthermore, a direct acyclic graph confirmed that all variables included in the stratification were true confounders (see supplement Figure S1).

Figure 1.

Restricted cohort design with multilevel stratification.

A formal nonparametric sample size calculation was performed although power estimates were indicative only. The calculation was set to allow the detection of “small” group differences with 90% power. The calculated total sample size resulted in 1745 subjects. To compensate for usual ambiguities, it was enhanced by a factor of 1.15 to a total of approximately 2000 subjects.

Outcomes and endpoints

The primary endpoint was the ordinal analysis of the mRS at day 90 after IS. Secondary endpoints were the NIHSS scores at days 21 and 90 and mRS at day 21. Furthermore, mRS scores were analyzed for excellent recovery (mRS 0–1) and functional independence (mRS 0–2). To evaluate cognitive declines the MoCA³¹ at day 90 after IS was used.

Further sensitivity analyses were performed, including worst rank imputation for missing mRS values in the Cerebrolysin group, analysis of non-fatal serious adverse events, MoCA outcome based on the dichotomized IQCODE (⩽3.3 and >3.3),³² thrombolysis and admission to stroke unit. With respect to the primary ordinal effectiveness outcomes, an analysis of covariance (ANCOVA) with adjustment for baseline NIHSS scores was performed as a sensitivity analysis (TP and Intention to Include (ITI) population).

For ordinal and binary data, the Mann–Whitney (MW) statistic was employed.^33
–35 Superiority testing was conducted under the null hypothesis of no difference between treatment groups, with interpretation thresholds ranging from inferiority (MW < 0.5) to superiority (MW > 0.5).

Results are given as P values as well as MW effect size measures with their associated confidence intervals.^33
–36 The MW statistic can also be re-expressed in other common effect size measures^28,33,37 like odds ratio (OR) or Number Needed to Treat. Details are shown in Table S3 in the Supplement.

For dichotomized outcomes ORs were calculated, supplemented by MW effect size for inter-outcome comparisons.³³

Safety analyses evaluated mortality, adverse events, and serious adverse events based on observed cases using ORs, with both confounder-adjusted and sensitivity analyses conducted on the ITI population. Non-fatal serious adverse events were evaluated in addition to serious adverse events to provide insight on the outcome of stroke survivors.

Data quality was maintained through an electronic case report form (eCRF) system with built-in validation and independent, centralized statistical monitoring. In agreement with ICH E6 (R2), EMA/269011/2013,³⁸ the Statistical Analysis Plan was finalized and placed in the ISRCTN repository before database lock and unblinding of the statistician for analysis.

The detailed methodology of C-REGS-2 was previously published.²⁹

Results

In total, 1865 patients were enrolled in 16 countries in Europe, Asia, Africa, and South America from April 2018 until December 2023. About 3.2% of the enrolled patients did not meet the ITI inclusion criteria. From the ITI population 2.0% did not meet the inclusion criteria for the TP. Thus, 94.5% of all enrolled patients were included in the TP and 90.0% of enrolled patients were finally included in the pre-specified analysis. The demographic and baseline characteristics of analyzed patients are shown in Table 1 (boxplots for continuous baseline variables are shown in supplement Figure S2).

Table 1.

Demographic baseline characteristics (TP).

Parameter	Total	Cerebrolysin	Control
Parameter	n = 1769	n = 1021	n = 748
Male sex, n (%)	1080 (61.1%)	630 (61.7%)	450 (60.2%)
Mean age, y (SD)	64.4 (11.3)	63.9 (11.1)	65.2 (11.4)
Median time until hospitalization, h (IQR)^a	5 (8)	5 (10)	6 (8)
Stroke Unit, n (%)	1156 (65.3%)	667 (65.3%)	489 (65.4%)
Thrombolytic treatment (i.v.), n (%)	312 (17.6%)	219 (21.4%)	93 (12.4%)
Prevalence of risk factors, N (%)
Diabetes mellitus	432 (24.4%)	257 (25.2%)	175 (23.4%)
Small Vessel Disease	350 (19.8%)	207 (20.3%)	143 (19.1%)
Previous Diagnosis of TIA	89 (5.0%)	48 (4.7%)	41 (5.5%)
Hyperlipidemia	801 (45.3%)	458 (44.9%)	343 (45.9%)
Congestive Heart Failure	213 (12.0%)	103 (10.1%)	110 (14.7%)
Hypertension	1509 (85.3%)	855 (83.7%)	654 (87.4%)
Atrial Fibrillation	322 (18.2%)	188 (18.61%)	134 (18.28%)
Current Smoker at Stroke Onset	443 (25.0%)	248 (24.3%)	195 (26.1%)
Baseline NIHSS
Mean ± SD	10.3 ± 2.2	10.4 ± 2.2	10.3 ± 2.2
Median (IQR)	10.0 (4.0)	10.0 (3.0)	10.0 (4.0)

Percentages based on ITI population.

Calculated from stroke onset.

NIHSS, National Institutes of Health Stroke Scale; IQR, interquartile range.

A study flow-diagram is available in Figure 2.

Figure 2.

CONSORT (Consolidated Standards of Reporting Trials) flow diagram.

Effectiveness evaluation

The pre-planned analysis of the ordinal mRS as primary outcome using the MW effect size measure showed a statistically significant small to medium superiority of Cerebrolysin treatment compared to the control (MW 0.6157; 95% CI 0.5910–0.6404; P < 0.0001; I² = 0.0000; Figure 3).

Figure 3.

Ordinal analysis of the mRS (all stages, TP) at 90 days after stroke onset (absolute scores, MW effect size, observed cases (OC), TP).

Regarding missing final mRS scores, a worst rank score imputation was performed for the Cerebrolysin group as a conservative sensitivity approach. The worst rank score results well supported the primary results (P < 0.0001 in both scenarios, data not shown), excluding bias due to premature discontinuation.

Ordinal mRS was also analyzed for early recovery after stroke (day 21), likewise showing a small to medium superiority for the Cerebrolysin group (MW 0.6065; 95% CI 0.5811–0.6319; P < 0,0001; I² = 0.1696) (see Supplement Figure S3).

The evaluation of the multilevel-adjusted proportion of patients with excellent recovery on day 90 (mRS 0–1, TP, OC), showed an MW of 0.66 (95% CI 0.62–0.70; P_Wei-Lachin < 0.0001; I² = 0.0000), with derived OR³³ of 2.74 (95% CI 2.12–3.60) in favor of Cerebrolysin. Similar results were found for the proportion of patients with functional independence (mRS 0–2) with an MW of 0.67 (95% CI 0.63–0.71; P_Wei-Lachin < 0.0001; I² = 0.0000), and an derived OR³³ of 2.88 (95% CI 2.28–3.68) (see Supplement Figures S4 and S5).

The analysis of the ordinal NIHSS showed a small superiority of Cerebrolysin compared to the control group (MW 0.5781; 95% CI 0.5561–0.6002; P < 0.0001; I² = 0.3134; Figure 4) on day 90 and a similar result on day 21 (MW 0.5792; 95% CI 0.5576–0.6008; P < 0.0001; I² = 0.3210; see Supplement Figure S6), again indicating clinical improvement of Cerebrolysin-treated patients already in the early recovery period.

Figure 4.

Ordinal NIH Stroke Scale (NIHSS) at 90 days after stroke onset (absolute scores, MW, OC, TP).

The analysis of the MoCA scale at day 90 post-stroke demonstrated a small, yet statistically significant superiority of Cerebrolysin compared to the control group (MW 0.5530; 95% CI 0.5282–0.5778; P < 0.0001; I²= 0.0000; Figure 5). To account for pre-stroke cognitive status, the MoCA data was dichotomized as previously described (IQCODE ⩽ 3.3, no evidence of pre-stroke cognitive impairment, IQCODE > 3.3 suggestive of pre-existing cognitive decline).³⁹ MoCA scores at day 90 were generally higher in patients with IQCODE ⩽ 3.3 (MW 0.5282; 95% CI 0.5040–0.5524; P = 0.0222; I²= 0,0000) compared to those with IQCODE > 3.3 (MW 0.5803; 95% CI 0.5306–0.6300; P = 0.0015; I²= 0.0000; Figure 6). Notably, the effect of Cerebrolysin on cognitive outcome was particularily pronounced in the subgroup with evidence of pre-stroke cognitive deficits (IQCODE > 3.3), indicating that these patients experienced the greatest treatment benefit in terms of cognitive outcome.

Figure 5.

Ordinal MoCA, 90 days after stroke onset (MW, OC, TP).

Figure 6.

Scatter plots of MoCA scores at 90 days post-stroke, stratified by baseline IQCODE (⩽ 3,3 vs > 3,3).

A sensitivity analysis compared the mRS at 90 days after stroke onset in patients with and without thrombolysis (Figure 7 Panel A and B). In the TP, 17.6% of the patients received thrombolysis. In the non-thrombolysis group (82.4%) the treatment effect of Cerebrolysin was comparable to the primary outcome (MW 0.6226; 95% CI 0.5943–0.6505; P < 0.0001; I² = 0.0000; Figure 7, Panel A).

Figure 7.

Ordinal analysis of the mRS at 90 days after stroke onset compared between patients without (Panel A) and with (Panel B) IV thrombolysis (absolute scores, MW, OC, TP).

Of note, in the group of patients that received thrombolysis, Cerebrolysin treatment still resulted in a statistically significant superiority in the pre-specified robust Wei-Lachin procedure, despite the substantially smaller sample size (MW 0.5723; 95% CI 0.5091–0.6354; P_Wei-Lachin = 0.0248; I² = 0.3537; Figure 7, Panel B).

Analysis of patients receiving stroke unit care or not, showed a similar result. Patients not treated in a stroke unit (34.70%) showed again a small to medium superiority for the Cerebrolysin group (MW 0.6673; 95% CI 0.6159–0.7188; P < 0.0001; I² = 0.0000). Patients receiving stroke unit care (65.30%) still showed a small superiority in the Cerebrolysin group (MW 0.5999; 95% CI 0.5696–0.6301; P < 0.0001; I² = 0.0000).

Safety analysis

Deaths, Serious Adverse Events, and non-fatal Serious Adverse Events showed no significant differences between groups (all P ⩾ 0.1). All analyses were performed in the ITI population, using a fixed-effects model with adjustment for stroke severity.

Discussion

The C-REGS2 study showed that the Cerebrolysin treatment had significant benefit on functional outcome at 90 days after IS compared to standard treatment alone, consistent with findings from controlled clinical trials.^19
–21 Based on the effect size and their benchmark values for relevance of difference³⁷ the effect size of MW 0.6157 corresponds to an OR of approximately 2.03 and a number needed to treat of 8,64 suggesting a moderate treatment effect.⁴⁰

Secondary endpoints, including NIHSS at days 21 and 90, mRS at day 21, and categorical functional outcomes (e.g. independence), yielded results consistent with the primary analysis. Notably, early recovery (day 21) benefits, which were in line with prior studies^22,41 indicate that treatment with Cerebrolysin can increase the pace of recovery mechanisms post-stroke indeed.

Given the approval status of Cerebrolysin in about 60 countries, a large, real-world, multinational observational CER study was initiated. While RCTs are considered the gold standard in clinical research, observational studies offer the significant advantage of broader patient selection and enhanced external validity.⁶ Such studies reflect routine clinical practice more accurately, improving the generalizability of outcomes and providing a valuable data set for decision-making by regulatory agencies and health technology assessment bodies.

Nevertheless, the lack of randomization, stringent patient selection, and blinding in observational studies may introduce potential bias.⁴² To mitigate these limitations, C-REGS2 was designed following the GRACE guidelines,^12,43 ensuring rigorous methodological standards. Key confounders, such as small vessel disease, prior stroke, prior disability, diabetes, and age,⁴⁴ were accounted for in a pre-specified case-mix standardization strategy.²⁹ Baseline stroke severity was addressed through appropriate patient stratification to ensure comparable patient cohorts.^30,45,46 To avoid ceiling effects due to spontaneous recovery only patients with baseline NIHSS 8 and above were included.

A risk-based centralized monitoring system was implemented ensuring consistent data quality and close patient follow-up. This approach resulted in a very low dropout rate (5.7%) and a high proportion of patients included in the final analysis (90%), minimizing potential attrition bias. The inclusion of over 1.800 patients across 16 countries worldwide ensures a wide representation of clinical practices, stroke etiologies, and healthcare systems.

Originally planned with a two-stage adaptive procedure according to Bauer-Köhne,²⁹ the scientific advisory board recommended a seamless transition due to delayed recruitment caused by the COVID19 pandemic. However, a sensitivity analysis was conducted using the originally specified Bauer-Köhne procedure, showing that Type I error control was maintained (details see supplement Table S4).

The thrombolysis rate of 17.6% (TP) across participating countries, though below optimal thresholds, exceeded expectations in many low- and middle-income geographies where the study has been performed.⁴⁷ Due to an observed baseline group difference in thrombolysis rates (21.4% vs 12.4%) a sensitivity analysis based on the principle of like-to-like comparison was performed, evaluating patients receiving thrombolysis and patients without thrombolysis as separate strata. Cerebrolysin was shown to be effective in both patient groups, suggesting that it offers a treatment option for patients not receiving thrombolysis (MW 0.6224, OR 2.11, Number Needed to Treat 8.2) but that it can also provide benefit as adjunctive therapy to thrombolysis. These results are in line with the meta-analysis of Bornstein et al.²² and Guekht et al.,²⁰ showing a combined Number Needed to Treat of 7.7 and 7.1 respectively for clinically relevant changes in early NIHSS. For further confirmation of this observation additional randomized studies are warranted. Similar results were obtained for Stroke-Unit and general hospital settings, affirming its role as a viable therapeutic option across diverse acute stroke care settings. Subgroup analyses of the 90 day mRS, stratified by thrombolysis status, provide clear evidence for the consistency of the treatment effect, with a stronger treatment effect in the group not receiving thrombolysis. Due to competitive effectiveness of thrombolysis and Cerebrolysin, the less-pronounced results in the group of patients with thrombolysis are well explainable.

Post-stroke cognitive impairment is a common sequel after stroke, and it can significantly affect stroke patient outcomes and caregivers. Post-stroke cognitive impairment is most prevalent in the first year after IS, with rates of up to 60% in survivors.⁴⁸ Post-stroke cognitive development, as assessed by MoCA at 90 days, likewise showed favorable effects for Cerebrolysin, particularly in patients with suspected pre-stroke cognitive decline (IQCODE > 3.3). This finding aligns well with prior studies demonstrating the cognitive benefits of Cerebrolysin^49,50 in various patient populations. Considering this fact, further research is needed to address these issues.^51
–53 RCTs aiming to study the long-term effects of Cerebrolysin on cognitive outcomes in post-stroke patients are ongoing with results expected in 2026.^21,54

Safety data from C-REGS2 were consistent with prior clinical trial findings,²³ with no significant differences in the incidence of death, serious adverse events, or non-fatal Serious Adverse Events between the Cerebrolysin and control groups, underscoring the intervention’s favorable tolerability profile.

Despite its strengths, C-REGS2 has some limitations as an observational CER study, including the possibility of residual confounding. While the study adhered to GRACE principles and advanced standardization techniques were applied to deal with key confounders, causal inferences remain more limited compared to RCTs. However, the main findings from C-REGS2 were in line with previous RCTs, suggesting overall validity of the results. All sensitivity analyses were in agreement with the results of the primary outcome, emphasizing the robustness of the data set. Furthermore, treatment duration and dosing were not standardized and administered according to local treatment standards, reflecting real-world conditions. In conclusion, these characteristics position C-REGS2 as a valuable complement to the more controlled settings of RCTs and the results further emphasize the added value of Cerebrolysin treatment for acute and sub-acute stroke care.

Supplemental Material

sj-docx-1-wso-10.1177_17474930251375439 – Supplemental material for C-REGS2—A multinational, high-quality comparative effectiveness study of Cerebrolysin in moderate acute ischemic stroke

Supplemental material, sj-docx-1-wso-10.1177_17474930251375439 for C-REGS2—A multinational, high-quality comparative effectiveness study of Cerebrolysin in moderate acute ischemic stroke by Milan R Vosko, Daniel Sanak, Youngrok Do, Jarungchai S Vatanagul, Tamer Roushdy, Natan M Bornstein, Johannes C Vester and Michael Brainin in International Journal of Stroke

Footnotes

Acknowledgements

The authors thank all sites participating in C-REGS2 (See ).

Author contributions

M.V., M.B., N.B., and J.V. conceptualized the study, contributed to methodology design, and drafted the manuscript. All authors reviewed and edited the manuscript and approved the final version. M.V., D.S., Y.D., J.V., and T.R. served as principal investigators, recruited patients, and contributed to data acquisition. J.V. curated the data and performed statistical analyses.

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: M.B., N.M.B., and J.C.V. have received honoraria from EVER Neuro Pharma. M.R.V. has received honoraria from EVER Neuro Pharma and was PI in one of the participating sites. D.S., J.S.V., Y.D., and T.R. were PIs in a participating site and received reimbursement for patient enrollment.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by EVER Neuro Pharma GmbH.

Ethical considerations

The study was submitted to the relevant Ethics Committee (EK Nr: 1026/2017, Ethikkommission des Landes Oberösterreich, Kepler Universitätsklinikum) and to the hospital ECs. As C-REGS2 is a non-interventional study some ECs waived the requirement of approval.

Consent to participate

All participants signed an Informed Consent form, specifically including information about data use and data protection.

ClinicalTrials registration

Registered at ISCRTN: ISRCTN98553245,

ClinicalTrials.gov: NCT03480698,

ORCID iDs

Milan R Vosko

Jarungchai S Vatanagul

Natan M Bornstein

Michael Brainin

Data availability

The data sets generated and/or analyzed during the current study are not publicly available due to ethical and data protection restrictions but are available from the corresponding author on reasonable request and subject to institutional approvals.

Supplemental material

Supplemental material for this article is available online.

References

GBD 2019 Diseases and Injuries Collaborators. Global, regional, and national burden of stroke and its risk factors, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Neurol 2021; 20: 795–820.

U.S. Food & Drug Administration. Real-world evidence, https://www.fda.gov/science-research/science-and-research-special-topics/real-world-evidence (2024, accessed 31 March 2025).

(CHMP/MWP) MWP. Reflection paper on use of real-world data in non-interventional studies to generate real-world evidence for regulatory purposes. Amsterdam: European Medicines Agency, 2025.

Moore

Carter

Nietert

, et al. Recommendations for planning pilot studies in clinical and translational research. Clin Transl Sci 2011; 4: 332–337.

Colli

Fraquelli

Casazza

, et al. The architecture of diagnostic research: from bench to bedside—research guidelines using liver stiffness as an example. Hepatology 2014; 60: 408–418.

Fernainy

Cohen

Murray

, et al. Rethinking the pros and cons of randomized controlled trials and observational studies in the era of big data and advanced methods: a panel discussion. BMC Proc 2024; 18: 16.

Sherman

Anderson

Dal Pan

, et al. Real-world evidence—what is it and what can it tell us? N Engl J Med 2016; 375: 2293–2297.

Barnish

Turner

The value of pragmatic and observational studies in health care and public health. Pragmat Obs Res 2017; 8: 49–55.

Williamson

Altman

Bagley

, et al. The COMET Handbook: version 1.0. Trials 2017; 18: 280.

10.

Dreyer

Schneeweiss

McNeil

, et al. GRACE principles: recognizing high-quality observational studies of comparative effectiveness. Am J Manag Care 2010; 16: 467–471.

11.

Hernán

Robins

JM.

Using big data to emulate a target trial when a randomized trial is not available. Am J Epidemiol 2016; 183: 758–764.

12.

Dreyer

Velentgas

Westrich

, et al. The GRACE checklist for rating the quality of observational studies of comparative effectiveness: a tale of hope and caution. J Manag Care Spec Pharm 2014; 20: 301–308.

13.

Acha

Barefoot

Juarez Garcia

, et al. Principles for good practice in the conduct of non-interventional studies: the view of industry researchers. Ther Innov Regul Sci 2023; 57: 1199–1208.

14.

Dreyer

Tunis

Berger

, et al. Why observational studies should be among the tools used in comparative effectiveness research. Health Aff 2010; 29: 1818–1825.

15.

Zhang

Chopp

Cui

, et al. Cerebrolysin enhances neurogenesis in the ischemic brain and improves functional outcome after stroke. J Neurosci Res 2010; 88: 3275–3281.

16.

Chen

Tung

, et al. Trophic factors counteract elevated FGF-2-induced inhibition of adult neurogenesis. Neurobiol Aging 2007; 28: 1148–1162.

17.

Teng

Zhang

, et al. Therapeutic effect of Cerebrolysin on reducing impaired cerebral endothelial cell permeability. Neuroreport 2021; 32: 359–366.

18.

Zhang

Chopp

Meier

, et al. Sonic hedgehog signaling pathway mediates Cerebrolysin-improved neurological function after stroke. Stroke 2013; 44: 1965–1972.

19.

Muresanu

Heiss

Hoemberg

, et al. Cerebrolysin and Recovery After Stroke (CARS): a randomized, placebo-controlled, double-blind, multicenter trial. Stroke 2016; 47: 151–159.

20.

Guekht

Vester

Heiss

, et al. Safety and efficacy of Cerebrolysin in motor function recovery after stroke: a meta-analysis of the CARS trials. Neurol Sci 2017; 38: 1761–1769.

21.

Homberg

Jianu

Stan

, et al. Speech therapy combined with Cerebrolysin in enhancing nonfluent aphasia recovery after acute ischemic stroke: ESCAS randomized pilot study. Stroke 2025;56: 937–947.

22.

Bornstein

Guekht

Vester

, et al. Safety and efficacy of Cerebrolysin in early post-stroke recovery: a meta-analysis of nine randomized clinical trials. Neurol Sci 2017; 39: 629–640.

23.

Strilciuc

Vécsei

Boering

, et al. Safety of Cerebrolysin for neurorecovery after acute ischemic stroke: a systematic review and meta-analysis of twelve randomized-controlled trials. Pharmaceuticals 2021; 14: 1297.

24.

Brainin

Cerebrolysin: a multi-target drug for recovery after stroke. Expert Rev Neurother 2018; 18: 681–687.

25.

Beghi

Binder

Birle

, et al. European Academy of Neurology and European Federation of Neurorehabilitation Societies guideline on pharmacological support in early motor rehabilitation after acute ischaemic stroke. Eur J Neurol 2021; 28: 2831–2845.

26.

Neurorehabilitation DGf. S3-Leitlinie: Rehabilitative Therapie bei Armparese nach Schlaganfall. Report no. AWMF-Registernummer 080/001, 2021. Berlin: AWMF.

27.

Williams

Skinner

James

, et al. Comparative effectiveness research for the clinician researcher: a framework for making a methodological design choice. Trials 2016; 17: 406.

28.

Colditz

Miller

Mosteller

Measuring gain in the evaluation of medical technology. The probability of a better outcome. Int J Technol Assess Health Care 1988; 4: 637–642.

29.

Vester

Bornstein

Heiss

, et al. C-REGS 2—design and methodology of a high-quality comparative effectiveness observational trial. J Med Life 2021; 14: 700–709.

30.

DeGraba

Hallenbeck

Pettigrew

, et al. Progression in acute stroke: value of the initial NIH stroke scale score on patient stratification in future trials. Stroke 1999; 30: 1208–1212.

31.

Nasreddine

Phillips

Bédirian

, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc 2005; 53: 695–699.

32.

Quinn

Fearon

Noel-Storr

, et al. Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE) for the diagnosis of dementia within community dwelling populations. Cochrane Database Syst Rev 2014; 4: CD010079.

33.

Rahlfs

Zimmermann

Lees

KR.

Effect size measures and their relationships in stroke studies. Stroke 2014; 45: 627–633.

34.

D’Agostino

Campbell

Greenhouse

The Mann–Whitney statistic: continuous use and discovery. Stat Med 2006; 25: 541–542.

35.

Kieser

Friede

Gondan

Assessment of statistical significance and clinical relevance. Stat Med 2013; 32: 1707–1719.

36.

Munzel

Hauschke

A nonparametric test for proving noninferiority in clinical trials with ordered categorical data. Pharm Stat 2003; 2: 31–37.

37.

Rahlfs

Zimmermann

Effect size measures and their benchmark values for quantifying benefit or risk of medicinal products. Biom J 2019; 61: 973–982.

38.

ICH. Statistical principles for clinical trials. ICH topic E9. Report no. CPMP/ICH/363/96, 1998. London: European Medicines Agency (EMA).

39.

Hénon

Pasquier

Durieu

, et al. Preexisting dementia in stroke patients. Baseline frequency, associated factors, and outcome. Stroke 1997; 28: 2429–2436.

40.

Davis

Vanenkevort

Understanding the number needed to treat. J Pain Symptom Manage 2024; 68: e190–e193.

41.

Lang

Stadler

Poljakovic

, et al. A prospective, randomized, placebo-controlled, double-blind trial about safety and efficacy of combined treatment with alteplase (rt-PA) and Cerebrolysin in acute ischaemic hemispheric stroke. Int J Stroke 2013; 8: 95–104.

42.

Anglemyer

Horvath

Bero

Healthcare outcomes assessed with observational study designs compared with those assessed in randomized trials. Cochrane Database Syst Rev 2014; 2014: MR000034.

43.

Bryant

Mendelsohn

Viswanathan

, et al. What impacts the quality of comparative-effectiveness research: a classification and regression tree analysis using the GRACE checklist. Value Health 2016; 19: A277.

44.

Johnston

Connors

Jr Wagner

, et al. A predictive risk model for outcomes of ischemic stroke. Stroke 2000; 31: 448–455.

45.

Wouters

Nysten

Thijs

, et al. Prediction of outcome in patients with acute ischemic stroke based on initial severity and improvement in the first 24 h. Front Neurol 2018; 9: 308.

46.

Heiss

Brainin

Bornstein

, et al. Cerebrolysin in patients with acute ischemic stroke in Asia: results of a double-blind, placebo-controlled randomized trial. Stroke 2012; 43: 630–636.

47.

Khatib

Arevalo

Berendsen

, et al. Presentation, evaluation, management, and outcomes of acute stroke in low- and middle-income countries: a systematic review and meta-analysis. Neuroepidemiology 2018; 51: 104–112.

48.

Ihle-Hansen

Thommessen

Wyller

, et al. Incidence and subtypes of MCI and dementia 1 year after first-ever stroke in patients without pre-existing cognitive impairment. Dement Geriatr Cogn Disord 2011; 32: 401–407.

49.

Al-Kuraishy

Al-Gareeb

Zekry

, et al. The possible role of cerebrolysin in the management of vascular dementia: leveraging concepts. Neuroscience 2025; 568: 202–208.

50.

Guekht

Moessler

Novak

, et al. Cerebrolysin in vascular dementia: improvement of clinical outcome in a randomized, double-blind, placebo-controlled multicenter trial. J Stroke Cerebrovasc Dis 2011; 20: 310–318.

51.

Cramer

Richards

Bernhardt

, et al. Cognitive deficits after stroke. Stroke 2023; 54: 5–9.

52.

Rost

Brodtmann

Pase

, et al. Post-stroke cognitive impairment and dementia. Circ Res 2022; 130: 1252–1271.

53.

El Husseini

Katzan

Rost

, et al. Cognitive impairment after ischemic and hemorrhagic stroke: a scientific statement from the American Heart Association/American Stroke Association. Stroke 2023; 54: e272–e291.

54.

Rosu OV. Can treatment with Cerebrolysin improve recovery after acute ischemic stroke? https://www.isrctn.com/ISRCTN88122184

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.79 MB