PDF: CARS
(2016)Objective
To investigate whether Cerebrolysin (30 mL/day IV for 21 days, starting 24–72 hours post-stroke) improves upper extremity motor function compared to placebo at day 90 in ischemic stroke patients undergoing early rehabilitation.
Study Summary
• Mean ARAT score improved from 10.1 to 40.7 with Cerebrolysin vs 10.7 to 26.5 with placebo (mean absolute change +30.7 vs +15.9)
• Global outcome across 12 scales also favored Cerebrolysin (MW 0.62, 95% CI 0.58–0.65, P<0.0001); favorable mRS 0–1: 42.3% vs 14.9%
• Safety comparable to placebo (SAEs 2.9% vs 6.7%); no deaths in Cerebrolysin arm vs 4 deaths (3.8%) in placebo
Intervention
Cerebrolysin 30 mL diluted in 100 mL normal saline IV once daily over 20 minutes for 21 days, starting 24–72 hours after ischemic stroke onset, combined with standardized rehabilitation (5 days/week, 2 hours/day for 21 days)
Inclusion Criteria
Age 18–80; ischemic supratentorial stroke confirmed by CT or MRI; infarct volume >4 cm³; ARAT score <50; Goodglass-Kaplan Communication Scale score >2; prestroke mRS 0–1; no stroke within previous 3 months
Study Design
Arms: Cerebrolysin 30 mL IV daily (n=104) vs Placebo — normal saline 100 mL IV daily (n=104)
Patients per Arm: Cerebrolysin: 104; Placebo: 104 (mITT: 104 vs 101)
Outcome
• mRS 0–1 at day 90: 42.3% vs 14.9%; mRS 0–2: 65.4% vs 33.7%
• Global outcome (Wei–Lachin, 12 scales): MW 0.62 (95% CI 0.58–0.65, P<0.0001)
• SAEs: 2.9% vs 6.7%; Deaths: 0% vs 3.8%
Bottom Line
Cerebrolysin produced a large superiority over placebo on the primary endpoint of upper extremity motor function (ARAT score at day 90: Mann–Whitney estimator 0.71, 95% CI 0.63–0.79, P<0.0001), with mean absolute ARAT change of +30.7 versus +15.9 points. Favorable functional independence (mRS 0–1) was achieved in 42.3% of Cerebrolysin patients versus 14.9% of placebo patients. Safety was comparable to placebo; however, as a phase II exploratory trial with a small sample, findings require confirmation in a phase III trial.
Major Points
- CARS was a phase II randomized, double-blind, placebo-controlled multicenter trial of Cerebrolysin — a porcine-derived multimodal neuropeptide preparation with neuroprotective, neurotrophic, and neuroplastic properties — initiated 24–72 hours after ischemic stroke and paired with intensive standardized rehabilitation (5 days/week, 2 hours/day for 21 days) in 208 patients across Romania, Ukraine, and Poland.
- The primary endpoint (ARAT score at day 90) showed a large Mann–Whitney superiority of Cerebrolysin over placebo (MW 0.71, 95% CI 0.63–0.79, P<0.0001), representing a mean absolute score gain of 30.7 vs 15.9 points; in OC analysis the median ARAT was 51.0 (Cerebrolysin) vs 22.0 (placebo).
- Functional independence (mRS 0–1) was achieved in 42.3% of Cerebrolysin patients vs 14.9% in placebo at day 90 — a clinically striking difference; mRS 0–2 was 65.4% vs 33.7%. Global outcome across 12 scales (Wei–Lachin) also showed small-to-medium superiority (MW 0.62, 95% CI 0.58–0.65, P<0.0001).
- Medium superiority (MW ≥0.64) of Cerebrolysin was observed for 6 of 12 outcome criteria: ARAT, NIHSS, Barthel Index, mRS, SF-36 Physical Component Summary, and Geriatric Depression Scale. Small superiority was seen for gait velocity, 9-Hole Peg Test, aphasia scale, and SF-36 Mental Component Summary. No benefit was seen for neglect (low baseline prevalence).
- Effect size on the ARAT increased continuously from day 7 to day 90, suggesting a cumulative neurotrophic/neuroplastic effect over time rather than an early acute neuroprotective effect — consistent with the known mechanism of Cerebrolysin on neurogenesis and synaptic plasticity.
- Safety profile was comparable to placebo: SAEs in 2.9% (Cerebrolysin) vs 6.7% (placebo); no deaths in the Cerebrolysin arm vs 4 deaths (3.8%) in the placebo arm; no drug-related SAEs in either group; premature discontinuation <5% (3.8% overall).
- The trial enrolled patients with moderate-to-severe stroke (median baseline NIHSS 8–9; median baseline ARAT 0–2), explicitly targeting a population with significant disability where spontaneous recovery would be limited — motivated by a prior CASTA subgroup analysis showing a trend toward Cerebrolysin benefit in NIHSS >12 patients.
- Unlike prior Cerebrolysin trials (10-day treatment courses, acute phase only), CARS used a 21-day infusion course beginning in the subacute phase, specifically designed to harness neuroplastic and neurorestorative effects during the critical window of post-stroke brain reorganization.
- The exploratory phase II design, Eastern European population, unusually low thrombolysis rate (1.9%), and poor placebo recovery rate (only 14.9% achieved mRS 0–1) all limit generalizability and represent important caveats; results require confirmation in a larger phase III trial.
- CARS provided the rationale and effect size estimates for the subsequent CAPTAIN (Cerebrolysin Asian Pacific Trial in Acute Brain Injury and Neurorecovery) phase III program; Cerebrolysin has regulatory approval in multiple countries (Austria, Germany, Russia, China, others) but not in the US or UK, where it remains investigational.
Study Design
- Study Type
- Prospective, randomized, double-blind, placebo-controlled, multicenter, parallel-group, phase II exploratory trial
- Randomization
- Yes
- Blinding
- Double-blind (patients, healthcare providers, data collectors, outcome assessors, and sponsor all blinded; infusion bags provided in sealed colored plastic sleeves to conceal Cerebrolysin's slight yellow tint)
- Sample Size
- 208
- Follow-up
- 90 days from stroke onset
- Centers
- Multiple centers (exact number not specified)
- Countries
- Romania, Ukraine, Poland
Primary Outcome
Definition: Change in Action Research Arm Test (ARAT) score from baseline to day 90. ARAT assesses upper limb motor function; scale range 0 (no function) to 57 (no functional limitation). Analyzed using nonparametric Wilcoxon–Mann–Whitney test with Mann–Whitney (MW) estimator as effect size.
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| - | - | - (0.63–0.79) | <0.0001 |
Limitations & Criticisms
- Phase II exploratory design with a relatively small sample size (n=208) — authors explicitly state results require confirmation in a large-scale phase III trial; limited statistical power for subgroup analyses.
- Exclusively Eastern European population (Romania, Ukraine, Poland) — limits generalizability to broader global stroke populations with different demographics, rehabilitation infrastructure, and standard-of-care practices.
- Both arms received active standardized rehabilitation (5 days/week, 2 hours/day) — benefits may partly reflect an unusually intensive rehabilitation program rather than Cerebrolysin alone; rehabilitation intensity was higher than typical clinical practice.
- Very low rate of thrombolytic treatment (1.9% per arm) — not representative of current acute stroke care, where IV alteplase and/or thrombectomy are standard for eligible patients.
- Poor recovery in placebo group — only 14.9% achieved mRS 0–1 compared to historical controls for moderate stroke, which may artificially inflate the apparent treatment effect of Cerebrolysin.
- No central or blinded adjudication of the primary outcome (ARAT) across sites — multisite variance in ARAT scoring is a recognized limitation in stroke trials.
- Lacunar and subtentorial strokes were excluded — study population enriched for supratentorial cortical/subcortical strokes with upper limb deficit, limiting applicability to all ischemic stroke subtypes.
- High baseline ARAT score of 0 (floor effect) in both groups (median 0 in Cerebrolysin arm) — creates floor effect concerns for measuring improvement, though pre-planned subgroup analysis of ARAT >0 was consistent.
- Funded by EVER Neuro Pharma (manufacturer of Cerebrolysin) — multiple authors are employees or paid consultants; risk of industry bias.
- Cerebrolysin mechanism of action involves low molecular weight porcine brain peptides — composition is not fully characterized, raising questions about regulatory and mechanistic reproducibility; unlike single-molecule drugs.
Citation
Muresanu DF, Heiss WD, Hoemberg V, et al. Cerebrolysin and Recovery After Stroke (CARS): A Randomized, Placebo-Controlled, Double-Blind, Multicenter Trial. Stroke. 2016;47(1):151–159.