← Back
NeuroTrials.ai
Neurology Clinical Trial Database

REACT

REACT: a randomized trial to assess the efficacy and safety of clazosentan for preventing clinical deterioration due to delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage

Year of Publication: 2025

Authors: Stephan A. Mayer, Nicolas Bruder, Giuseppe Citerio, ..., E. François Aldrich

Journal: Journal of Neurosurgery

Citation: J Neurosurg 142:98–109, 2025

Link: https://doi.org/10.3171/2024.4.JNS232191


Clinical Question

Does clazosentan (15 mg/hour IV), an endothelin receptor antagonist, reduce clinical deterioration due to delayed cerebral ischemia when administered for up to 14 days alongside standard care (including nimodipine) in patients with aneurysmal subarachnoid hemorrhage and thick diffuse clot?

Bottom Line

Clazosentan 15 mg/hour IV for up to 14 days did not significantly reduce clinical deterioration due to DCI (15.8% vs 17.2%, p=0.734) in high-risk aSAH patients, despite significantly reducing the need for rescue therapy by 42.6%. There was no benefit on 12-week functional outcome. The placebo DCI rate (17.2%) was much lower than the expected 28%, likely contributing to the negative result. These findings contrast with positive results in Japanese phase 3 trials that used clazosentan without concomitant nimodipine.

Major Points

  • Phase 3, double-blind RCT of 406 treated patients across 74 ICUs in 15 countries
  • Primary endpoint (clinical deterioration due to DCI up to day 14): 15.8% clazosentan vs 17.2% placebo, RRR 7.2% (95% CI –42.6% to 39.6%, p=0.734)
  • Main secondary endpoint (clinically relevant cerebral infarction at day 16): 7.4% vs 11.3%, RRR 34.1% (95% CI –21.3% to 64.2%, p=0.177)
  • Rescue therapy significantly less needed with clazosentan: 10.4% vs 18.1%, RRR 42.6% (95% CI 5.4–65.2%), confirming antivasospastic activity
  • Sensitivity analysis adding rescue therapy to primary endpoint definition: RRR improved to 26.7% (95% CI –8.5% to 50.5%) but remained nonsignificant
  • Poor 12-week outcome (mRS ≥3 or GOSE ≤4) nonsignificantly worse with clazosentan: 24.8% vs 20.1%, RRI 25.4% (95% CI –10.7% to 76.0%)
  • After adjustment for baseline imbalances (geography, securing procedure), RRI for poor GOSE decreased to 14.0%
  • Placebo DCI rate (17.2%) was much lower than the expected 28% based on CONSCIOUS 2/3 data, reducing statistical power
  • Key difference from positive Japanese phase 3 trials: clazosentan was given alongside nimodipine in REACT vs without systemic vasodilators in Japanese studies
  • Deaths within 24 weeks: 7/207 (3.4%) clazosentan vs 3/199 (1.5%) placebo; 1 death (brain edema) potentially related to clazosentan
  • TEAEs more common with clazosentan included lung complications (25.6% vs 10.6%), edema/fluid retention (19.8% vs 4.0%), hypotension (10.1% vs 4.0%), and pulmonary edema (6.3% vs 1.0%)
  • Higher premature treatment discontinuation with clazosentan (13.0% vs 5.5%)

Design

Study Type: Multicenter, randomized, double-blind, placebo-controlled, parallel-group, phase 3 trial

Randomization: 1

Blinding: Double-blind. Patients, investigators, and site personnel blinded to treatment until study closure. Study drug kits similarly packaged. Independent clinical event committee and independent radiology committee adjudicated endpoints blinded.

Enrollment Period: February 2019 to May 2022

Follow-up Duration: 24 weeks post-aSAH (primary endpoint assessed at day 14; main secondary at day 16; clinical outcome at week 12)

Centers: 74

Countries: Austria, Belgium, Canada, Czechia, Denmark, Finland, France, Germany, Hungary, Israel, Italy, Poland, Spain, Sweden, USA

Sample Size: 406

Analysis: Full analysis set (all randomized patients who initiated study treatment). Cochran-Mantel-Haenszel test stratified by WFNS grade (1–2 vs 3–5) and age (≤60 vs >60 years). Hierarchical testing of primary and secondary endpoints at two-sided alpha 0.05. Sample size calculated for 90% power assuming DCI rate of 28% (placebo) and 14% (clazosentan) using Pearson chi-square test.


Inclusion Criteria

  • Adults aged 18–70 years
  • Angiographically confirmed ruptured saccular aneurysm
  • Aneurysm successfully secured by surgical clipping or endovascular coiling within 72 hours of rupture
  • WFNS grade 1–5 at admission, but required WFNS grade 1–4 after securing procedure and after EVD if required
  • Thick and diffuse clot on admission CT (confluent clot >4 mm in thickness involving ≥3 basal cisterns)
  • Randomization within 96 hours of aSAH symptom onset

Arms

FieldClazosentanControl
InterventionClazosentan 15 mg/hour continuous IV infusion for up to 14 days (minimum 10 days), in addition to standard of care including oral or IV nimodipineMatching placebo continuous IV infusion for up to 14 days, in addition to standard of care including oral or IV nimodipine
DurationUp to 14 days, with clinical follow-up to 24 weeks post-aSAHUp to 14 days, with clinical follow-up to 24 weeks post-aSAH

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Occurrence of clinical deterioration due to DCI up to 14 days after study drug initiation, defined as worsening of ≥2 points on modified GCS or abbreviated NIHSS lasting ≥2 hours not entirely attributable to causes other than CVS; also met by death during 14-day period, unevaluable neurological status with rescue therapy for vasospasm, or rehospitalization for symptomatic vasospasm after early dischargePrimary35/204 (17.2%)32/202 (15.8%)1.32%0.734
Clinically relevant cerebral infarction at day 16 (all-cause infarct ≥5 cm³ or infarct <5 cm³ with DCI)Secondary23/204 (11.3%)15/202 (7.4%)0.177
Rescue therapy required (up to hospital discharge)Secondary37/204 (18.1%)21/202 (10.4%)
Poor outcome at week 12 (mRS ≥3 or GOSE ≤4)Secondary41/204 (20.1%)50/202 (24.8%)0.198
Sensitivity analysis: DCI including any rescue therapy in primary endpoint definitionSecondary47/204 (23.0%)34/202 (16.8%)
Death or vegetative state at end of studySecondary8/199 (4.0%)7/207 (3.4%)
Death within 24 weeks of aSAHSecondary3/199 (1.5%)7/207 (3.4%)
Any TEAEAdverse165/199 (82.9%)182/207 (87.9%)
Treatment-related TEAEAdverse36/199 (18.1%)58/207 (28.0%)
Severe adverse eventsAdverse26/199 (13.1%)33/207 (15.9%)
Premature treatment discontinuationAdverse11/199 (5.5%)27/207 (13.0%)
Lung complicationsAdverse21/199 (10.6%)53/207 (25.6%)
Edema and fluid retentionAdverse8/199 (4.0%)41/207 (19.8%)
HypotensionAdverse8/199 (4.0%)21/207 (10.1%)
Brain edemaAdverse18/199 (9.0%)27/207 (13.0%)
Pulmonary edemaAdverse2/199 (1.0%)13/207 (6.3%)
Pleural effusionAdverse3/199 (1.5%)16/207 (7.7%)
Cerebral vasoconstriction (TEAE)Adverse50/199 (25.1%)33/207 (15.9%)
HypertensionAdverse26/199 (13.1%)10/207 (4.8%)

Subgroup Analysis

The study did not report formal prespecified subgroup analyses on the primary endpoint. A supportive analysis adding any administration of rescue therapy to the DCI definition increased RRR from 7.2% to 26.7% (95% CI –8.5% to 50.5%), still nonsignificant. Post hoc analysis of 12-week outcome adjusted for baseline imbalances in geographical location and aneurysm-securing procedure reduced the RRI for poor GOSE from 25.4% to 14.0% (95% CI –64.6% to 21.1%). An early treatment subgroup (11 patients with moderate-severe vasospasm without neurological deterioration) was discontinued due to low recruitment and included in overall analyses.


Criticisms

  • Placebo DCI rate (17.2%) was far lower than the assumed 28%, substantially reducing statistical power despite adequate sample size
  • Study was powered for 90% at 50% RRR; actual RRR was only 7.2%, suggesting either the drug is ineffective for this endpoint or the endpoint was poorly designed to capture treatment effect
  • Clazosentan was given alongside nimodipine (95–99.5%), unlike positive Japanese phase 3 trials where systemic vasodilators were avoided — possible ceiling effect or pharmacological interaction
  • Primary endpoint definition differed from positive Japanese studies, which included hemodynamic therapy and inotropes as rescue therapy triggers
  • Japanese studies required routine angiography at day 9 for all patients, potentially improving identification of vasospasm-attributable deterioration; REACT relied more on clinical judgment
  • Imbalance in surgical clipping between arms (35.6% vs 24.5%) and geographic distribution, potentially confounding outcome analyses
  • Higher premature treatment discontinuation with clazosentan (13.0% vs 5.5%) may have limited drug exposure
  • Notable adverse event profile: lung complications (25.6% vs 10.6%), edema/fluid retention (19.8% vs 4.0%), and pulmonary edema (6.3% vs 1.0%) without clear efficacy benefit
  • Trend toward more deaths with clazosentan (7 vs 3 within 24 weeks) and worse functional outcome (24.8% vs 20.1% poor), raising safety concerns
  • Nimodipine co-administration varied (95.5% clazosentan vs 99.5% placebo), introducing potential confounding
  • mRS and GOSE may lack sensitivity to detect cognitive improvements relevant to aSAH survivors
  • The early treatment group protocol change (discontinued after only 11 patients) introduces design instability
  • Study funded by Idorsia Pharmaceuticals (clazosentan manufacturer); multiple authors with financial ties to sponsor

Funding

Idorsia Pharmaceuticals Ltd.

Based on: REACT (Journal of Neurosurgery, 2025)

Authors: Stephan A. Mayer, Nicolas Bruder, Giuseppe Citerio, ..., E. François Aldrich

Citation: J Neurosurg 142:98–109, 2025

Content summarized and formatted by NeuroTrials.ai.