PDF: CONSCIOUS-3
(2012)Objective
To assess whether clazosentan (5 or 15 mg/h) reduced vasospasm-related morbidity and all-cause mortality in patients with aneurysmal subarachnoid hemorrhage secured by endovascular coiling
Study Summary
• Clazosentan 15 mg/h significantly reduced primary endpoint (15% vs 27% placebo, OR 0.474, p=0.007)
• Clazosentan 5 mg/h showed no significant benefit (24% vs 27% placebo, OR 0.786, p=0.340)
• Neither dose improved functional outcome (GOSE): placebo 24%, 5 mg/h 25%, 15 mg/h 28%
• Marked reduction in rescue therapy with 15 mg/h (7% vs 21% placebo)
Intervention
Intravenous clazosentan 5 mg/h or 15 mg/h or placebo for up to 14 days after aneurysmal SAH
Inclusion Criteria
Age 18-75 years, aSAH from ruptured saccular aneurysm secured by endovascular coiling, any thick clot (short axis ≥4 mm) on baseline CT, WFNS grades I-IV before coiling, able to start study drug within 56 hours of rupture
Study Design
Arms: Clazosentan 5 mg/h vs Clazosentan 15 mg/h vs Placebo
Patients per Arm: Placebo: 189, Clazosentan 5 mg/h: 194, Clazosentan 15 mg/h: 188
Outcome
• Poor outcome (GOSE ≤4) at 12 weeks: Placebo 24%, Claz 5 mg/h 25% (p=0.748), Claz 15 mg/h 28% (p=0.266)
• Rescue therapy: Placebo 21%, Claz 5 mg/h 15%, Claz 15 mg/h 7%
• Mortality at 12 weeks: Placebo 6%, Claz 5 mg/h 4%, Claz 15 mg/h 6%
Bottom Line
CONSCIOUS-3 was halted prematurely after enrolling only 38% of planned patients following negative CONSCIOUS-2 results. Clazosentan 15 mg/h significantly reduced vasospasm-related morbidity/all-cause mortality (15% vs 27% placebo, p=0.007), primarily through reduction in rescue therapy and DIND. However, neither the 5 mg/h nor 15 mg/h dose improved functional outcome at 12 weeks. The lack of functional benefit despite reduction in vasospasm-related events may be due to early trial termination (low statistical power), adverse events (pulmonary complications, hypotension, anemia), increased rescue therapy use in placebo group, or contributions from non-vasospasm mechanisms of injury.
Major Points
- Phase III randomized, double-blind, placebo-controlled trial across 106 centers in 27 countries
- Trial halted prematurely (October 2010) after CONSCIOUS-2 showed no benefit with 5 mg/h dose
- Only 577 of 1500 planned patients (38%) enrolled; 571 treated when stopped
- Clazosentan 15 mg/h significantly reduced primary endpoint: 15% vs 27% placebo (OR 0.474, 95% CI 0.275-0.818, p=0.007)
- Clazosentan 5 mg/h showed no significant benefit: 24% vs 27% placebo (OR 0.786, 95% CI 0.479-1.289, p=0.340)
- Neither dose improved functional outcome (GOSE ≤4): placebo 24%, 5 mg/h 25%, 15 mg/h 28%
- Marked 3-fold reduction in rescue therapy with 15 mg/h: 7% vs 21% placebo
- DIND significantly reduced with 15 mg/h: 10% vs 21% placebo
- Vasospasm-related new cerebral infarcts: placebo 13%, 5 mg/h 16%, 15 mg/h 7%
- Mortality at 12 weeks similar across groups: placebo 6%, 5 mg/h 4%, 15 mg/h 6%
- More adverse events with clazosentan: lung complications (21% placebo vs 36-37% clazosentan), hypotension (7% vs 11-16%), anemia (10% vs 13%)
- Oral nimodipine used in 94-95% of patients across all groups
- Treatment started mean 18-20 hours after aneurysm coiling
- Median total infarct volume at week 6: placebo 14.44 cm³, 5 mg/h 8.26 cm³, 15 mg/h 8.26 cm³
Study Design
- Study Type
- Phase III, prospective, multicenter, international, randomized, double-blind, placebo-controlled trial
- Randomization
- Yes
- Blinding
- Double-blind. Randomization (1:1:1) using independent interactive web-response system with predefined scheme. Randomization code only available to authorized individuals until unblinding after study completion. Randomization stratified by site
- Sample Size
- 577
- Follow-up
- 12 weeks
- Centers
- 106
- Countries
- 27 countries
Primary Outcome
Definition: Composite endpoint of vasospasm-related morbidity and all-cause mortality within 6 weeks of aSAH, defined by at least one of: death; vasospasm-related cerebral infarction (where vasospasm was primary cause or relevant contributing factor); DIND caused by vasospasm (decrease ≥2 modified Glasgow Coma Scale points or increase ≥2 abbreviated NIHSS points lasting ≥2 hours; for non-assessable patients, administration of valid rescue therapy); or neurological signs/symptoms in presence of positive angiogram leading to rescue therapy
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 50/189 (27%) | - | - (5 mg/h: 0.479-1.289; 15 mg/h: 0.275-0.818) | 5 mg/h: 0.340; 15 mg/h: 0.007 |
Limitations & Criticisms
- Trial halted prematurely - only 577 of 1500 planned patients (38%) enrolled
- Stopped after negative CONSCIOUS-2 results and Data Safety Monitoring Board recommendation
- Low statistical power due to early termination limits conclusions
- Despite significant reduction in vasospasm-related morbidity/mortality with 15 mg/h, no improvement in functional outcome (GOSE)
- Clazosentan 5 mg/h showed no benefit, consistent with negative CONSCIOUS-2 results
- Higher adverse event rates with clazosentan: lung complications, hypotension, anemia
- Possible interaction between oral nimodipine and clazosentan contributing to hypotension
- 3-fold more frequent rescue therapy use in placebo group may have obscured benefit
- Tolerability profile may have negated therapeutic benefit
- Clazosentan only targets large vessel vasospasm - effect on microcirculation unknown
- Other mechanisms (microthromboembolism, cortical spreading ischemia, delayed neuronal injury) may contribute to poor outcome
- No assessment of whether clazosentan affects microcirculatory dysfunction
- Confirmatory statistical analyses not performed due to premature termination
- Per-protocol analysis results not fully reported
- Missing data substitution methods may have influenced results
- Only exploratory analyses conducted (not confirmatory)
- Oral nimodipine used in 94-95% of patients - difficult to separate individual drug effects
Citation
Stroke. 2012;43:1463-1469