FASTEST
Recombinant factor VIIa versus placebo for spontaneous intracerebral haemorrhage within 2 h of symptom onset (FASTEST): a multicentre, double-blind, randomised, placebo-controlled, phase 3 trial
Clinical Question
Does recombinant factor VIIa given within 2 hours of ICH onset improve functional outcomes compared to placebo?
Bottom Line
Recombinant factor VIIa 80 µg/kg given within 2 hours of ICH onset significantly slowed haematoma growth but did not improve functional outcomes at 180 days and was associated with a 3-fold increased risk of life-threatening thromboembolic events. The trial was stopped for futility. Subgroup signals in patients with a CT spot sign or treated within 90 minutes are hypothesis-generating and being investigated in FASTEST part 2.
Major Points
- No improvement in functional outcome: mRS distribution at 180 days was identical between groups (OR 1.09, 95% CI 0.79-1.51, p=0.61); trial stopped for futility at second interim analysis (conditional power <1%).
- rFVIIa significantly slowed haematoma growth: ICH volume change from baseline to 24h was -3.7 mL less with rFVIIa (p=0.0011) and ICH+IVH growth was -5.2 mL less (p=0.0011).
- Slowing bleeding did not translate to improved outcomes, suggesting that a mean reduction of ~3-4 mL may be insufficient; modelling suggests 6-12 mL may be needed to improve functional outcomes.
- Life-threatening thromboembolic complications within 4 days were significantly higher with rFVIIa: 15 (<5%) vs 4 (1%), RR 3.41 (95% CI 1.14-10.15, p=0.020).
- Subgroup signals (non-significant but prespecified): spot sign OR 1.86 (0.94-3.68) and treatment within 90 min OR 1.82 (0.98-3.40) -- patients with both features showed the greatest exploratory benefit.
- Adaptive enrichment to age <=70 years was added during the trial but also showed no benefit (conditional power <1%).
- FASTEST part 2 (NCT07227246) is ongoing, recruiting additional sites specifically targeting patients with a spot sign or treated within 90 min.
- CT angiography should be integrated into standard ICH assessment workflows to rapidly identify patients with spot signs and ongoing bleeding risk.
Design
Study Type: Multicentre, double-blind, randomised, placebo-controlled, adaptive, phase 3 trial
Randomization: 1
Blinding: Double-blind (all investigators and participants masked)
Allocation: 1:1 (step-forward randomisation using lowest pre-randomised kit at a site)
Enrollment Period: December 3, 2021 – October 1, 2025
Follow-up Duration: 180 days
Centers: 93
Countries: USA, Japan, Canada, Spain, Germany, UK
Sample Size: 626
Analyzed: 626
Analysis: Ordinal logistic regression for primary outcome, adjusted for age, baseline ICH volume, baseline IVH volume, pre-stroke mRS; modified ITT
Power Calculation: 388 participants per group (860 total) for 80% power to detect 10% absolute difference in mRS 0-2 at 90 days; adaptive re-estimation allowed up to 1330 participants
Registration: NCT03496883; EudraCT 2019-003722-25; EU CTR 2024-517383-28-00
Inclusion Criteria
- Age 18-80 years
- Spontaneous ICH 2-60 mL on baseline CT
- IVH in less than two-thirds of one lateral ventricle or less than one-third of both lateral ventricles
- Glasgow Coma Scale score at least 8
- No evidence of recent ischaemic stroke or myocardial infarction (past 90 days)
- No anticoagulation use within the past 7 days
- No other structural cause of ICH
- Treated within 2 hours of stroke onset or last known well
Exclusion Criteria
- Ischaemic stroke or MI within past 90 days
- Recent anticoagulation use within 7 days
- Known structural cause of ICH (AVM, aneurysm, tumour, etc.)
- Planned neurosurgical evacuation within 24 hours of enrollment
- Pregnancy
- ICH volume >60 mL or <2 mL
- Significant IVH (>2/3 of one lateral ventricle or >1/3 of both)
Arms
| Field | Recombinant factor VIIa | Control |
|---|---|---|
| N | 328 | 298 |
| Intervention | Recombinant factor VIIa 80 µg/kg IV over 2 min (maximum dose 10,000 µg or 10 mg) | Identical volume placebo (same solvent and process) IV over 2 min |
| Duration | Single dose | Single dose |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Ordinal modified Rankin Scale (mRS 0-2, 3, and 4-6) at 180 days, measured by Rankin Focused Assessment Tool; analysed by ordinal logistic regression | Primary | 134/298 (45%) mRS 0-2; 76/298 (26%) mRS 3; 88/298 (30%) mRS 4-6 | 151/328 (46%) mRS 0-2; 80/328 (24%) mRS 3; 97/328 (30%) mRS 4-6 | 1.09 | 0.61 |
| mRS 0-2 at 90 days | Secondary | 113/298 (38%) | 132/328 (40%) | 0.87 | |
| mRS 0-2 at 180 days | Secondary | 134/298 (45%) | 151/328 (46%) | 0.87 | |
| Change in ICH volume from baseline to 24h (mL) | Secondary | 5.5 mL (SD 13.7) | 1.9 mL (SD 7.0) | 0.0011 | |
| Change in ICH plus IVH volume from baseline to 24h (mL) | Secondary | 7.4 mL (SD 19.6) | 2.2 mL (SD 8.4) | 0.0011 | |
| All-cause mortality at 180 days | Secondary | 22 (7%) | 20 (6%) | ||
| Subgroup: mRS at 180 days -- spot sign present | Secondary | 77 patients | 71 patients | NS (exploratory) | |
| Subgroup: mRS at 180 days -- treated within 90 min | Secondary | 92 patients | 82 patients | NS (exploratory) | |
| Life-threatening thromboembolic complications within 4 days (primary safety) | Safety | 4 (1%) | 15 (<5%) | 0.020 | |
| Life-threatening thromboembolic complications within 90 days | Safety | 11 (4%) | 21 (6%) | 0.15 | |
| Life-threatening thromboembolic events within 4 days | Adverse | 4 (1%) | 15 (<5%) | 3.41 | 0.020 |
| Deep venous thrombosis | Adverse | 4 (1%) | 7 (2%) | 1.59 | |
| Acute myocardial infarction | Adverse | 4 (1%) | 3 (1%) | 0.68 | |
| Acute cerebral infarction | Adverse | 11 (4%) | 18 (5%) | 1.49 | |
| Acute pulmonary embolism | Adverse | 3 (1%) | 5 (2%) | 1.51 | |
| Mortality at 180 days | Adverse | 22 (7%) | 20 (6%) | 0.83 |
Subgroup Analysis
No prespecified subgroup analysis reached statistical significance. Prespecified subgroup signals favoring rFVIIa (non-significant): spot sign present (OR 1.86, 95% CI 0.94-3.68) and treatment within 90 min (OR 1.82, 95% CI 0.98-3.40). Exploratory analysis showed the greatest potential benefit in patients with both a spot sign AND treatment within 90 min. Post-hoc logistic regression interaction modelling suggested patients with both features may derive meaningful functional benefit. These findings are being prospectively tested in FASTEST part 2 (NCT07227246). Age (<=70 vs >70 years), country/region, ultra-early haematoma growth rate, and ICH volume subgroups showed no differential benefit.
Criticisms
- Trial stopped early for futility -- underpowered for subgroup analyses, which remain exploratory
- Mean haematoma growth reduction (~3.5 mL) appears insufficient; modelling suggests 6-12 mL may be needed to improve functional outcomes
- Thromboembolic risk is a real and significant concern -- limits widespread use even if efficacy were confirmed
- Asian patients comprised ~55% of the sample (due to Japan enrollment); findings may not be fully generalisable to other populations given different ICH etiologies (hypertensive-dominant in Japan vs more diverse in USA)
- Adaptive enrichment to younger patients (<=70 years) was added mid-trial but also showed no benefit, reducing confidence in age-based patient selection
- Only 45% of participants received treatment within 90 min or had a spot sign -- the 'ideal' subgroup was a minority, limiting the ability to test these hypotheses definitively
- Emergency consent procedures and mobile stroke units add operational complexity not universally available
Funding
National Institute of Neurological Diseases and Stroke (NINDS); Japan Agency for Medical Research and Development (AMED); Novo Nordisk (supply of study medication)
Based on: FASTEST (The Lancet, 2026)
Authors: Broderick JP, Naidech AM, Elm JJ, ..., for the FASTEST Investigators
Citation: Lancet. 2026;407:773-783
Content summarized and formatted by NeuroTrials.ai.