PDF: SAS
(2024)Objective
To evaluate the safety, pharmacokinetics, and efficacy of SFX-01 (sulforaphane delivery system) in reducing vasospasm and improving outcomes after aneurysmal subarachnoid hemorrhage
Study Summary
• CSF sulforaphane penetration was lower than anticipated, with most samples below quantification limits
• No significant reduction in cerebral vasospasm (MCA flow velocity ratio 1.04, p=0.545) or improvement in functional outcomes (mRS OR 1.647, p=0.237)
Intervention
SFX-01 300 mg oral capsules twice daily for up to 28 days (started within 48 hours of SAH) vs matching placebo capsules containing α-cyclodextrin only
Inclusion Criteria
Age 18-80 years, spontaneous aneurysmal SAH with radiological evidence, Fisher grade 3 or 4, within 48 hours of ictus, aneurysm treatment not ruled out, previously independent, informed consent from patient or legal representative within 24 hours of first dose
Study Design
Arms: SFX-01 300 mg BD (n=54) vs Placebo (n=51); administered orally or via nasogastric tube for up to 28 days
Patients per Arm: 54 patients allocated SFX-01, 51 allocated placebo (105 total randomized); Per protocol analysis: 46 SFX-01, 44 placebo
Outcome
• CSF haptoglobin ratio 1.981 (p=0.052), malondialdehyde ratio 1.12 (p=0.572)
• Maximum MCA flow velocity ratio 1.046 (95% CI 0.903-1.211, p=0.545)
• mRS at 180 days OR 1.647 (95% CI 0.721-3.821, p=0.237)
• SAHOT OR 1.082 (95% CI 0.464-2.525, p=0.855)
• Plasma SFN levels influenced by age and GSTT1 status; CSF levels predicted by BBB permeability (QAlb)
Bottom Line
SFX-01 was safe and achieved excellent plasma sulforaphane levels in acutely unwell SAH patients, but CSF penetration was lower than expected. Despite a trend toward increased CSF haptoglobin, SFX-01 did not reduce large vessel vasospasm or improve clinical outcomes.
Major Points
- First randomized controlled trial of sulforaphane (via SFX-01) in acute SAH patients
- Phase II multicenter double-blind placebo-controlled parallel-group trial
- SFX-01 was safe with no serious adverse events attributed to the drug
- Only adverse event difference was nausea (16.7% SFX-01 vs 2.0% placebo)
- Excellent plasma sulforaphane and metabolite levels achieved (SFN-GSH AUC 277, SFN-NAC AUC 415 h×ng/ml)
- CSF sulforaphane levels were unexpectedly low, with most samples below quantification limits
- Plasma SFN influenced by age and GSTT1 genotype status but not weight or BMI
- CSF SFN presence predicted by BBB permeability (QAlb ratio) rather than plasma levels
- Trend toward increased CSF haptoglobin (1.98-fold, p=0.052) but no change in malondialdehyde
- No reduction in MCA flow velocity (vasospasm) or improvement in functional outcomes
- Independent DSMB oversight with interim analysis after 30 patients
Study Design
- Study Type
- Phase II multicenter double-blind placebo-controlled parallel-group randomized clinical trial
- Randomization
- Yes
- Blinding
- Double-blind (patients, nurses, clinicians, lab staff, and investigators blinded; capsules identical in appearance; randomization by blinded third party using sequential pre-numbered treatment packs)
- Sample Size
- 105
- Follow-up
- 180 days
- Centers
- 3
- Countries
- United Kingdom
Primary Outcome
Definition: Three co-primary outcomes: (1) Safety via treatment-emergent adverse events, (2) Plasma and CSF sulforaphane and metabolite levels, (3) Maximum middle cerebral artery flow velocity on transcranial Doppler ultrasound
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| Placebo group: 48/51 (94.1%) experienced TEAE; Maximum MCA flow velocity measured | SFX-01 group: 53/54 (96.1%) experienced TEAE; SFN plasma AUC 16.2, SFN-GSH 277, SFN-NAC 415 h×ng/ml | - (Maximum MCA flow velocity ratio 95% CI 0.903-1.211) | Maximum MCA flow velocity p=0.545 |
Limitations & Criticisms
- Study limited to Fisher grade 3 and 4 patients, may not generalize to all SAH severities
- CSF sulforaphane levels unexpectedly low, with most samples below lower limit of quantification
- Enrollment up to 48 hours may be considered long, though actual mean time to treatment was 29-32 hours
- TCD-measured vasospasm may not be the most clinically relevant endpoint compared to delayed cerebral ischemia
- Imbalance in posterior circulation aneurysms between groups (21.8% SFX-01 vs 7% placebo)
- 15 patients in ITT analysis excluded from per-protocol analysis (mainly due to death before day 7)
- CSF sampling from either EVD or LP rather than standardized single source
- CSF sample timing not standardized relative to dosing time
- Some analyses conducted post-hoc to explain unexpected findings (exploratory in nature)
- Assay LLOQ limitations - samples below LLOQ may still contain detectable drug with more sensitive assay
- Unable to determine if lack of benefit due to inadequate CSF penetration vs failure to engage pathways vs pathway saturation
- No measurement of mode of delivery impact (oral vs nasogastric) on bioavailability
- Relatively short follow-up (180 days) for long-term cognitive/functional assessment
- Small sample size for phase II trial limits power for clinical outcome detection
Citation
Translational Stroke Research. https://doi.org/10.1007/s12975-024-01278-1