PREMIUM Trial: Amplatzer PFO Closure vs Sham for Migraine, Co-Primary Efficacy and Safety Endpoints (Tobis et al., 2017)
In patients with 6 to 14 migraine days per month refractory to at least 3 preventive medications, does Amplatzer PFO closure increase the proportion of patients with at least a 50% reduction in migraine attacks compared with a sham right heart catheterization?
Tobis JM et al. (JACC 2017;70(22):2766-2774) · doi:10.1016/j.jacc.2017.09.1105 · 230 patients
Population
Included
- 6 to 14 migraine days per month
- Failure of at least 3 preventive medications
- Right-to-left shunt defined by transcranial Doppler
Verifier-confirmed inclusion
- 6 to 14 migraine days per month
- Failure of at least 3 preventive medications
- Right-to-left shunt defined by transcranial Doppler
Co-Primary Outcomes
Efficacy: responder rate. Safety: adverse events. Report both.
EFFICACY CO-PRIMARY NOT MET. Responder rate, defined as at least a 50% reduction in migraine attacks, was 38.5% with the device and 32.0% with sham (P=0.32). The safety co-primary WAS met. That is a real and separate finding: in this population the procedure was safe, it simply did not prevent migraine. Quoting either co-primary without the other misrepresents the trial.
Difference not statistically significant (p = 0.32); no difference band is drawn.
Responder rate: at least a 50% reduction in migraine attacks
No confidence interval was recoverable for any PREMIUM result, so the between-arm difference is not stated as a number and no NNT is computed. The secondary endpoints did separate: headache days fell (P=0.025) and complete migraine remission reached 8.5% versus 1.0% (P=0.01). Those are secondary results on a trial whose efficacy co-primary was null.
Study Arms
- Agent
- Amplatzer PFO Occluder
- Route
- Percutaneous transcatheter device implantation
- Frequency
- Single procedure
- Route
- Sham right heart catheterization
- Frequency
- Single procedure
Double-blind. This is the strongest control arm in the PFO-for-migraine literature, and the trial still did not meet its efficacy co-primary.
Trial Design
Type
- Multicenter, double-blind, SHAM-CONTROLLED randomized trial (the control arm underwent a right heart catheterization)
- Transcatheter PFO closure with the Amplatzer PFO Occluder vs sham
- CO-PRIMARY endpoints: responder rate (at least a 50% reduction in migraine attacks) AND adverse events
- Efficacy co-primary NOT met (38.5% vs 32.0%, P=0.32); safety co-primary MET
- Enrolled 6 to 14 migraine days per month, failure of at least 3 preventive medications, transcranial-Doppler-defined right-to-left shunt
Timeline
Published Journal of the American College of Cardiology 2017;70(22):2766-2774
N
230
Enrollment
230 patients randomized, 117 device and 103 control analyzed (NCT00355056). Amplatzer PFO Occluder against a sham right heart catheterization, double-blind. Co-primary efficacy and safety endpoints. Published JACC 2017.
ClinicalTrials.gov
NCT00355056Bedside Pearl
PREMIUM (JACC 2017) had CO-PRIMARY endpoints. The efficacy co-primary, responder rate with at least a 50% reduction in attacks, was NOT met: 38.5% device vs 32.0% sham, P=0.32. The safety co-primary WAS met. Headache-day reduction (P=0.025) and complete remission of 8.5% vs 1.0% (P=0.01) are secondary and sit on a failed primary. This is the double-blind sham-controlled trial in the group, and it found no efficacy signal. No NNT.