← Back
NeuroTrials.ai
Neurology Clinical Trial Database

PREVeNT

Early Treatment with Vigabatrin Does Not Decrease Focal Seizures or Improve Cognition in Tuberous Sclerosis Complex: The PREVeNT Trial

Share Share Copied! Compare

Year of Publication: 2024

Authors: Bebin EM, Peters JM, Porter BE, ..., Krueger DA; on behalf of the PREVeNT Study Group

Journal: Annals of Neurology

Citation: Ann Neurol. Author manuscript; available in PMC 2025 February 28.

Clinical Question

Does preventative vigabatrin started at first epileptiform EEG activity, before clinical seizure onset, improve neurocognitive outcomes and prevent drug-resistant epilepsy in infants with TSC?

Bottom Line

Preventative vigabatrin initiated based on EEG epileptiform activity prior to seizure onset did not improve neurocognitive outcomes at 24 months in infants with TSC, nor did it reduce or delay the onset of focal seizures or drug-resistant epilepsy. However, it was associated with lower incidence and later onset of infantile spasms.

Major Points

  • Early preventative vigabatrin did NOT improve Bayley-III cognitive composite scores at 24 months in TSC infants
  • No reduction in overall epilepsy incidence or drug-resistant epilepsy at 24 months
  • No delay in time to first seizure after randomization
  • Vigabatrin DID lower incidence of infantile spasms and delay their onset
  • 19 of 27 placebo participants (70%) crossed over to open-label vigabatrin (median 44 days after randomization)
  • EEG epileptiform biomarker had modest predictive performance: sensitivity 0.826, specificity 0.611, PPV 0.731, NPV 0.733
  • Adverse event profile was similar between vigabatrin and placebo arms

Design

Study Type: Phase IIb multicenter randomized, double-blind, placebo-controlled trial

Randomization: 1

Blinding: Double-blind (participants, clinicians, psychologists, EEG readers blinded; only DCC lead statistician, independent medical monitor, and site research pharmacists unblinded)

Allocation: 1:1 randomization to vigabatrin or placebo at first detection of EEG epileptiform discharges

Enrollment Period: December 2016 to March 2020

Follow-up Duration: Up to 36 months of age (primary outcome at 24 months)

Centers: 12

Countries: United States

Sample Size: 84

Analyzed: 56

Analysis: General linear model with repeated measures for cognitive scores, controlling for sex and randomization age (<7 vs ≥7 months); Cox proportional hazards for time-to-event; Fisher's exact/chi-square for proportions; Hotelling-Lawley Trace for multivariate developmental outcomes

Power Calculation: Sample size based on prior biomarker study; expected ~37.5% would not develop epileptiform EEG. Needed 75-80 enrolled to randomize ~48 (24/group). Fisher's exact test at α=0.05 had >95% power to detect difference between placebo seizure rate 0.917 and vigabatrin 0.208, and >90% power if vigabatrin was 50% less effective (0.417)

Inclusion Criteria

  • Age ≤6 months at enrollment
  • Met diagnostic criteria for Tuberous Sclerosis Complex (TSC)
  • No history of clinical seizures
  • No evidence of subclinical electrographic seizures on baseline EEG

Exclusion Criteria

  • Born prematurely (<30 weeks' gestation)
  • Prior receipt of any anti-seizure medication (ASM)
  • Prior receipt of an mTOR inhibitor
  • Enrolled or planned enrollment in an experimental behavioral intervention study
  • Electrographic seizures on baseline EEG (post-enrollment exclusion)

Arms

FieldEarly VigabatrinControl
N2927
InterventionVigabatrin 50 mg/kg/day for 3 days, then 100 mg/kg/day, started at first EEG epileptiform discharge prior to clinical seizure onset; escalated to 150 mg/kg/day open-label upon any clinical or electrographic seizureIdentical placebo sachets started at first EEG epileptiform discharge; transitioned to open-label vigabatrin (100 mg/kg/day, escalating to 150 mg/kg/day) upon seizure onset
DurationUntil 24 months of age (with 2-week blinded transition to open-label thereafter)Until first seizure or 24 months of age

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Bayley-III cognitive composite standard score at 24 months of agePrimarySimilar to vigabatrin arm (specific value not reported in available text)Similar to placebo arm (specific value not reported in available text)Not significant (NS)
SecondaryNo significant difference between vigabatrin and placebo
SecondaryNo significant difference between groups
SecondaryNo significant difference between groups
SecondaryNo significant differences between groups at 12 or 24 months
SecondaryLower in the vigabatrin group than placebo
SecondaryLater onset in vigabatrin group
Secondary19 of 27 (70%) crossed over, median delay 44 days after randomization
SecondarySensitivity 0.826; Specificity 0.611; PPV 0.731; NPV 0.733
SafetySimilar between vigabatrin and placebo groups
SafetySerial ophthalmologic surveillance performed throughout study

Subgroup Analysis

Analyses controlled for participant sex and whether randomization occurred before or after 7 months of age (chosen because prior data showed mean first epileptiform discharges at 4.5 months and mean first seizure at 7.5 months)

Criticisms

  • Small sample size (n=56 randomized) limits power for cognitive outcome differences
  • High crossover rate from placebo to open-label vigabatrin (70%) dilutes any potential treatment effect on primary outcome
  • COVID-19 pandemic required substitution of ADOS-2 with BOSA and ADI-R for autism assessment, precluding planned autism analysis
  • Single country (US-only) limits generalizability
  • Median 44-day delay between randomization and crossover means placebo participants received vigabatrin relatively early after seizure onset
  • EEG biomarker specificity (0.611) and PPV (0.731) were modest, meaning some randomized infants may not have been on a path to refractory epilepsy

Based on: PREVeNT (Annals of Neurology, 2024)

Authors: Bebin EM, Peters JM, Porter BE, ..., Krueger DA; on behalf of the PREVeNT Study Group

Citation: Ann Neurol. Author manuscript; available in PMC 2025 February 28.

Content summarized and formatted by NeuroTrials.ai.