TARGET-EU: Risk of adverse birth and neurodevelopmental outcomes in children born alive to fathers exposed to valproate versus levetiracetam for generalised epilepsy

Study type

• Study topic:
  Human medicinal product
• Study type:
  Non-interventional study
• Scope of the study:
  Safety study (incl. comparative)
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Other
• Non-interventional study design, other:
  Sequential trial design (combining multiple, partially overlapping, cohorts)

Study drug and medical condition

• Study drug International non-proprietary name (INN) or common name:
  LEVETIRACETAM

  VALPROIC ACID
• Anatomical Therapeutic Chemical (ATC) code:
  (N03AG01) valproic acid

  valproic acid

  (N03AX14) levetiracetam

  levetiracetam
• Medical condition to be studied:
  Epilepsy
• Additional medical condition(s):
  Focus is on general epilepsy

Population studied

• Short description of the study population:
  Study population includes males exposed to Valproate or Levetiracetam diagnosed with generalized epilepsy, with ages between 18 and 50, fertile, and no contraindications for any of the two drugs, during January 2010 and March 2016 in the region of Valencia, Spain, as captured in the Valencia Health System Integrated Database. It also includes their offspring and the mothers of their offspring (when a linkage between the three subjects is possible).
• Age groups:
  • Adults (18 to < 65 years)

    • Adults (18 to < 46 years)
    • Adults (46 to < 65 years)
• Special population of interest:
  Other
• Special population of interest, other:
  Males diagnosed with generalized epilepsy and their offspring

Study design details

• Main study objective:
  To estimate the cumulative incidence of death, major congenital malformations, autism and ADHD in offspring born alive to fathers diagnosed with generalised epilepsy exposed to valproate compared to levetiracetam
• Setting:
  The study attempts to emulate a hypothetical target trial using routinely collected electronic health records in the region of Valencia, Spain, as recorded in the Valencia Health System Integrated Database in the periods 2010 to 2024. It uses primary care records, outpatient hospital records, pharmacy prescription and dispensing records, congenital anomaly registry records.
• Comparators:
  In the study population that satisfies the inclusion and exclusion criteria, the comparators are 1) males exposed to Valproate, in any dose, formulation or regimen vs 2) males exposed to Levetiracetam, in any dose, formulation or regime. For both arms, concurrent use of any other antiseizure medication is not allowed. Because patients with generalized epilepsy are unlikely to go untreated, an active comparator allow us to reduce confounding by indication.
• Outcomes:
  The primary endpoint is cumulative incidence by 8 years of age of a composite outcome of ADHD diagnosis, autism diagnosis, offspring death and offspring's congenital malformation. ADHD diagnosis and autism diagnosis were the initial endpoints of interest. However, offspring death and offspring's congenital malformation are part of the outcome because they are intercurrent events dealt with the composite strategy.
• Data analysis plan:
  The overall analysis tries to mimic the analysis reported in a hypothetical trial protocol, while dealing with electronic health records complications such as confounding.; ; For the primary estimand, the main estimand to support decision making, the cumulative incidence ratio will be estimated in the principal stratum of interest using a weighted Poisson regression model, using inverse probability of treament weights. To identify the principal stratum, we will assume that men who fathered a child in their assigned treatment would have also done so had they been assigned to the other treatment. Missing outcome and covariates data are imputed using Multiple Imputation algorithms. Treatment switch is handled via hypothetical strategy, meaning outcomes are set to missing after treatment switch. Sensitivity analyses will be performed to assess the robustness of the results to this and other assumptions made in the primary analysis. ; ; The analysis for estimand 2 is kept the same with one difference: how the intercurrent event of treatment discontinuation is handled. In Estimand 1 this is dealt with via treatment policy. In Estimand 2, this is dealt with via hypothetical strategy 3 months before conception (which implies missing outcome data), and via treatment policy in the 3 months before conception (same as in estimand 1). ; We refer to the protocol for details on the data analysis plan.
• Summary results:
  Not yet available