Valproate is a drug approved and commonly used in the treatment of epilepsy and bipolar disorder in Europe. However, literature evidence and clinical reports alerting on the risk of babies with birth defects and future developmental disorders born from mothers treated with valproate during pregnancy triggered a safety review from the European Medicine Agency (EMA) / Pharmacovigilance Risk Assessment Committee (PRAC) in 2014. The safety concerns were specifically for pregnant women, female children, and fertile women treated with valproate. This prompted the implementation of risk minimisation measures such as education materials and applying restrictions on valproate use to the label, to be set up by the pharmaceutical companies. However, various studies showed that the use of valproate by fertile women had not significantly changed in 2014-2016. In 2017-2018 the PRAC recommended new changes that were adopted by the European Commission on 31 May 2018: the prescribing conditions in the product information, and educational measures were revised, and a Pregnancy Prevention Programme was launched. The PRAC also requested companies that market valproate to carry out an observational study to evaluate and identify the best practices for valproate treatment discontinuation and switch in fertile women and pregnant women. The importance of this study lies in 1) understanding what are the best clinical practices of switching from valproate to another medication, in order to comply with the prescribing conditions and to avoid exposure during pregnancy, while avoiding the medical issues associated with medication discontinuation and 2) providing recommendations for the prescribing physicians
Valproate is licensed to treat epilepsy and bipolar disorder in Europe; however, it is associated with significant teratogenicity risks. Discontinuation of valproate and switching to another medication in women of child-bearing potential (WCBP) and pregnant women was discussed with the EMA and clinical experts during an Article 31 referral procedure, and it was agreed that the available recommendations were limited for clinical practice. Therefore, the PRAC requested the pharmaceutical companies marketing valproate to conduct an observational retrospective study in order to identify the best practices for discontinuation or switch of valproate to support the guidelines release. CPRD was selected as one of the appropriate data sources because of its national coverage and high representativeness of the primary care setting.
The primary objective is to identify patterns of switching or discontinuation of valproate associated with a successful switch after valproate discontinuation (defined below), and to evaluate the association between patients’ and disease characteristics with a successful switch after valproate discontinuation, in WCBP chronic users of valproate and in a sub-population of pregnant women.
This is a retrospective cohort study of WCBP chronic users of valproate for epilepsy or bipolar disorder, who discontinued valproate during the inclusion period (2014-2017), with up to one-year follow-up.
The primary outcome is the occurrence of a successful switch after valproate discontinuation, defined as 12 months continuous use of the newly switched drug (more details in the next section). The secondary outcomes are: 1) absence of clinical relapse within the follow-up period after index date as a proxy of clinical stability and 2) absence of valproate re-initiation.
Descriptive analyses will be conducted to describe the patterns of valproate discontinuation and use of new medication(s). Identification of risk factors associated with a successful switch will be assessed using a multivariable Cox proportional hazards regression model
The primary outcome will be the occurrence of a successful switch after valproate discontinuation, defined as 12 months continuous use of the newly switched drug, which represents an appropriate proxy of good tolerance and efficacy. Please, refer to the next section 'Objectives, specific aims and rationale' for justification. 12-months continuous use will be estimated based on prescription records and quantity supplied of the newly switched drug without any gap in supply of more than 30 days.
The patterns of valproate discontinuation and switch to other medication(s) will be described among patients who meet this primary outcome; then, the association between specific patterns and this success criteria will be evaluated.
The following secondary outcomes will be considered:
• Absence of clinical relapse for either epilepsy or bipolar disorder within the 1-year period after index date, as a proxy of clinical stability. Relapse will be defined as at least one hospitalization or emergency room (ER) visit for indication-specific reasons. Information related to specialist outpatient care delivered at hospitals will be considered to build an algorithm for proxy of clinical relapse in patients without linkage to inpatient hospital data,
• Absence of valproate re-initiation during the follow-up period.
The patterns of valproate discontinuations and switch to other medication(s) will be described among patients who meet the primary outcome and the secondary outcome ‘absence of clinical relapse’ as well as in patients who meet the primary outcome and both secondary outcomes.
The following other outcomes will also be considered:
• Hospitalization or ER visits for selected diagnoses (e.g. for suicide attempt, for psychiatric conditions other than bipolar disorder),
• Number of office visits to GP, Neurologist or Psychiatrist or other medical specialty (gynecologist) or other relevant health care professionals (epilepsy nurse in the UK),
• Death.
Sandrine Colas - Chief Investigator - Sanofi Aventis Groupe (France)
Sandrine Colas - Corresponding Applicant - Sanofi Aventis Groupe (France)
Cindy Li - Collaborator - Sanofi US Services Inc (USA)
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation;Pregnancy Register