2G-03 - Paternal exposure to valproate and risk of neurodevelopmental disorders in children – a nationwide registry-based study in Norway

Background: The use of valproate by male patients during the three months prior to conception has gained attention due to concerns about a potential increase in the risk of neurodevelopmental disorders (NDDs) in their offspring. Following a review of relevant data, the European Medicines Agency's Pharmacovigilance Risk Assessment Committee issued a recommendation in January 2024 to exercise caution when prescribing these medications to male patients. However, this recommendation was made as a precautionary measure, and the signal has not yet been confirmed.

Objectives: To investigate the association between paternal exposure to valproate before conception and the risk of NDDs in children, utilizing nationwide registries in Norway.

Methods: This population-based health-registry linkage study included 339,500 offspring with available paternal data from 2010 to 2015, followed up until 2021. We compared the risk of NDDs among children born to fathers exposed to valproate during sperm development period (three months before conception) with those unexposed, including both population controls and restricted to those with related indications, and those exposed to lamotrigine or levetiracetam (composite active comparator). Comparisons were made for both antiepileptic drug monotherapy and combination therapy. Hazard ratios (HRs) were estimated using pooled logistic regression models in unadjusted analyses and using propensity-score fine stratification weighting (PS-FSW). Several sensitivity analyses were conducted to assess the impact of potential misclassification and confounding by indication and genetic factors.

Results: The fathers of 319 children used valproate as monotherapy during the period of sperm development period. Among these, 24 children (7.5%) developed NDDs during follow-up. Increased risks of any NDDs associated with paternal valproate use were observed in the unadjusted analysis but were attenuated as confounding control improved. The HRs were 1.67 (95% CI 1.10–2.54) in the unadjusted analysis using population controls, 1.43 (95% CI 0.93–2.18) in the analysis restricted to fathers with related indications, and 1.20 (95% CI 0.75–1.94) with additional adjustment using PS-FSW. In the active comparator analysis with PS-FSW, the HR was 1.02 (95% CI 0.57–1.82). No associations were observed between paternal valproate use and specific types of NDDs. These findings were consistent when considering combination therapy and across sensitivity analyses.

Conclusions: Our findings suggest that paternal use of valproate before conception is not associated with an increased risk of NDDs in children. Further studies, also in non-Nordic populations, are warranted to replicate these findings.