Clermond-Ferrand – For a long time, the teratogenicity of anti-crisis drugs in epilepsy has occupied the majority of pharmacovigilance concerns concerning these molecules. However, neuro-developmental and autistic risks specific to these molecules have been gradually mentioned. International registers exist, fortunately, but are not enough to clearly rule on all risks and all of the molecules that have appeared in the last 20 years on the market. This situation creates a double problem, mentioned by the Pre Sophie Dupont (Neurology, Pitié-Salpêtrière, Paris) during French-language neurology days (April 15-18, 2025, Clermont-Ferrand) : « Future parents are worried, and doctors perplexed on the nature of the information to provide them, given their highly fluctuating nature ». However, she has taken stock of knowledge today robust, and the uncertainties that will have to be raised in the years to come.
Risk clarification
From the 1970s, the teratogenic risk of certain premiere and second generation antiepileptic drugs was mentioned. Then the number of new drugs exploded. Many countries participate in registers, such as EURAP, in which the registration of pregnant women processed by anti-crisis drugs makes it possible to develop knowledge. This is how this register recently made it possible to specify the dose-dependence bonds of teratogenicity linked to these molecules: sodium valproate and carbamazepine have progressive dose toxicity for the first, and from a threshold dose for the second. Things are less clear for topiramate, pregabalin. Finally, if levetiracetam and lamotrigine seem to be free from this risk, the data is insufficient for recent molecules such as lacosamide, zonisamide, or cenobamate.
Same observation for the neurodevelopmental risk and autistic spectrum disorders linked to these molecules: the main suspicions were mentioned in 2009 for sodium valproate, then topiramate, with a risk multiplied by 2 to 4 in the event of exhibitions in utero. However, recent register data contradict the question of autistic overrisque under topiramate and evoke on the contrary that this overview would be directly attributable to epilepsy itself. In addition, with regard to other anti-crisis drugs, data is nonexistent, limited or uncertain.
-Persistence of discordant data
« The folate deficit has been suspected of being involved in the teratogenicity of these drugs, but the discrepancy of data concerning the taking of folate in preconception is still not resolved ». Thus, it is not certain that they allow to prevent the occurrence of a Spina Bifida, while they could be associated with a better cognitive becoming of the children to be born. The problem could be based on the choice of the dose (beyond 0.4 mg/d), but the methodological biases associated with these studies do not make it possible to decide. Another uncertainty that should be raised is that of the neurodevelopmental risk linked to taking anti-crisis drugs by the father at the time of design. Studies evoke a higher risk in children born of fathers treated by ValProate. “” However, this study does not know the indication of prescription, which can induce a methodological bias. Another study, which is free from this bias, does not find the same risk, commented the specialist. We are now awaiting the results of an important dedicated study, which has been requested by the European Medicines Agency and which should soon be published. »
« The biggest problem to come is the total absence of data for the third generation of antiepileptics, except concerning lamotrigine, levetiracetam and oxcarbazepine “, Complete the Pre Sophie Dupont. Moreover, “ Perhaps we will never have these data because we know that we have third generation drugs devoid of teratogenicity ». This could cause difficulties for women who are not controlled by the latter. Also, ” The best would be to understand what this teratogenicity is based on ». The mechanisms have been particularly studied with valproate (Metabolism disorders of folate with reduction of folic embryonic acid disturbing the expression of genes, increase in embryonic oxidative stress, epigenetic modifications, acetylation or hypomethylation of histones …) or with topiramate (gabaergic signaling, DNA methylation, etc.). The fact that “ The teratogenicity of certain drugs only manifests itself in some children also suggests the existence of a field of genetic susceptibility in mothers ».
For the future, it will be necessary to structure the information so that the risks are as best as possible in real time, and to continue the registration of women in the registers. So, ” Given that it is not certain that there are robust data on all the most recent molecules, it is imperative to develop reliable and predictive animal models of potential teratogenicity and neurodevelopmental and possible autism over these drugs “Concluded the neurologist.
