Therapeutic Considerations of Antiseizure Medications-The Teratogenic Risks of Antiseizure Medications: From Malformations to Cognitive Development

Video Transcription

Thank you. Thank you for the introduction. Thank you to the chairs, Dr. Weatherspoon and Dr. Pack, for having invited me. So today I'll be speaking, this evening I'll  be speaking about the teratogenic risks of anti-seizure medications from malformations, I have to look over the podium, to cognitive development.  Actually, what I may ask is, I'm just going to move to this side, because I can't see my slides here at all.  Oh, no, no, I'll stay here. So these are my disclosures. So the learning objectives for this evening are to review the live birth prevalence of major congenital malformations  with anti-seizure medications, ASMs, to describe newer data regarding neurodevelopmental outcomes with certain ASMs, and then to have a bit of a discussion  to allow us to appreciate some challenges and nuances with regards to tritogenicity.  So prevalence of major congenital malformations.  So this is a slide that's courtesy of Torbjorn Thompson and Paige Pinnell. So that summarizes the results of three major registries, pregnancy registries, in epilepsy.  They summarize the European registry, the North American registry, and the UK and Ireland registry. The major message is that amongst these different  anti-seizure medications, the general signal is that lebatoracetam, lamotrigine, and oxcorpazepine appear to have the, well, do have the lowest prevalence  of major congenital malformations, and that valproic acid has the highest prevalence of major congenital malformations.  What are these malformations? We often summarize them as many of them being midline defects. So these are neurotube defects, including the brain and encephaly.  These are spina bifida. These are cardiac defects, urogenital defects, oral and facial clefts. And they can also include skeletal defects and utero growth retardation.  It's worth noting that certain anti-seizure medications are more strongly associated with certain malformations. For instance, that phenobarbital is especially  associated with cardiac defects. Oh, yes, now I can see the slide here. Thank you. So those are major congenital malformations. And we've discussed, this has been  something that's been discussed in the literature for many, many years, well-described, well-known. What is more recent, there is newer data on the neurodevelopmental outcomes,  the adverse neurodevelopmental outcomes that we can see with certain anti-seizure medications.  This is a seminal study that was published by Kimford-Meader in 2013, an analysis of the North American registry, looking at the association of different anti-seizure  medications and a child's IQ at six years. And we see here that with valproic acid, the IQ, especially the verbal IQ, but also non-verbal,  is decreased as compared to other anti-seizure medications, as compared to phenytoin, as compared to lamotrigine carbamazepine. And not just a small degree, but we see here  verbal IQs of 105, 106 with the others, and only 97 with valproic acid.  These are findings that we've seen with other registries, with other data. So this is data from the UK and Ireland, showing again that valproic acid has an important impact  on the IQ of a child at six years after in utero exposure. Here, the comparative group was not other anti-seizure medications. It was individuals who did not have epilepsy,  who there was no in utero exposure to anti-seizure medications. These same effects were not seen with lamotrigine, phenytoin, or carbamazepine in this study.  What was additional with this analysis was that we see there's a dose effect, that higher doses of valproic acid are associated with lower IQs.  It's worth pointing out that a high dose of valproic acid in this study was 800 milligrams a day. So it was defined as high dose, but I  think in clinical practice, when we're treating adults, 800 milligrams is not that much. That's 250 in the morning and 500 at night. Anything more than that is high dose.  This is a very recent study, the Nordic Registry, where it's an administrative study using administrative data. And so the sample size is very large. They had 25,000 individuals.  They were able to look at their data, pregnant people with epilepsy. And in the middle graph, you see findings that are, again, consistent, consistently showing  that valproic acid is associated with a decrease in IQ that is associated with an increased prevalence of learning disability. What is different with this study  is that they also looked at autism spectrum disorder, the diagnosis of it as per ICD codes. They showed an effect of valproic acid.  They also showed an effect, however, with topiramate. So it's not just valproic acid that has a potential for adverse neurodevelopmental outcomes, but also topiramate.  So I've discussed the potential for physical tritogenicity, major congenital malformations with anti-seizure medications, how some are better than others, certainly,  that valproic acid is particularly problematic. I've discussed about neurodevelopmental, adverse neurodevelopmental outcomes that we see, especially with valproic acid,  potentially also with intermittent  also with, in terms of autism spectrum disorder, with topiramate. I haven't mentioned there are other studies looking at school performance with carbamazepine  and other anti-seizure medications, a bit beyond the scope of the talk this evening. After having presented that, I wanted to present some nuances, some things  that we should consider when we're making decisions with our patients, things that we should consider when counseling our patients.  So this is an article that was published seven years ago now by Torbjorn Thompson and his colleagues that I think does a wonderful job at explaining  some of the nuance around valproic acid, when the risks associated with valproic acid, but when sometimes it may be appropriate.  Because when talking about valproic acid, we cannot forget that it is the most effective at treating generalized tonic-clonic seizures in individuals with generalized epilepsy.  We've seen this with SANAD-1, where it was compared to lamotrigine and topiramate. We saw this relatively recently with SANAD-2, where it was compared to levetiracetam.  That if a person becomes pregnant and you try to withdraw the valproic acid, that there are risks associated with that as well. This is analysis of the European registry carried out  by, again, Torbjorn Thompson. And this is a, I won't go into the details, but this is each curve. This is the results of a Bayesian analysis. This is a posterior probability figure.  And what we see here is that if you are maintained on your valproic acid, the probability of having a generalized convulsive seizure  would most likely be 0.16, somewhere beneath 20%, that if you were to be switched to another anti-seizure medication, that risk would increase quite a bit,  would become much closer to 30%, and that if it was removed altogether, not surprisingly, then your risk of having a generalized tonic-clonic seizure  during that pregnancy would be well over 30%. I've included this second study just to highlight not the specific numbers, because the outcomes here are different,  and the sample size is much smaller, but that to highlight rather that what you are switched to doesn't have too much of an impact,  that if you're switched from valproic acid to levetiracetam or switched from valproic acid to lamotrigine, with both of them, the probability  of having a worsening of your epilepsy are equivalent.  Another nuance to the discussion is that the risks with valproic acid are dose-dependent. We see here how I said earlier, I  showed earlier, that the risk of major congenital malformations with valproic acid are, on average, 8%, 9%, 10%. But we see here how that risk can be much higher  with higher doses, that it can be as high, in this analysis of the North American data, can be as high as 25%. I admit the confidence intervals are very wide,  because the sample sizes were smaller, but nevertheless, we have a sense that it's quite high. But reassuringly, perhaps, that with lower doses,  the risk becomes much lower, and that with the lowest doses, and here it's up to 500 milligrams, so 250 BID, which is a dose that you can be successful with in some  patients, the prevalence of major congenital malformations approaches begins to approach that seen with lamotrigine. These are analysis of the European registry and the UK  and Ireland to demonstrate, again, that demonstrate that underlying this dose-dependent effect, that low doses of valproic acid approach the risk seen here.  We're looking at high dose lamotrigine.  Final caveat is the prevalence of major congenital malformations is non-independent across pregnancies, that what happens with the first pregnancy  can help you allow you to at least somewhat predict what will happen with the second. In this analysis of the Australian registry, they looked at 1,200 people, female sex,  who had had close to 3,000 pregnancies, that if there was a malformation, and here we are talking about major congenital malformations,  we aren't talking about adverse neurodevelopmental outcomes. If there was a malformation with the first pregnancy, the risk with the second pregnancy was very high.  It was 35% on average. It was as high as almost 60% if we were in the context of valproic acid. But if there was no malformation with the first pregnancy,  the risk with the second was actually lower. It was as low as 3%.  Final comment is about folic acid and the role it plays in teratogenicity. So we know from older studies, in fact, a particular RCT  that was carried out in Hungary in the late 80s, early 90s, showing that there is, on average, a 50% decrease in the prevalence of neural tube defects with folic acid.  It is worth noting that this benefit of folic acid on neural tube defects has never been demonstrated in people of childbearing potential with epilepsy.  This may be because of the general beneficial effect of increasing folic acid supplementation in our diets. This may also reflect the fact that the mechanism  of the neural tube defects seen in epilepsy is different than that in the general population. So with neural tube defects, it's uncertain. It's a bit unclear.  But I think it's very interesting, and what is newer is an analysis of MONID showing that periconception folate is associated with a mean 5.9 increase in full-scale IQ at six years.  So we're not sure about neural tube defects, but potentially we have data now that suggests there are cognitive neurodevelopmental benefits to folic acid.  There's a Norwegian study where they looked at 335 pregnancies with exposure to an anti-seizure medication. There was an eightfold increased odds of autistic traits  at 36 months if there was no folate, suggesting that folate was protecting against autistic traits in pregnancies to people with child-rearing potential with epilepsy.  What was interesting is that it appeared to be a plasma concentration-dependent effect, that a higher plasma concentration of folic acid was associated  with a lower risk or prevalence of autistic traits.  Higher may be better, but there may be a ceiling to what higher should be. That's with a very recent study that was re-emphasized recently, the data from Denmark, Norway, Sweden,  looking at 27,000 people with epilepsy, showing that in pregnancies where there was exposure to a high dose of folic acid, and they defined high dose  as anything of one milligram or more, so not exceptionally high. It wasn't four or five milligrams. It was one or more, that the instantaneous risk,  the hazard of cancer, mostly hematological cancers, was increased in the newborn.  So impact on clinical care and practice. The prevalence of major congenital malformations is lowest with lamotrigine, levotriazetam, and oxcarbazepine.  Valproic acid and likely topiramate are associated with adverse neurodevelopmental outcomes. These risks must be balanced. Here's the nuance.  These risks must be balanced against potential advantages of certain anti-seizure medications. Folic acid is likely beneficial, especially for neurodevelopmental outcomes, but there  may be risks associated with higher doses. Thank you.

Video Summary

The presentation addresses the teratogenic risks associated with anti-seizure medications (ASMs), particularly focusing on congenital malformations and neurodevelopmental outcomes. Key points include that levetiracetam, lamotrigine, and oxcarbazepine have the lowest prevalence of causing major congenital malformations, while valproic acid has the highest risk. Valproic acid also notably decreases a child's IQ after in utero exposure. The study shows valproic acid and topiramate can be linked to adverse neurodevelopmental outcomes, including autism spectrum disorders. However, valproic acid is highly effective for treating generalized seizures, posing a challenge when withdrawal is considered during pregnancy. Risks are dose-dependent, with higher doses leading to higher risks. Folic acid has shown potential benefits in reducing neurodevelopmental risks, though not conclusively for neural tube defects, and could be linked to increased cancer risk at higher doses. Balancing these risks with the therapeutic benefits of ASMs is crucial in clinical decision-making.

Asset Subtitle

Presenter:

Mark Keezer MDCM, PhD, FAES, is a clinician researcher at the Centre Hospitalier de l'Université de Montréal, Associate Professor (clinical) in the Department of Neurosciences and Adjunct Professor at the School of Public Health of the Université de Montréal. Dr Keezer is a member of the editorial board for Epilepsia, the epilepsy surgery commission of the International League Against Epilepsy, and the guidelines subcommittee of the American Academy of Neurology. Dr Keezer's research program is focused on the comorbidities of epilepsy.

Keywords

teratogenic risks

anti-seizure medications

congenital malformations

neurodevelopmental outcomes

valproic acid