Plus, learn more about this severe form of pediatric epilepsy from Child Neurology Foundation Executive Director Amy Brin in this episode of our Seizing Life podcast.

Download Full Transcript

Audience Q&A with Dr. Jeff Noebels

Do all IS cases have a genetic cause? And if they don’t, what are some of the other presumed causes?

Right. So, IS, it’s not incredibly rare, but it’s uncommon. And among those who have it, we have now discovered that there are certain genes that are now well known that can underlie the syndrome. I mentioned tuberous sclerosis and Arx. There are some others. I won’t give you their names. But there are probably five to ten genes that we know about and we can look for in the case. And that can likely explain features of the syndrome that the infant has. As far as acquired causes, I mentioned perinatal insults and maternal CNS infections have been associated with it. Beyond that, that seems like a general category. We don’t know specifically which viruses, what kind of insults, because they often created a lot of damage in the developing brain, and so we can’t always predict what kind of epilepsy will emerge from that. So there are both genetic and acquired causes of this syndrome.

I understand that IS results secondary to a brain injury. Does the research differentiate between genetic causes of IS and those secondary to a brain injury?

What we know from these and the research is usually the firmest information we can get is from animal models at a very basic level. So we don’t have too many of those models in the animal. There are various drugs that can be injected into a developing brain that seem to produce spasms and even the EEG counterpart, the abnormality, the hypsarrhythmia to study. But we’re not confident always that those are the same as what happens in human. It’s just a model. They allow us to study basic parameters of this disorder without really telling us this is what any child might have. So there are ways of approaching this scientifically. The genes are probably the clearest and most reproducible way of learning more about each child with infantile spasms.

Do infantile spasms have an autoimmune etiology? And have auto antibody prevalence been studied in this population?

So in the group that CURE  Epilepsy brought together, we’re a group of scientists and clinicians, and we sat down and quickly realized that there are many questions we don’t understand. The autoimmune concept, just to explain to people, is that the nervous system in the body will start to generate antibodies against itself. Obviously a very dangerous chain of events, because you begin to attack normal structures in your brain. And there are several forms of childhood epilepsies in particular that have this autoimmune cause, where if you look at the cerebrospinal fluid, you can find antibodies directed against the very molecules that you need to properly signal the brain. These receptors can become degraded by your immune system and seizures can result. I’m not certain that we have a clear knowledge base on whether this is one form, whether infantile spasms can arise from it, but it’s a very tempting hypothesis that deserves to be explored. The forms that I’m aware of don’t cause infantile spasms, but they do cause seizures.

Now on the other side of the question, is that well why does prednisone and ACTH these steroids, why are they so effective as treatments? Because we know that they do impair, they tune down the inflammatory response in your body. That’s why they’re usually prescribed. So therefore, does that mean that the infantile spasms was an inflammatory disorder? Not necessarily, because these steroids have many other effects, including actually acting on GABA receptors which is the target for many of our best antiepileptic drugs. So there is some evidence that perhaps there could be an inflammatory response. I don’t believe it’s ever been well studied or firmly demonstrated. But just because a prednisone works so well when it does in blocking infantile spasms, doesn’t mean that that’s the mechanism for generating.

How often are anti inflammatory drugs used in this population?

Well in the sense of that prednisone is the front line treatment for these infants, so I would suspect that whenever it’s properly diagnosed and the prednisone is available, that is the first medicine that the child will see. It doesn’t always work. Also as I mentioned, infantile spasms sometimes can go away anyway without treatment. Another interesting aspect of the infantile spasms part of this syndrome, it’s what brings the child to the doctor’s attention. It is not necessarily itself a epileptic seizure in the sense that it may not actually be damaging the brain. So as I mentioned, one of the fascinating features of this is that it looks like a normal reflex. It’s just present in the nervous system at an age when that normal reflex would have disappeared. There’s no sign in the EEG, in the rest of the brain, of a seizure when this happens.

And so to my mind, we don’t have to worry as much as we do with convulsive seizures about whether the event itself is actually damaging to the brain. We would like it to go away, wish it wasn’t there in the first place, but it’s really just a tell tale that something developmentally is wrong in the wiring of the brain that these reflexes persist in some way or return. And as nice as it is to get rid of them, the infantile spasms themselves to me aren’t as much of a danger as just an important warning sign that there’s something else the matter, and if we know how to treat seizures that are to come, maybe we can prevent them.

Do you know whether combination treatment with estradiol and ACTH or vigabatrin will provide a greater level of effect?

Some of those studies are ongoing. So one of the things we decided among the CURE Epilepsy group would be to find out if in our model does our mouse model respond to prednisone, and we found out that it does. But what we haven’t done yet is to combine the two, prednisone plus estradiol, and see if that would really allow us to either use smaller doses or treat for a shorter period of time. Those are studies that we would really like to complete and get an answer for.

Do you think that there would be a difference in the actions in males compared with females in your study?

Well that’s interesting, because this was an excellent gene. And for everyone, that means that males will only have a single X chromosome. If they have the mutation on that chromosome that they inherited, they will have the disease. Whereas a female has only 50% chance and possibly no chance because she has a second normal X chromosome that could protect her from the disorder. Whereas the male has only one X chromosome to rely on, and if it has the mutation, you’re affected. So all of our studies are actually, and for this particular gene, are done in males, because they’re the only ones that are affected. Now other genes may not show that sex difference and they’ll be very interesting to see if estrogen levels, which might be different in females, are actually protective in that sense. So we don’t know the answer to that question, but it’s interesting.

Do we know what proportion of patients with IS have a TSC mutation? And on these lines, could you share your thoughts on the potential of mTOR inhibitors for the treatment of IS?

I’m sorry, I don’t know the answer. This is still really rare, and I’m sure there’s some studies that could answer that, your first question. But I don’t have that fact at my fingertips. But we’re talking about an uncommon … both disorders are somewhat uncommon. You hear about them a lot because we have the genes and we’re working on them intensively. But I don’t have the epidemiology of that question yet. But the mTOR inhibition issue is certainly an interesting one, and just like the estradiol, mTOR inhibition seems to be remarkably effective in certain kinds of epilepsy, including TS. But I don’t know whether it’ll be effective in all forms, or even same forms as the ones where estradiol will work. So again, these are open questions, and see how well we can prevent damage using these different sort of targeted therapies for very specific molecular pathways and cells.

What is your opinion on the existence of truly idiopathic infantile spasms where they were normal prior to diagnosis, rapid response to treatment, and normal tests, and normal developmental outcome?

Well my opinion, that word idiopathic helps a lot of people out because it’s a Greek term. That means we have no idea of what’s going on. So idiopathic infantile spasms up until recently was all of them were idiopathic. Now we have a few genes for some of them. But I can’t really comment as a group on what the natural history of idiopathic infantile spasms are. If you go into the old literature of infantile spasms before we knew these subtypes, then everything you read would be true, because they couldn’t distinguish between different types. Now you wonder whether something is idiopathic because you haven’t looked for all the things that we do know about yet. So you begin to question the use of the term idiopathic because we know there are certain answers. And in fact, the field of epilepsy as a whole has seemed to move away from the term idiopathic epilepsy, and now they call it genetic epilepsy under what I guess is the optimistic assumption that everything that doesn’t have an observable cause during the lifetime must be genetic, and therefore has a discoverable genetic problem.

We won’t probably ever be able to discover every single gene that contributes to epilepsy in every single person. But there is a sense that the entire field is now assuming that most cases, the more we study them, if we can get a pure culture of a specific type of epilepsy and study it intensely, we will find the cause of that particular type. But I think all of our listeners know that epilepsy comes in so many varieties. Different ages of onset, different seizure types, different responses to antiepileptic therapy, that we know we’re dealing with a very complicated disorder. Certainly it’s as complicated as all the cancers you’ve heard about, and that we just need to learn a lot more about these disorders and we need to split them apart and look at them, and then see if we can lump them back together.

Do you know of chloride regulation deficits in interneurons that impair migration or development of interneurons to cause infantile spasms?

Most ion channel mutations, chloride channel is one of them, are not known to interfere terribly with migration. At the moment, there’s a migration disorder in one of a number of forms is probably the best explanation for infantile spasms as far as I know. So of the chloride channels, I can’t think of one. There’s one in mice that obliterates the hippocampus, and that’s not anything like what you see in infantile spasms. But the other ones that are either triggered by electricity or opened by GABA inhibitory transmitter, I don’t know that there are … there should be defects in migration. I don’t know of a specific model where that’s been shown, and it shows to have the infantile spasms phenotype, but I would not be surprised if there’s one about to be reported.

The next question actually is regarding CURE Epilepsy’s team science approach to the IS initiative. And the question is can you explain what this means and do you think that this approach may lead to discoveries that could lead to breakthroughs sooner?

Yes, absolutely. And this is something that I think is happening all over science, but really needs to happen fast in disease or oriented sciences because we’re all counting on these breakthroughs to come. And what’s happening in science right now is the enormous growth of very powerful tools, both ways of manipulating brain cells at the molecular level and of tracking their functions with amazing tools of micro imaging and electrophysiology of single cells and of large groups and networks of cells. We’re really starting to be able to powerfully examine the nervous system, when it works and when it doesn’t work properly. So all of those techniques require time and skill, and no one laboratory really can do it. The ones that really have a tool that they can perform correctly don’t know as much about disease, and the people with a lot of disease background don’t necessarily have all the tools. So the idea of getting together with a focus group and attacking a problem, such as was done with the CURE Epilepsy initiative, was really a smart thing to do, and I think we need to see a lot more of it in the future.

Has the ketogenic diet been used as a first line approach to treatment for these babies?

I’m not a pediatric neurologist, I’m an adult neurologist studying the developing brain. So I don’t have firsthand information. The ketogenic diet keeps coming up in every form of epilepsy where the conventional drugs don’t work well is not the front line treatment for infantile spasms, but it is certainly something to try if the front line treatments haven’t given the desired effects. It’s a difficult treatment. I think it’s been tried in most different forms of epilepsy. I don’t know that anyone is claiming that it is the second line, but it’s certainly available and could work in some cases.

Audience Q&A with Dr. Hussain

If an infant gains control over their infantile spasms and normalize the hypsarrhythmia, is there a greater likelihood of normal neurological development?

That’s a tough question. That is certainly the goal of therapy in the short term, and it seems to be pretty clear that if you don’t completely abolish spasms and the hypsarrhythmia, development will not turn out well. At UCLA, we follow almost 500 patients, and I can’t think of a single one who had a long-term burden of infantile spasms or long-term burden of hypsarrhythmia and did well.

In order to get a good developmental outcome, you have to abolish spasms. You have to abolish hypsarrhythmia. And you have to be a little bit lucky. You have to find that the cause of the infantile spasms is not itself something that can damage development. But in short, I would say you simply can’t tolerate any infantile spasms or hypsarrhythmia. If either of those are present, it means that you need to try different therapy.

If a gene panel doesn’t reveal a known cause genetic cause of infantile spasms, but a child has been seizure- and med-free for several years, do you recommend further genetic testing?

It’s tough. The answer to that question has to be pretty individualized, but usually the answer is no. The odds of us identifying a meaningful genetic abnormality that impacts our treatment – would tell us to begin or stop a therapy – is actually very, very low at that point.

On the other hand, if the parents are thinking about having another kid, it would be very nice to do that genetic testing and get a sense of whether there is any risk posed to the next child. There is unfortunately, very little data about the risk in general. In thinking about our cohort here at UCLA of nearly 500 patients, there have been only a couple cases in which a sibling also had infantile spasms after the first child.

Let’s say overall risk is pretty low. If it was me, I would want to do that genetic testing to figure that out. But there are also risks of genetic testing. Sometimes you might find out that a child, sibling, or parent are at risk for some other disorder associated with infantile spasms. It’s a big discussion, a difficult decision often, and you have to go into that decision-making process knowing all the risks and benefits of genetic testing.

If the first course treatment was not effective, does this equate to a delay in treatment and, therefore, a poor prognosis for future development?

We don’t actually know the answer. My sense is that the delay in finding an effective therapy probably poses some risk, but we don’t know how much. The first therapy not working probably means that the child has overall worse infantile spasms and may be at risk for bad outcomes on that basis.

The short answer is, I don’t know. But the second answer is that it doesn’t matter. If you’ve got ongoing spasms, you just got to change your treatment approach and be aggressive. Don’t be afraid of those side effects and focus on eradicating infantile spasms.

Is surgery only considered when seizures don’t respond to certain medications? Or can it also be an option when the MRI and PET scans indicate a specific portion of the brain causing abnormal EEG activity, indicating a risk of seizures, even after hypsarrhythmia has been resolved?

There is certainly no consensus in answering that, so I want point out a couple things. There are patients who have structural abnormalities – things like cortical dysplasia – who will respond to first line therapy and not need that surgery. We have multiple individuals in our cohort who responded to therapy and seemed to be doing just fine.

I will also tell you that there are patients with that exact story ; they had infantile spasms,  they had a lesion that could be removed, they responded to therapy, and we said, “Well, you don’t have hypsarrhythmia, you don’t have spasms, you seem to be doing well developmentally.” Then several years later or a decade later, we have either the return of infantile spasms, epileptic spasms, or other types of seizures. Then we have to think, “Whoa, it would have been nice to have removed that piece of brain back when that patient was an infant and when the risks and costs of surgery would have been less.”

It’s a really tough decision. I think if you asked a neurosurgeon, they would be hard pressed to remove pieces of brain in a patient who did not have ongoing infantile spasms or hypsarrhythmia. But I’m not sure that’s the right decision.

I would put it this way though: if you have ongoing spasms or hypsarrhythmia and you have identified a lesion that can be removed to potentially cure the infantile spasms, that is almost always the right path. We’ve seen that all the relapse rates are pretty high across the board, they’re much lower and those patients who are good surgical candidates and undergo a successful surgery.

You mentioned the importance of immediate diagnosis. To clarify, does that mean from the first visible spasm or from the onset of hypsarrhythmia? Could there have been hypsarrhythmia for weeks or months prior to the first visible spasm?

It refers to the interval from the first identified spasm to effective treatment. And it’s very possible that hypsarrhythmia is brewing or emerging or growing before that first spasm.

A big focus of ongoing research is to identify patients who are at risk and then sequentially check their EEG every few weeks looking for the possibility that hypsarrhythmia or infantile spasms are on the way. That might be an opportunity to treat infantile spasms when they aren’t as bad, meaning there maybe a higher opportunity to prevent them from ever happening at all.

What are the chances of stopping spasms with the combination of ACTH and vigabatrin, if the spasms have been going on for one year or more?

The odds are not great, but they’re not zero. When we look at the patients who are relatively new to the treatment combination, the response rate is in the mid-70s, about 75% is our best estimate. The odds of response after months or years of ongoing spasms – especially with ongoing hypsarrhythmia – is considerably lower. We don’t have good estimates of what that risk looks like.

But there have been plenty of patients who have had spasms for a year and failed specific therapies, like the first line therapies, but when the medications were tried a year later, they did work.

I think it’s worth consideration. Just because a therapy didn’t work a year ago doesn’t mean that it won’t work now. Infantile spasms and hypsarrhythmia are dynamic process and they’re changing. And just because it didn’t work in the past doesn’t mean that it won’t work now.

Unfortunately, the reverse is also true. You can imagine a patient who was diagnosed quickly, who responded to say ACTH and that spasms returned six months later. Just because they responded the first time doesn’t mean that they’ll respond the second time.

If a child’s spasms are brought under control, is this child more likely than the general population to develop epilepsy down the road?

It’s absolutely true. We don’t quite know the magnitude of that risk. But I would say that it’s pretty substantial. It also depends on the cause of epilepsy. For example, if a patient has one of the most common structural abnormalities that cause the infantile spasm (something like a focal cortical dysplasia), then the risk of epilepsy down the road in the absence of surgery is pretty substantial. I would say it’s probably in the neighborhood of 50 or more percent.

But it really varies. It’s hard to predict. I would just say that, yes, you’re at elevated risk. You have to be on the lookout for the return of infantile spasms or the emergence of other types of epilepsy.

You mentioned that vision loss can accompany infantile spasms. Do you have any advice if delayed speech is concerned in a child who has suffered from IS?

That is actually one of the most common concerns. To the extent that infantile spasms and hypsarrhythmia can hurt development, it seems that they disproportionately affect language. But we actually don’t know why that is.

When we conceptualized infantile spasms, they are a form of seizure and epilepsy that hijacks the entire brain, but seem to have a disproportionate impact on the temporal lobes. And the temporal lobes are very important, especially the left temporal lobe, for processing and understanding language. That seems to be a pretty big barrier for graduates of infantile spasms, including those who have responded pretty quickly to therapy.

We have a very low threshold for referring patients for speech therapy. I would say many cases, especially those which have responded robustly and quickly to therapy, are actually very good responders to speech therapy. It is something actionable, but not something too alarming. I would say that there are many, many patients who have good outcomes after infantile spasms, and many of them had some degree of speech delay.

Can other seizure types early on mask the presence of infantile spasms?

Absolutely. Another factor is that infantile spasms are sometimes subtle and don’t really register as seizures with most people. When you think about the general public, if they saw a video of a child having infantile spasms, most people would not say, “Oh, that looks like a seizure.” They would say, “I’m not sure what that is. It’s probably nothing. Maybe it’s an infant heartburn or gastroesophageal reflux.”

Infantile spasms visually don’t register and don’t register on that fear meter. But when you think about most other types of seizures, they’re rather dramatic, especially if you think about something like a generalized tonic-clonic seizure seizure, those are much scarier, much more obvious.

If you imagine a patient who’s having both generalized tonic-clonic seizures and infantile spasms, it’s pretty easy to see how everyone’s attention – parents, pediatricians, neurologists – would likely be focused on those generalized tonic-clonic seizures because they’re so dramatic. That is also probably part of the challenge. We have to keep infantile spasms in mind as part of the possibilities of seizures and infancy.

When we teach our trainees, our residents and fellows, we’re telling them, “Look, this is an infant. You need to have infantile spasms in the back of your mind, no matter what kind of seizure they’re showing you right now.”

Is vigabatrin known to cause myoclonic jerks? If these jerks are not epileptic, can you stop once the child is weaned?

It’s not exactly clear. There’s a pretty widespread rumor that vigabatrin can cause myoclonic jerks or myoclonic seizures, and that they can be classified as epileptic or nonepileptic. I would say, don’t worry about any of that.

I think it’s probably true that a minority of patients who are treated with vigabatrin have either the emergence or worsening of myoclonic seizures. These seizures are reversible after stopping vigabatrin therapy.

The big question is; if vigabatrin is working to stop the spasms, should you stop the vigabatrin to make those myoclonic jerks and myoclonic seizures better? I think that has to be considered on a case by case basis.