All right, I appreciate the invitation to come and speak to all of you this morning. This is me, these are my disclosures, and you've had the privilege to be in the room for all of these creative talks this morning where people have asked you to interact with them, to interact with each other. I'm just gonna drone at you for the next 30 minutes, so strap in. The learning objectives are to talk a little bit about the clinical implications of genetic testing associated with the corresponding yield of genetic testing, just to set some baseline. How do we then approach thinking about disease-modifying therapies as those become increasingly prevalent, the role that clinical trials might play incorporated within our clinical practices, and then thinking a little bit about how do we measure success as we move into this era of disease-modifying therapies. And I gotta show the most boring slide. Everybody's seen it 1,000 times, but I invite you to look at this a little bit differently. Because this is a treatment talk, the point I wanna make here is that the typical timeline for development of novel therapies is 15 to 20 years. So if you look at where we're at, from when these genes were discovered to when there might be a treatment that's targeting a particular genetic mutation, we're about to hit the upslope, the big upslope in the next five years or so. So we've got a few things that have hit clinical trial space now, and we have an incredibly large number of things that have preclinical work being done. So we need to get ready. This is just to make the point that it's the most common that actually correspond to a very large number of diagnoses that you get, but there's more than 860 genes now currently associated with epilepsy, and so there's this very long tail of what each gene individually has a very low yield, but collectively provide thousands of answers for families. When we think about what is a monogenic epilepsy, we just wanna level set a little bit here that we're talking about the left end of this spectrum where one gene is impactful enough to result in disease. You may have variability associated with that disease, you may have a number of things that change what that looks like for that particular individual, like their genomic background or environmental factors, but that gene by itself is gonna result in disease, and that's important to anchor to when you start to think about can we treat that or modify that disease with a targeted approach. I'm gonna actually be pretty quick here because I can't say this better than Tristan said this already in terms of the value of genetic testing. Regardless of whether there's a treatment approach, the fact that you have a name that you can put on it gives a hope that there is a direction you can go, and we know many, many families out there that have taken this on beyond all expectations and started their own organizations and worked to drive research in this space. So being able to find that diagnosis is really the first step. Obviously stopping the diagnostic odyssey, family planning implications, prognosis, and then of course most of this talk is gonna focus more on the management implications that there may be. And that will not just be about treatments, it's also about other aspects of management plans. This is some of our data between 2016 and 2020, just looking at diagnostic yield for genetic testing panels. It's not all one panel, there was a couple of different panels that were involved with this. But just looking at that yield that we see associated with genetic panels based on age of onset, it's not a surprise that the age of onset of epilepsy is associated with your diagnostic yield, but there still is yield in genetic testing even with onset at an older age. Even if there is normal development, there is still a yield for genetic testing, and regardless of the MRI findings, there is a yield to genetic testing even if it's higher when you have abnormalities, developmental anomalies associated with your MRI. Another way of thinking about it is in terms of seizure subtypes, and it may not be surprising that things like tonic seizures and epileptic spasms have the highest yield, where genetic generalized epilepsies, particularly absence, have sort of the lowest yield. But again, there is a diagnostic yield in genetic testing across any range of epilepsy seizure types. So impact, this is a way that we looked at this building off the study by Trudy et al, where we took a handful of genes that had known implications in terms of seizure selection, medications, and that's this category of precision medicine implication, and of those disease-causing variants that were identified from our panels, half of patients, half of patients had a treatment implication. So the concept that I shouldn't send genetic testing because it won't make a difference for treatment is gone. We can't claim that anymore. Half of patients had implications for the treatment specifically that we were prescribing, and you can see the breakdown of the types of genes that were involved in those implications. In some cases, a rare sliver, this 4%, we were trying to decide about do we pursue surgery or not, and that actually helped us to be able to inform on that decision. About a quarter, there were other medical management implications, and that's referring to specialty clinics, there may be published care guidelines associated with those genes for other things we need to think about, or recommended surveillance that we should consider associated with those genes. I know we're epileptologists, so all we care about is the epilepsy, but sometimes we have to think about the other stuff that might come with these genes as well. We had a wonderful talk about variants of uncertain significance earlier. For panels, because they're highly curated gene sets, if there is a variant there that is not known to be benign, it has to be reported as variant of uncertain significance. So interestingly, based on this recommendation to think about whole genome and whole exome, you often have less variants of uncertain significance reported on exome and genome compared to panels, and that's a misconception that most people don't realize. So that's not a reason to be scared about exome and genome. On panels, this data shows that we had a very large number of variants of uncertain significance, and you can see a little bit of this shift here, that when you are not white, then you have a higher rate. That is probably exacerbated a little bit on larger things like exome and genome, because the data for minority groups in that space is even worse, because they're less highly curated gene panels. So how did that come out? We had an average of 1.2 variants of uncertain significance per 100 genes on the panels. 15.7 VOOs per pathogenic variant. That's a burden. That's something we have to be prepared to deal with, and that we have to be able to deal with with families. So the concept of talking to families ahead of time and preparing them that the most likely scenario may actually be that we don't know what this means is really important. And again, we saw 1.7 compared to 2.1 if you are of a minority group, and so there is a statistically significant shift there that is important to consider. The impact of VOOs on families and on those discussions is important, but I think has already been covered, so I'm gonna move forward. All right, the modern era of disease-modifying therapies, we can think about them in a pretty basic schema. We can target things at the DNA level, the RNA level, or the protein level, and we have some examples of these that have happened in other spaces, certainly significantly neuromuscular, and in things that are in the pipeline, so to speak, for a number of epilepsy conditions. To make the point again that this is an onslaught that's coming our way, this is actually two years old, and was a collection of epilepsy-related disease-modifying therapies that were in development. So this list is out of date, and probably much larger at this point in time. The point isn't to read all of this, obviously, but to make the point, this is coming. I think one of the things that many people may not think about is how complex these treatments are to administer and to be able to operationalize. These are complex disorders to begin with, requiring multidisciplinary care, and now you add to that a scenario of having to engage procedure centers, infusion centers, inpatient services, outpatient services, and it may be for one-time treatments or for recurrent treatments. That becomes quite complicated. Our approach to this at Children's Hospital Colorado, really spearheaded by Julie Parsons, one of our neuromuscular docs, was to create this team of physicians, nurses, and administrative support that's able to create that operationalization. Our nurses help to support protocol development for each of our drugs so that we can implement those with the highest level of safety, quality, and support families through what can be quite arduous. This is just a brief list of the things on here, and you'll notice it's all neuromuscular except for CLIN2, baser liponase alpha enzyme replacement. So they're really ahead of us, and we need to learn from them as we continue to move forward with disease-modifying treatments for epilepsy. We like to say there is not disease-modifying treatments for enough things, and this is just making the point that this snowballs quickly. So 100 patients at our institution have had disease-modifying treatments within neurology alone. Again, many of those are in the neuromuscular space but not isolated to that space, and this is where we're gonna be looking at in epilepsy when we fast forward five years. We're gonna have hundreds of patients across the country who have been treated by epilepsy disease-modifying therapies. I think this is raising a really important point here around how we approach shared decision-making with families when we start to talk about treatments. It's something we haven't had to face a whole lot, historically, because this is a new space, but one of the lessons from looking at neuromuscular and other areas is that you actually often don't just get one new treatment or disease-modifying treatment in a particular gene, and if you look at SMA, where there's three FDA-approved drugs, now you have to talk about risk benefits and guide families about which treatment to go for, and you're like, yeah, yeah, that's not gonna happen for the, but there's multiple genes that have multiple therapies either in clinical trial or headed to clinical trial now, and part of that's because when you know how to do a successful trial in a particular condition, you do risk it, and so it actually incentivizes companies to go after additional treatments in that space. So we have to be prepared for this. You can also look at, this month, the FDA is reviewing two drugs for sickle cell, beta thalassemia has two drugs as well. This is not isolated to neurology either. This is gonna be the reality of these types of treatments. I'm not gonna go through this in an exhaustive fashion, but I will provide this as a reference that you can sort of think about this in terms of pluses and minuses of different approaches. So if you're talking about a gene replacement or a gene editing that may be one time versus ASOs or protein replacements that may require recurrent infusions or injections, and some of the advantages or disadvantages that you might have to consider from a biological perspective around off-target effects and those sorts of things. I think the other side is when we think about this in a parent's perspective, we're used to thinking about epilepsy, seizure counting as our measurement of success. This is a really nice study by Jenny Downs that was a qualitative study doing parent interviews looking at what their expectation was of a clinical trial outcome that would be meaningful based on looking at a traditional small molecule that might be taken orally for something like a gene therapy that was more invasive. And the point is, without reading all of this, it's a higher bar. Families are expecting more out of a disease-modifying therapy compared to an oral treatment. And we need to be prepared to assess that with the families and think about, are we meeting that bar for them? So some considerations in terms of how I think about this, because as was brought up earlier, I don't have a single appointment now where families don't ask about gene therapies. Whether there's an FDA-approved treatment or there's something going on or somebody's working in the lab. And so I think we have to be able to talk about this. And I think the most important point here is I don't use the word cure. Alter the course of disease, I certainly hope so. Disease-modifying, absolutely, that's the goal. Cure means you don't have disease. And we have to be really careful about what the words that we're using mean to the family and what expectations we're then creating. Because when you get to the point of then doing an informed consent, what do they think they're getting out of this for the risk that they're taking is a really important piece. There may be certain aspects that we have to think about that are unique to the child, may not necessarily want that to change. I've had conversations with families about improvements that also unmask additional challenges. For example, behavioral challenges that they now can express that they previously couldn't. It's not only upsides that we have to think about. In this case, even when we are doing what we are seeking to achieve, addressing the gene deficit and being able to improve things, we can create some challenges at the same time. And then of course there's things that we need to think about like is this disqualifying for other future treatments, et cetera, particularly in the vector-based gene therapy space. So these conversations start to become really complicated and we need to learn as a community about how to do these well. Some unique challenges to think about in terms of gene therapy trials or disease-modifying trials. So these are often rare diseases. They're gonna be relatively small cohorts that we have to be prepared to implement trials for. Traditional anti-seizure medications usually can be larger, they usually are shorter. If we are interested in development and other outcomes, we may need a longer time period to be able to assess that. The role of historic controls become increasingly important and raises the idea that we need to learn from every patient in rare disease as a community and work together for that. And again, the point around outcome measures being beyond seizure control in terms of what we measure success with. And to sort of summarize this and again point you to this resource, you can think about conventional anti-seizure medication, epilepsy trials in comparison to a disease-modifying trial that may be a bit different from a population perspective, duration, trial designs, and outcome measures. And I'll point to trial designs to offer one alternative that one could consider in comparison to the classic randomized control trials. Some of these trials are actually hard to blind, for example. And so you can do a randomized start or step wedge trial where you actually bin into cohorts. And I bring this up because whether you're involved in these or not, we all have to be able to assess the science behind these types of trials and be involved in thinking about whether or not these make sense for our patients. So this is kind of an interesting newer design that might be used in some disease-modifying trials going forward. So what about outcome measures? Well, just to anchor to some concrete things, this is CDKL5 data from two different sources, a natural history study and the FDA meeting with the family group that looked at what were the top concerns for families. So yeah, epilepsy's up there and seizures are very severe in CDKL5, so that's not surprising. But it's not the only thing and it's not actually the highest in both categories. So developmental delay, communication, GI issues, sleep are all things that were brought up as concerns. And so we need to be prepared to think about outcome measures both for clinical trials and for assessing efficacy in clinical care that are speaking to these challenges. This is another example from Angelman syndrome where again, seizures were identified as being quite high, but communication, motor, IQ and cognitive were right behind that as also being incredibly important for families. So there may be some issues with using seizures as the primary outcome for disease-modifying therapies and we have to think about this a little bit. There may be a range of epilepsy in a particular gene. Some disorders where not all patients have epilepsy. So if you're targeting that gene, do you exclude those patients that could still benefit potentially from that? That raises some issues. There's some, and I point to STXBP1 as an example here where seizures are very frequent in the first years of life but tend to come under control as they move into toddler and past five years of age. So if you use that as your primary thing you're looking at and you expect it to get better anyway, how do you judge the success of the trial? SCN2A is an example where there's a loss of function or a gain of function scenario that may influence whether they have epilepsy at all. And then this last piece I think is an interesting one because we're still learning about how we deal with the immune reactions to vector-based gene therapies. And often that involves a pretty decent course of steroids at least in current standard of care which we know could have the impact of modifying epilepsy control anyway. So sort of what time points do you start to think about that or judge that or how do you disentangle the immune modulatory regiments associated with gene therapies from the effect on their epilepsy? So a few takeaways for clinical trials and clinical care, while epilepsy is a major concern, it's not the only concern. So obviously we're gonna measure epilepsy but it shouldn't be the only thing we're thinking about. We should be aiming to improve the majority of symptoms, not just seizures. We'll need outcome measures that can then accurately, appropriately and ideally without too much burden on families be able to assess those things. So when we think about that in the era of precision medicine this is sort of summarizing various things that we talked about. You have to have access to the genetic testing that starts with the genetic testing as the base of the pyramid we can build on. We are increasing constantly through advances in science, technology, informatics, the ability to increasing, to get the larger percentage of patients diagnosed out of those tests. We are getting more and more to therapeutics and I'm drawing a little bit of a distinction here between something that may be FDA approved compared to clinical trial or drug discovery. But this is all happening rapidly and this is expanding at all levels of the pyramid. So if we wanna be able to support precision medicine therapies, there's a few implementation things that we need to think about as a community. How do we guide clinical care to learn from every patient across these rare diseases? Maybe learning health systems as an opportunity in that space. How do we build decision support tools both on the diagnostic side but also pointing towards clinical care? This is hard to keep up with. We have to have ways in which we're being informed as we go along and get these diagnoses and I think Dennis was pointing a little bit to some of these ideas as well and has done some work in this space around how do we democratize the knowledge that can support us in making these decisions in tight schedules. And then just thinking a little bit about how we do cross learnings here. So whether we're talking about vectors or we're talking about the chemistry associated with ASOs, the sequence that we're trying to address may not be the biggest source of safety concerns. And so how do we learn from two vectors that are the same in two different trials what we need to be watching for? Because the vector is gonna carry a lot of our morbidity rather than the gene that's actually being replaced. And we need to consider how as a community we can combine that data so that we can have aggregate learnings across these various mechanisms of novel therapeutics. And then I think we're gonna have a lot to learn from our immunology colleagues partnership around how we maximize our immune suppression regimens going forward as well. And with that, thank you for having me and I may have done okay on time. Thank you.

genetic testing

epilepsy treatment

disease-modifying therapies

clinical implications

novel therapies

multidisciplinary care