AES 2024: THE FUTURE IS NOW

I often write a summary of the main lessons from the American Epilepsy Society meeting, but this year there was so much about epilepsy syndromes (which you also see referred to as Developmental and Epileptic Encephalopathies, or DEEs) that I didn’t have a chance to see many of the more general epilepsy presentations at AES, or even everything about DEEs!

Therefore this summary is not intended to be a summary of what was presented at AES, but simply the highlights of what caught my eye at the conference and where I see the field going.  

You can see my summaries from previous years at AES here: 2017, 2018, 2019, 2020, 2021, 2022, and 2023.

 

1 - THE FUTURE IS NOW: TANGIBLE PROGRESS IN DISEASE-MODIFICATION

In 2022 I talked about crossing the line from developing treatments for symptoms, to developing treatments for the cause of the disease. And the line from focusing on a few syndromes only (those with the most patients), to seeing treatments in development for many more. I called that escape velocity.  

At that time in 2022, that progress was illustrated by ongoing trials for Dravet syndrome (SCN1A haploinsufficiency) and preclinical progress in other DEEs. We were not there yet, we were just taking off. 

After a slower 2023, 2024 caught up with the speed and delivered us a fantastic year for DEEs, and in particular for Dravet syndrome. This was the first classic DEE (if we don’t count other neurodevelopmental disorders like Angelman syndrome and Rett syndrome, more on this below) to cross the line between having promising treatments for the cause of the disease in development, to having clinical proof of disease-modification.

For Dravet syndrome, 2024 has marked a before-and-after in therapy development, and two news at AES made all this difference.

First Encoded Therapeutics presented during the Dravet Syndrome Foundation Roundtable the evening before AES and announced that they have already dosed 5 patients with their gene therapy ETX101 using a virus to deliver a transcriptional activator for SCN1A. They expect to be able to share some safety and efficacy data in the second half of next year. Encoded is only recruiting very young children for their POLARIS Phase 1/2 program in Australia, the US and the UK, as young as 6 months of age. This means that if this gene therapy is successful, these might be the first few kids with Dravet syndrome that will be spared from developing the disorder, or that will only develop a milder form of it. And as a reminder, Dravet syndrome like all other DEEs comes with neurocognitive delay, behavioral problems, motor problems, drug-refractory epilepsy and (in this case) a high rate of seizure-related mortality. Because it can often be diagnosed during the first year of life, a gene therapy like this one could be used in the future soon after diagnosis and PREVENT the development of Dravet syndrome. At least that would be the ultimate success scenario.

But what if you already developed Dravet syndrome? That was the second big news at the conference, and for me the highlight of the entire meeting and of my 13 years working in this disease. A few days after the announcement from Encoded, Stoke Therapeutics showed the complete results from their Phase 2 in children and teenagers ages 2 to 18 with zorevunersen (STK-001), an ASO designed to increase the levels of productive SCN1A mRNA, including 24 months follow up of these patients, and we learnt that:

  • Stoke already identified the right dosing for the Phase 3 trials: 2 or 3 loading doses of 70mg followed by 45mg maintenance doses every 4 months.

  • They see sustained seizure reduction of over 80% at that optimal dose (this is more than fenfluramine, and with half of the patients already on fenfluramine!)

  • They documented growing improvements after 12 months and 24 months in …wait for it: receptive communication, expressive communication, personal skills, interpersonal relationships, play and leisure, coping skills, gross motor and fine motor skills. That’s it, the entire Vineland-3 scale improved in all domains, getting even better into the second year of treatment, while we know from the natural history studies that these patients don’t make improvements in any of these skills within a year.

To better understand the transformational improvement that these numbers meant, one of the principal investigators from the Phase 2 trial showed some videos of the before and after of a 12-year-old treated with zorevunersen. In my experience, what I saw is what many Dravet syndrome families have described over the years as “the cure”, and what we can medically call a truly life-transforming improvement across all the disease domains. Here is some description for what we saw in the video:

  • Before treatment the teenage girl had problems with motor coordination and balance, hesitation in movements when trying to kick a soccer ball and ataxia. She looked like many people with Dravet syndrome at her age. Yet the video of her playing soccer 8 months after treatment with zorevunersen showed a completely different posture, much more fluid, with balance, even able to control the ball with her foot to center it before kicking. This level of improvement in this short amount of time was not something I thought would be possible, and that’s the power of seeing videos. And then again, remember that patients on zorevunersen keep getting better over time and this improvement was already within the first year.

  • We also saw videos of her buttoning her shirt with much improved fine motor skills, and then some videos of her talking to her doctor and following some basic directions that showed a dramatic difference between a shy distracted kid who could not keep much eye contact or follow basic instructions as in “touch my finger and then touch your nose” to a completely transformed teenager within a year, who would sit up and cheerfully follow the instructions and chat with her doctor about her school and favorite subject. And she went from 5 seizures a month to 1 or 2 seizures a month so this improvement is clearly related to the genetic upregulation and not to any dramatic effect secondary to seizure reduction. The transformation is hard to explain with words so I hope you can see the video in Stoke’s website (during this first month you can see it here) and how she does not look like the same kid after 8-12 months of treatment.  

  • Stoke said that the improvements that we saw in this girl map into the Vineland as an average patient after treatment with zorevunersen. So this was not a super-responder case, this was a representative patient.

This clinical trial with zorevunersen in Dravet syndrome is the first proof of disease-modification in a DEE. The first successful transition from seizure management to SYNDROME management, by targeting the root cause of the disorder. Any improvements in cognition or behavior (mainly focus) that we have seen over the years with seizure drugs like fenfluramine dwarf in comparison to the magnitude and sustained progress that we are now seeing with zorevunersen. Finally the future is now.

The results with zorevunersen in Dravet syndrome are so strong that I worry about the possibility of unblinding during the Phase 3, because presumably (the design has not yet been announced) the Phase 3 will be about one year long yet clinicians from the Phase 2 report clear improvements as soon as 6 weeks into the trial. Unblinding due to efficacy already happened in the spinraza trials in SMA, and indeed zorevunersen is looking like a spinraza in the making.

And now that we know what the human brain can do, and how much plasticity is still in it if only we can come with the right genetically-targeted treatment, we look at the other DEEs and ask when will they get their Stoke moment.

For SCN2A developmental and epileptic encephalopathy, Praxis is developing an ASO to downregulate the ion channel expression in cases of gain-of-function mutations (elsunersen), but it is so far only used in a few cases under emergency use. My eyes are on the gene therapy from Capsida for STXBP1-related disorders, which is on track to get an IND in the first half of 2025 and it will be an intravenous-administered gene therapy to restore STXBP1 levels, so it is a first-of-a-kind for the DEE world and even for the entire neurology field. SYNGAP1 held a scientific conference the day before AES and they also seem to be next in line for disease-modifying trials, with the ASOs including the one from CAMP4 Therapeutics in toxicology evaluation which is the last step before filing for an IND and several other programs in development. CDKL5 deficiency disorder is also preparing for next-generation gene therapies, and we saw a presentation from UCDavis with a viral-delivered epigenetic gene therapy that can open up the inactive X chromosome to specifically drive expression of the CDKL5 gene copy and that is also in late preclinical stage.

So the future is now for Dravet syndrome, and around the corner for several other DEEs. Based on the speed of progress in Dravet syndrome, I estimate that the AES conference in about three years will probably be the one with clinical results for disease-modifying therapies in multiple DEEs since we are on track for multiple trials to start in 2025 and 2026. But hey, maybe things go even faster now that we know (and investors know) that cures, or life-transforming improvement across all disease domains, are a real possibility.

2 - BEYOND SEIZURES IN DEEs

I mentioned before that Dravet syndrome is not alone, and that if we broaden our focus to include other neurodevelopmental disorders that are often less featured at AES (despite also having epilepsy) we see that 2024 has been an unprecedented year when it comes to showing neurodevelopmental improvements in monogenetic disorders with ASOs and gene therapies:

  • In Angelam syndrome, Ultragenyx and Ionis have announced that they are progressing to Phase 3 trials with their ASOs that increase UBE3A expression. Ionis showed data from their Phase 1/2 HALOS showing improvements in communication, cognition and motor function in Angelman patients ages 2-50. Ultragenyx also released data this year from their Phase 1/2 trial in Angelman patients ages 4-17, also showed rapid improvement across many disease domains.

  • Some diseases are not obvious targets for ASO treatments, and in Rett syndrome we have one of the first neurodevelopmental disorders to get clinical data from gene replacement gene therapies. Neurogene recently shared data from the first Rett syndrome patients (ages 4-7) treated with their viral gene therapy, also showing improvements in fine and gross motor skills, communication and cognition, while girls with Rett syndrome don’t re-gain skills after their early regression. Neurogene run into safety problems at a higher dose of the gene therapy, but the lower dose had substantial efficacy and should be a viable dose moving forward.

This brings us to 3 neurodevelopmental disorders that this year reached proof of disease-modification with genetically-targeted treatments, and their value goes beyond Angelman, Rett and Dravet syndrome because what they tell us is that the brain keeps much plasticity that is ready to go as soon as we get the right therapies.

To be able to document these improvements, companies had to move beyond seizure counting and figure out which scales could be used for Phase 3 trials in these disorders to document non-seizure improvements. It seems from Dravet syndrome that the non-seizure scale for the Phase 3 will be the Vineland-3 scale, while in Angelman both Ultragenyx and Ionis are choosing the Bayley scale for their Phase 3 trials. In fact all these companies were able to interpret their open-label Phase 1/2 trials because there had been previous efforts to document the performance of this type of scale in a particular type of Natural History Studies called endpoint-enabling studies, showing minimal or no skill development beyond certain age for these disorders. None of these findings would have been possible without those natural history studies.

In Dravet syndrome, the endpoint-enabling studies were run by Stoke and Encoded specifically targeted to the ages of their interventions. Encoded deep-phenotyped Dravet patients under 5 in their ENVISION study, and was able to show an inflection point around 2 to 2.5 years of age when patients slow down in their neurodevelopment despite the best standard of care. Stoke further extended this age range in their Butterfly endpoint-enabling study to show lack of changes in seizure frequency and skills over 12 months in this population, which leads to a growing neurodevelopmental gap when compared to the neurotypical population. These results were presented at AES as the reference to interpret the progresses that we are now seeing with the disease-modifying therapies and also to justify trial design.

While in Dravet syndrome those studies were initiated and run separately by two biotech companies, in CDKL5 deficiency disorder the Loulou Foundation was able to convene several companies in their space to participate jointly in a single endpoint-enabling study called the CANDID study by creating a pre-competitive consortium. This year the CANDID consortium presented the baseline results from this study at AES, which includes scales like the Vineland-3 and the Bayley that seem to be becoming the frontrunners for Phase 3 trials in this space, and the study is now collecting longitudinal data for all these scales.

And other communities chose to run before the companies in their space were ready, and to launch endpoint-enabling studies to start documenting the suitability of those preferred-scales for their particular disorder and their progress over time. This is the case of the STARR study for STXBP1-related disorders and the ProMMiS study for SYNGAP1 which were also presented at AES.

These studies are key for a disease to be “trial ready” for beyond-seizure trials, and key for the transition from treating only seizures to treating the syndrome.

3 - BEYOND THE LARGEST DEEs

At this point you are probably tired of hearing me mention the same group of DEEs over and over. That is because they have the most patients and the most advanced pipelines, but 2024 has been, again, a very different year when it comes to going beyond the top DEEs.

It started with Longboard pharmaceuticals announcing last summer that they have agreed with FDA to run a Phase 3 trial for bexicaserin (a second-generation fenfluramine) in “all DEEs” as opposed to cherry picking two or three of the largest syndromes. Longboard, recently acquired by Lundbeck, was one of the stars at AES and I have to praise them for making the time to meet with as many of the rare disease advocacy groups as they could during the conference, which is NOT an easy task when you have expanded your scope to include literally all DEEs!

The first-ever Phase 3 trial for “all DEEs”, called DEEp OCEAN (with bexicaserin), has already started in the US and will open up sites in Europe and beyond in this coming 2025. And at least one other company has already announced that they will be following on these steps: Praxis expects to being enrollment into the EMERALD trial with relutrigine (a next-generation sodium channel blocker) in the first half of 2025.  

This means that for so many other DEEs, include very rare ones, the future of being chosen for clinical trials is also now, and arrived in 2024. What a year.

These combined “all DEE” trials are possible because they are counting seizures, and their approval will be for the treatment of seizures in DEEs. But we were just talking about how DEEs are “neurodevelopmental disorders with seizures” and we really need treatments that address the neurodevelopmental part.

I expect the next few years to get us closer to clinical trials across DEEs with treatments that can target some of the core biology of the diseases and that will need to show improvement in non-seizure outcomes. One example are genetic approaches like the one from Tevard Bio, that rescues non-sense (premature stop) mutations regardless of the gene that has them. Tevard was present at AES including at the SYNGAP1 conference. Another example are small-molecule drugs that might improve seizures and non-seizure outcomes like the KCC2 activators (some were presented at AES, other companies were in attendance but not presenting). For all those, we will need to identity clinical scales that work across different DEEs and to come up with a “core outcome set” for the non-seizure aspects of DEEs which might be a combination of some elements of the Vineland-3 and the Bayley, and perhaps other scales. Here again, we rely on those endpoint-enabling studies that the different communities are running to help us unlock the next-generation treatments for DEEs that can work across disorders and not only on their seizures.

But that is the future for the next few years. Today we celebrate a 2024 that was unprecedented in terms of tangible progress and change, with Stoke’s data adding to the news on Angelman and Rett syndrome that makes us believe in disease-modifying treatments, and with Longboard opening the door for innovation to reach everyone. What a year.

Ana Mingorance, PhD

Disclaimer: I write these texts with the parents of people with rare epilepsy syndromes in mind, so excuse also my lack of technical accuracy in parts.