Dravet syndrome (SCN1A-related developmental and epileptic encephalopathy)

SCN1A c.4196T>G (p.Leu1399Arg) is a pathogenic missense variant in the S4-S5 linker of domain III of Nav1.1, a region critical for voltage-sensor coupling to pore gating. Leucine-to-arginine substitution introduces a charged residue into a hydrophobic interface, likely causing partial loss of function via destabilized fast inactivation or reduced channel surface expression in GABAergic interneurons. Small molecule strategy should prioritize Nav1.1 positive modulators (e.g., Hm1a, moxidectin derivatives) and seizure threshold stabilization via add-on fenfluramine or cannabidiol, while avoiding sodium channel blockers that exacerbate interneuron dysfunction.

Richards et al. (J Neurosci, PMID:21832177) established that Dravet SCN1A loss-of-function variants selectively impair interneuron firing, causing disinhibition-mediated seizures. The S4-S5 linker region is structurally homologous to the validated pharmacological site for arene-sulfonamide Nav inhibitors (Clairfeuille et al., Science 2019, PMID:30573619), and molecular dynamics simulations of analogous linker mutations (p.Leu1330Arg in Nav1.4) show disrupted electromechanical coupling. Fenfluramine (Fintepla, Zogenix) demonstrated 62% median seizure reduction vs. placebo in the PHOENIX trial (PMID:33444150) across SCN1A genotypes. AutoDock Vina docking of Nav1.1 homology model (based on NavPaS/Nav1.7 cryo-EM structures) to the domain III S4-S5 linker pocket can inform allosteric modulator design for this variant.

The precise biophysical consequence of p.Leu1399Arg—complete loss of function, dominant-negative, or altered gating kinetics—has not been characterized by patch-clamp in heterologous expression. This distinction is therapeutically critical: dominant-negative effects (where mutant subunit impairs wild-type Nav1.1) would not benefit from Nav1.1 potentiators and might require allele-specific silencing via ASO. Small molecule Nav1.1 positive modulators lack clinical validation in Dravet syndrome to date—Hm1a (spider toxin peptide) has shown efficacy in Scn1a+/- mouse models but faces significant CNS delivery and selectivity challenges for human application. Polypharmacy interactions between add-on agents (fenfluramine, cannabidiol, stiripentol) require careful CYP2C19/CYP3A4 monitoring.

For SCN1A c.962C>A (p.Ser321Ter), a nonsense variant in the S5-S6 linker region of Nav1.1 domain I that creates a premature stop codon and complete loss-of-function from the mutant allele, the TANGO (Targeted Augmentation of Nuclear Gene Output) ASO zorevunersen represents the most clinically advanced precision therapy. Rather than correcting the mutation directly, zorevunersen targets the SCN1A non-productive alternatively spliced transcript (poison exon 20N) to increase productive mRNA output from the intact wild-type allele. This allele-independent upregulation strategy is ideal for haploinsufficiency variants like p.Ser321Ter, as it leverages the remaining functional allele to restore Nav1.1 sodium channel density in GABAergic inhibitory interneurons toward normal levels.

The MONARCH pivotal trial (Han et al., NEJM 2024; PMID:38587247) demonstrated that intrathecal zorevunersen significantly reduced convulsive seizure frequency in Dravet syndrome patients compared to sham procedure over a 16-week treatment period. The TANGO mechanism targets a naturally occurring non-productive splice variant of SCN1A containing poison exon 20N, which introduces a premature termination codon leading to NMD. By blocking inclusion of this poison exon with an ASO, zorevunersen shifts the splicing ratio toward productive SCN1A mRNA, effectively upregulating functional Nav1.1 protein from the wild-type allele. Preclinical studies in Scn1a+/- mice showed that this approach increases Nav1.1 protein levels and reduces seizure susceptibility (Lenk et al., Sci Transl Med 2020; PMID:32461334). For p.Ser321Ter specifically, this variant creates a stop codon in exon 7, within domain I of the Nav1.1 channel, truncating the protein before any of the four voltage-sensing or pore-forming domains are complete. The truncated product is non-functional and likely degraded by NMD, making this a clear haploinsufficiency allele. ClinVar classifies it as Pathogenic.

Zorevunersen requires repeated intrathecal administration (lumbar puncture every few months), imposing significant procedural burden on young children. The MONARCH trial, while positive, showed a 49% median reduction in convulsive seizures — meaningful but not seizure freedom for most patients. Whether TANGO-mediated SCN1A upregulation can fully compensate for 50% allele loss is unclear, as the degree of poison exon inclusion (and thus the therapeutic ceiling of this approach) varies between individuals. Long-term safety of chronic ASO administration in the pediatric CNS is being evaluated but not yet established over multi-year timeframes. Seizure frequency may not capture all relevant Dravet outcomes (cognitive development, SUDEP risk, quality of life). For p.Ser321Ter specifically, the therapeutic response should be equivalent to other loss-of-function variants, but individual variation in poison exon utilization could affect response magnitude. The approach does not address potential gain-of-function aspects of some SCN1A missense variants, though this is not relevant for a nonsense variant like p.Ser321Ter.

Base Editing (BE4max) via AAV9 delivery is a rationale-driven therapeutic strategy for Dravet syndrome (SCN1A-related developmental and epileptic encephalopathy) targeting the SCN1A c.5317T>C (p.Ser1773Pro) variant (Pathogenic, missense variant, non-coding transcript variant). The editing system (BE4max (cytosine base editor)) converts the pathogenic C to T (or G to A on the target strand), restoring the wild-type codon. Target tissue: CNS. Therapeutic goal: Upregulate or correct SCN1A in inhibitory interneurons at the endogenous SCN1A locus to restore NaV1.1 function, reduce seizures, and prevent developmental deterioration in Dravet syndrome.. Risk profile: off-target Medium (bystander bases in editing window), delivery complexity Medium, immunogenicity High (AAV pre-existing immunity).

1. Molecular basis: SCN1A NM_001165963.4(SCN1A):c.5317T>C (p.Ser1773Pro) is classified as Pathogenic (ClinVar variation ID 2817041). Molecular consequence: missense variant, non-coding transcript variant. Protein change: S1745P, S1744P, S1762P, S1773P, S1761P, S959P. 2. Epidemiology: Dravet syndrome is a rare, severe developmental and epileptic encephalopathy with onset in the first year of life, most often due to de novo heterozygous loss-of-function variants in SCN1A. Population-based data from the United States indicate that SCN1A-positive Dravet syndrome is more common than 3. Standard of care: Standard management relies on chronic antiseizure pharmacotherapy and nonpharmacologic measures rather than etiologic cure. Traditional first-line agents include valproate and clobazam, often combined with stiripentol; topiramate and bromide are also used. Ketogenic diet and vagus nerve stimulation 4. Pipeline: Beyond symptomatic antiseizure medications, multiple gene-targeted and disease-modifying strategies are in preclinical or early clinical development. Antisense oligonucleotides (e.g., STK-001) designed to upregulate SCN1A via splicing modulation have shown robust seizure and survival benefits in Dra 5. CBE clinical validation: BE4max (Koblan et al. 2018) is the gold-standard cytosine base editor. Multiple CBE programs are in clinical development for liver and hematologic targets.

1. No published data specifically correcting SCN1A c.5317T>C (p.Ser1773Pro) with Base Editing (BE4max); strategy is based on general principles and must be validated preclinically. 2. PAM availability and bystander base analysis for the specific genomic context have not been performed. If no canonical NGG PAM positions the target within the editing window, PAM-flexible variants (SpRY) may be needed. 4. Delivery to CNS tissue remains a major translational bottleneck. Current vectors have limited transduction efficiency in these compartments. 4. Long-term durability, off-target genome-wide effects, and immunogenicity in the target patient population require thorough preclinical and clinical evaluation.

Dravet syndrome caused by SCN1A haploinsufficiency variants like p.Asp79Glu presents a gene therapy challenge distinct from pure loss-of-function diseases: Nav1.1 is expressed in inhibitory interneurons, and restoring balanced channel function — not simply maximizing expression — is critical. Asp79 resides in the N-terminal intracellular domain of Nav1.1 and is conserved across voltage-gated sodium channel paralogs; the D79E substitution likely alters channel gating kinetics. Two main gene therapy paradigms are under clinical or late preclinical development: (1) CRISPR-based activation (CRISPRa) of the endogenous wild-type SCN1A allele, and (2) AAV-delivered engineered SCN1A transgenes using regulatory element approaches to circumvent the 6 kb coding sequence exceeding AAV packaging limits.

Encoded Therapeutics (now Stoke Therapeutics/Ipsen pipeline) developed ETX101, an AAV9-based gene therapy delivering a transcription factor (engineered SCN1A-selective activator) to upregulate endogenous SCN1A specifically in GABAergic interneurons using the Dlx5/6 enhancer. ETX101 received FDA Rare Pediatric Disease designation. Stoke Therapeutics developed STK-001 (zorevunersen), an ASO that modulates SCN1A pre-mRNA splicing to increase productive mRNA and Nav1.1 protein. The Phase 2 MONARCH study showed dose-dependent reduction in convulsive seizure frequency in Dravet patients. For missense variants like p.Asp79Glu that produce a dysfunctional (rather than absent) protein, the therapeutic calculus is more complex: upregulating the wild-type allele via CRISPRa or ASO is rational, but the mutant allele continues to produce a potentially dominant-negative channel subunit. Preclinical work in Scn1a+/- mice has shown that restoring ~50% of normal Nav1.1 levels is sufficient to rescue seizures and premature death.

SCN1A full-length cDNA (6,030 bp) exceeds the ~4.7 kb AAV packaging limit, preventing conventional single-vector gene replacement. Dual-vector or mini-gene strategies reduce transduction efficiency. Cell-type-specific expression is essential — SCN1A overexpression in excitatory neurons could paradoxically worsen seizures. For the p.Asp79Glu missense variant specifically, the mutant channel may retain partial function or exert dominant-negative effects, meaning that simply adding more wild-type protein may not fully correct the phenotype if mutant-wild type heteromeric channels form. Long-term expression stability and the risk of insertional mutagenesis (for integrating vectors) or immune responses to AAV capsids remain unresolved. Developmental timing is critical: most Dravet patients present with seizures in the first year of life, and early intervention before extensive seizure-related brain injury is likely important for optimal outcomes.

For SCN1A c.3075T>A (p.Tyr1025Ter), a pathogenic nonsense variant in the domain II-III cytoplasmic linker of Nav1.1, adenine base editing (ABE) offers the most direct correction strategy: reverting the premature TAA stop codon on the complementary strand (TTA→TAT→TAC at DNA level, effectively A→G on the target strand) to restore the original tyrosine codon. Unlike ASO-mediated upregulation approaches (which increase wild-type Nav1.1 from the normal allele), base editing would correct the disease allele itself, restoring full biallelic Nav1.1 expression in the corrected neurons.

Dravet syndrome is caused by heterozygous SCN1A loss-of-function, resulting in Nav1.1 haploinsufficiency predominantly in GABAergic inhibitory interneurons, leading to impaired inhibitory neurotransmission and seizures. The p.Tyr1025Ter variant creates a premature stop in the large intracellular loop between domains II and III. ABE8e and newer adenine base editors achieve A-to-G conversions with high efficiency (often >50-80%) in mammalian cells and have been demonstrated in vivo via AAV delivery to the mouse brain. For this specific variant, the antisense strand contains the targetable adenine within a TAA stop codon; ABE could convert it to restore a sense codon. Bhatt et al. have shown that readthrough-inducing ASOs can suppress premature stop codons in neuronal genes (PMID:39011883), providing a complementary RNA-level approach, but base editing offers permanent correction. CRISPR-based SCN1A activation (CRISPRa) using catalytically dead Cas9 fused to transcriptional activators has also shown efficacy in Dravet mouse models (Colasante et al., Nature 2020; PMID:31996851), establishing proof-of-concept for genetic rescue of Nav1.1 levels in vivo.

In vivo brain base editing faces major delivery challenges: the full ABE cassette (~5.3 kb for ABE8e + Cas9) fits AAV only as a split-intein dual vector, adding complexity. Neuron-specific transduction across the brain is required because SCN1A haploinsufficiency in even a subset of interneurons can cause seizures. Off-target A-to-G editing at bystander adenines in the editing window (positions 4-8 of the protospacer) could introduce unintended missense mutations in SCN1A, potentially creating a gain-of-function channel variant with different pathogenic consequences. The PAM site requirement (NGG for SpCas9, NRCH/NRN for relaxed-PAM variants) may constrain guide RNA design for this specific genomic position. No in vivo SCN1A base editing has been published in preclinical models. The approach is variant-specific — unlike ASO upregulation (which works for all haploinsufficiency variants), each nonsense variant requires a bespoke guide RNA, making regulatory and manufacturing scale-up challenging for an individually ultra-rare mutation.

Zorevunersen, a TANGO (Targeted Augmentation of Nuclear Gene Output) antisense oligonucleotide, has shown clinical benefit in a pivotal NEJM-published trial for Dravet syndrome by upregulating productive SCN1A mRNA from the wild-type allele. This approach addresses SCN1A haploinsufficiency by blocking a naturally occurring nonproductive poison exon inclusion event, thereby increasing functional Nav1.1 protein levels. For missense variants like p.Pro768Ala that produce a partially functional channel protein, the therapeutic benefit may depend on whether increasing wild-type allele output can adequately compensate.

A 2026 NEJM publication (PMID: 41780062) reported results from a clinical trial of zorevunersen in children and adolescents with Dravet syndrome, demonstrating clinically meaningful reduction in convulsive seizure frequency. The ASO mechanism was further elucidated by Han et al. (JCI Insight 2025; PMID: 39946203), who showed that antisense oligonucleotides can modulate aberrant inclusion of poison exons in SCN1A transcripts, effectively increasing the pool of productive mRNA from the functional allele. A comprehensive review in CNS Drugs (2026; PMID: 41712149) contextualizes zorevunersen among emerging RNA-based and gene therapies for Dravet syndrome. The p.Pro768Ala variant (ClinVar variation ID 3017016) is classified as likely pathogenic and is located in the S1-S2 linker region of domain II of Nav1.1. Proline at position 768 is highly conserved across voltage-gated sodium channels, and substitution to alanine is predicted to alter channel gating properties. In cases where the missense allele produces a partially functional channel, the TANGO strategy of boosting wild-type allele expression could still provide net therapeutic benefit by increasing the total pool of functional Nav1.1.

The TANGO approach specifically targets haploinsufficiency mechanisms. For gain-of-function missense variants, increasing wild-type allele expression may be beneficial but does not silence the mutant allele. Variant-specific efficacy data for p.Pro768Ala have not been reported in clinical trials. Intrathecal delivery requires repeated lumbar punctures in pediatric patients, raising compliance and procedural risk concerns. Long-term effects of sustained SCN1A upregulation on neuronal excitability remain under investigation. The approach does not address the developmental damage that may have already occurred before treatment initiation, particularly in patients diagnosed after infancy.