Dr Melody Bacon and Dr Michael Yoong

Among epilepsies caused by a single gene mutation, the sodium channel neuronal type 1α subunit (SCN1A) gene, is the most common. SCN1A seizure disorders encompass a spectrum that ranges from simple febrile seizures, generalized epilepsy with febrile seizures plus (GEFS+) at the mild end to intractable childhood epilepsy with generalized tonic-clonic seizures (ICE-GTC) and Dravet syndrome (DS) at the severe end. The most severe associated condition is Dravet Syndrome, which is characterized by intractable epileptic seizures and a slowing of the psychomotor development in the second year of life, resulting in mild to severe intellectual disability. 

Single-gene epilepsies have a collective minimum incidence of about 1 per 2000 live births. The incidence of SCN1A related epilepsy is at least 1 per 12,200 live births, whilst the incidence of Dravet syndrome 1 in 15,500 live births. 

Pathophysiology

SCN1A encodes for the α-subunit of a neuronal sodium channel, Na_v 1.1. Pathogenic variants cause a reduction in sodium currents in gamma-aminobutyric acid (GABA)-ergic inhibitory interneurons, which leads to hyperexcitability of neuronal networks and the occurrence of seizures. These reduced sodium currents also impair Purkinje cells, causing motor disorders and contribute to the development of behavioural problems and cognitive disabilities. 

The association of SCN1A pathogenic variants with multiple syndromes can be partly explained by the consequences of different mutation types: pathogenic variants that lead to a complete loss of function of the channel are virtually always associated with severe phenotypes whereas milder disturbances in channel function usually cause milder phenotypes. However, in clinical practice, it remains difficult to fully predict the effects of all variants on channel function.

Clinical presentation

Sophie is a 5 year old who had her first febrile generalised tonic clonic seizure following her immunisations at 8 weeks. When she was 6 months old she had an episode of status epilepticus whilst unwell with bronchiolitis. She is a previously well child with normal development. Her EEG and MRI brain are normal. Her genetic test showed SCN1A variant, likely pathogenic. She was initially started on Sodium Valproate. Following this she had further episodes of status epilepticus and Stiripentol was added. She then developed myoclonic and focal seizures, mainly triggered by illness and fever. Since then, Clobazam has been added. School have recently raised some concerns with her development and she now receives 1:1 support. She has been seizure free for a few months. 

Patients with SCN1A-related disease may have a similar presentation at onset despite having different phenotypes later in life. Below is a table outlining the presentation of Dravet syndrome.

Age	Development	Seizure	EEG
Birth to 1 year old	Usually normal Some may have mild learning disability	Febrile tonic-clonic seizure may be hemiclonic Status epilepticus  Fever/illness/Immunisations precipitants  Focal, myoclonic seizures	Usually normal
1 – 5 years old	Mild – severe learning disability	Increase in seizure frequency, decrease in status epilepticus episodes Convulsive seizures – tonic clonic, clonic, hemiclonic Other seizure types: myoclonic seizures, atypical absence seizures, focal seizures, atonic seizures, non convulsive status epilepticus  Developmental of nocturnal seizures	No pathognomonic EEG findings.   Polymorphous may show background slowing, generalised spike waves, isolated or in brief bursts
5 years old to adulthood	Moderate to profound learning disability	Multiple seizure types  Refractory to AED Nocturnal seizures	No pathognomonic EEG findings.   Polymorphous may show background slowing, generalised spike waves, isolated or in brief bursts

Clinical features predicting a worse outcome includes increased frequency convulsive seizures, status epilepticus, interictal EEG abnormalities in the first year of life, early appearance of myoclonus/absence seizures/focal seizures, presence of a motor disorder (including hypotonia, ataxia, spasticity, and dyskinesia), and/or truncation SCN1A mutations.

When to test for the SCN1A? 

Clinical suspicion of SCN1A-related epilepsies / Dravet syndrome may prompt a clinician to consider sending blood for the SCN1A molecular genetic test. These are:

• < 12 months old and had prolonged febrile seizure(s)
• < 6 months old and had febrile seizure(s) 
• Normal development prior onset of seizures 
• Multiple seizures types – especially if hemiclonic 
• Vaccination associated seizure(s)
• Drug resistant epilepsy

85-90% of Dravet syndrome patients will have an SCN1A mutation identified – most are de novo (90%). However up to 15% will not! 

What does the genetic report mean? 

There are over 6,000 places for variants to occur on the SCN1A gene which is why one child/adult with Dravet Syndrome is unlikely to have exactly the same mutation as another child/adult and partly why Dravet Syndrome is a spectrum condition, affecting each individual differently. Of those with an SCN1A mutation 51% will have truncating phenotypes that leads to a loss of protein function which is associated with a more severe phenotype. 49% are associated with a missense mutation which results in a reduced protein function and that is where we get a range of phenotypes and we don’t know how the phenotype will get affected. 

What if the report shows “variant of unknown clinical significance”? 

In theses cases it is important to refer to a geneticist. A geneticist may be able to help determine whether the genetic variant is clinically significant by using other information available. These are some of the key questions that may help understand the clinical significance:

• Is it de novo or inherited from parent? If it is de novo then it is more likely to be significant than if it was inherited by a parent. 
• Does this change occur in an important part of the protein? If it does then it is likely to be more significant. 
• Is that area conserved in other species? If it is preserved in species then we know it must be important and thus likely more significant. 
• Is the change in amino acid similar or completely different? If is completely different then it is more likely to be significant. 
• Are functional data available? 
  • For example, in a laboratory setting, you can express a variant in human embryotic kidney cells. You can then measure the current through that channel which then gives you an idea how deleterious this mutation may be. This is time consuming – it can take up to a year on one single variant!
• Is it present in population databases? i.e. is there someone else with the same mutation? 
• When a variant cannot explain the phenotype, it may be due to the phenomenon of mosaicism. Mosaicism is present in 7.5% of symptomatic patients with de-novo pathogenic SCN1A variants. 

Investigations (other than genetics)

EEG studies are surprisingly normal in the first 1-2 years of life despite prolonged and frequent episodes of status epilepticus. From about the age of two years, the EEG may start to show generalized and multifocal epileptiform activity.

Imaging studies are often normal or may show non-specific abnormalities such as cerebral atrophy. Hippocampal sclerosis has been observed to a varying degree in different series ranging from 2% to 70% of cases; it probably occurs in about 30% of patients overall. Given the well-known association of hippocampal sclerosis with prolonged febrile status epilepticus, it is perhaps surprising that not all patients with Dravet syndrome have this acquired lesion.

Management

Management can be classified into treatment of Dravet syndrome, and emergency seizure management. However, beware of treatments to AVOID – sodium channel blockers: Carbamazepine, Oxcarbamazepine, Lamotrigine, Vigabatrin, Phenytoin (as daily medication). In individuals with SCN1A mutation, giving sodium channel blockers may cause more frequent and severe seizures. If given in the first year of life or there is prolonged use, sodium channel blockers are associated with worse cognitive outcomes. 

 In individuals with SCN1A mutation, giving sodium channel blocker may cause more frequent and severe seizures.

Associated features of Dravet Syndrome 

Developmental outcomes and associated behavioural difficulties

Development tends to be normal up until age 1 years age in children with Dravet syndrome however development then plateaus and most will have a degree of learning disability, ranging from mild to profound.

Clinical features predicting a worse developmental outcome included status epilepticus, interictal EEG abnormalities in the first year of and motor disorder.

Individuals with Dravet syndrome are more likely to have behavioural problems and acquired autistic features. Around 46% of individuals with Dravet syndrome will have some behavioural problems – two thirds ‘hyperactivity/inattention’ and one third ‘conduct problems’. 

Motor disorder

Functionally, a crouch gait is inefficient, requires higher energy cost, increases the stress on joints (especially knees). Many children over the age of 13 need a walker or wheelchair for longer distance mobility.

Treatment options are aimed at managing consequences of gait rather than preventative measures. Children may benefit from orthotics, physiotherapy, spasticity treatments – baclofen, L-dopa and some may even require multi-level orthopaedic surgery. 

Sleep problems

Many children with severe developmental and epileptic encephalopathies experience significant sleep disturbance, causing a major disruption to the family’s quality of life. >70% of individuals with Dravet syndrome have sleep problems. Difficulty initiating and maintaining sleep was most common, particularly in those older than 20 years. Second-most common were sleep-wake transition disorders, affecting more than 50% of those younger than 5 years. Sleep breathing disorders were a frequent problem in all age groups. 

Other co-morbidities include: growth and nutrition issues, frequent infections, and dysautonomia – of varying extent.  

Mortality

Dravet syndrome is associated with an increased mortality. It is estimated 10% of children with DS die of SUDEP before their 20^th birthday. Dravet specific mortality is 9.32 per 1000-person-year. 

Beware of the Dravet mimics

This illustration looks at Dravet mimics and it highlights other genetic disorders that present in similar way to Dravet syndrome. As you can see, the majority are caused by SCN1A, but there are other number of genes that may present like Dravet. Notably PCDH19, presents mainly in girls with clusters of febrile seizures early on.

Gene therapy

We are moving into a new era in genetic epilepsies, where we have opportunities for precision medicine – gene therapy. The aim of gene therapy is to restore the deficient SCN1A protein production. In theory this can be done by supplying a new copy of SCN1A via use of a virus: adeno-associated virus (AAV) or Lentivirus however inserting a gene into specific populations of neurons in the brain is neither possible nor proven safe and effective in humans yet. Further complicating this potential solution is the fact that the safe and effective methods of transport in gene therapy are too small to fit SCN1A, a relatively large gene. Direct editing of the mutation in SCN1A is not possible in humans yet, either.

The aim of gene therapy is to restore the deficient SCN1A protein production.

However, there are other genetic approaches that, while not altering the DNA directly, capitalize on steps in the gene-to-protein pathway cells naturally use therefore increasing more functional sodium channel proteins. Stoke Therapeutics was the first company to publicly announce their research for a treatment for Dravet syndrome. Their treatment is called “Targeted Augmentation of Nuclear Gene Output” or TANGO. The idea is you inject intrathecally antisense oligonucleotides (ASOs) that bind to pre-mRNA to up-regulate – protein production. This trial is open for recruitment and has enrolled their first patient in August 2020.

Another company, Encoded Therapeutics, is using a slightly different approach to arrive at the same solution: more healthy sodium channels. They do this not by focusing on increasing the cell’s processing efficiency, but rather by upregulating the reading of the gene in the first place. A trial is expected to start in 2021. 

What we don’t know though is what the effect of over-expression of SCN1A may have on patients? 

Take home messages 

• Consider sending blood for the SCN1A molecular genetic test in any infant who presents with features suggestive of Dravet syndrome. 
• Dravet syndrome is the most well‐recognized epilepsy phenotype associated with SCN1A.
• Avoid sodium channel blockers in individuals with SCN1A related epilepsy / Dravet syndrome. 

Dr Melody Bacon, paediatric registrar with special interest in neurology, Royal London Hospital. Consultant review by Dr Michael Yoong, paediatric neurology consultant, Royal London Hospital

Further Reading

• Dravet Syndrome UK
• Dravet Syndrome Foundation