Adults with a history of possible Dravet syndrome: An illustration of the importance of analysis of the SCN1A gene

Summary

Most patients with Dravet syndrome have de novo mutations in the neuronal voltage-gated sodium channel type 1 (SCN1A) gene. We report on two unrelated fathers with severe childhood epilepsy compatible with a possible diagnosis of Dravet syndrome, who both have a child with Dravet syndrome. Analysis of the SCN1A gene revealed a pathogenic mutation in both children. One father exhibited somatic mosaicism for the mutation detected in his son. A relatively favorable cognitive outcome in patients with Dravet syndrome patients may be explained by somatic mosaicism for the SCN1A mutation in brain tissue. A mild form of Dravet syndrome in adult patients is associated with a high recurrence risk and possibly a more severe epilepsy phenotype in their offspring.

Dravet syndrome, also known as severe myoclonic epilepsy of infancy (SMEI) is an intractable epilepsy, starting in the first year of life with febrile and afebrile hemiclonic or generalized tonic–clonic seizures, followed by myoclonic seizures, absences, and partial seizures (Dravet et al., 2005). Following the first year of life, developmental delay becomes apparent, leading in a majority of cases to mental retardation. Patients lacking one of the core features of the disease can be diagnosed with borderline Dravet syndrome (Harkin et al., 2007).

Heterozygous mutations in the neuronal voltage-gated sodium channel type 1 (SCN1A) gene are detected in 70–80% of patients with Dravet syndrome; 89–95% of mutations are de novo. Five percent to 11% of patients with Dravet have inherited an SCN1A mutation from one of their parents (Mulley et al., 2005; Depienne et al., 2010). In general, these parents have either no epilepsy phenotype or a milder form of the disease, which includes febrile seizures. We report on two father–child pairs with a severe epilepsy phenotype compatible with a diagnosis of (borderline) Dravet syndrome, caused by a SCN1A mutation.

Case Reports

Family 1

The proband was a 3-year-old boy. At the age of 5 months he had a generalized tonic–clonic seizure (GTCS), provoked by fever, that lasted for >30 min. Hereafter, he had weekly unprovoked generalized seizures or alternating hemiconvulsions lasting 20–120 min, despite treatment with sodium valproate (VPA). Seizures were not provoked by vaccination. In the subsequent months, he also developed absences, tonic-, atonic-, myoclonic-, and complex partial seizures, and continued to have frequent seizures under the treatment of VPA, topiramate (TPM), and stiripentol (STP). At the age of 3 years and 7 months, his level of development was that of a child of 17 months Bayley Scales of Infant Development—Second Edition (BSID-II) and he could not yet speak. His behavior was hyperactive and autistic. Up to age 4, interictal electroencephalography (EEG) showed background slowing but no epileptic discharges. A magnetic resonance imaging (MRI) scan of the brain showed a small, left temporal arachnoidal cyst.

The proband’s father had his first seizure, provoked by fever, at the age of 8 months. He developed refractory generalized epilepsy with daily seizures that, at the age of 3 years and 9 months, necessitated admission to an institution. No medical data on seizure semiology were available. EEG showed generalized epileptiform abnormalities. At the age of 6 years, cognitive impairment and aggressive outbursts were noted. Anticonvulsive medication was withdrawn at the age of 11 years and 6 months. His last seizure occurred at the age of 14 years. Around the age of 12, he learned to read and write and was subsequently educated to lower technical school level. The father was 41 years old and employed, when his son was diagnosed with Dravet syndrome. The family history of first- and second-degree relatives was negative for febrile or afebrile seizures.

Family 2

The proband is a 3-year-old girl. She had her first generalized tonic–clonic seizure (GTCS) with fever, not provoked by vaccination, at the age of 10 months. The following day she had an afebrile seizure. An interictal EEG was normal. She was seizure free and without treatment until the age of 17 months. At age 17 months she had five febrile and afebrile generalized seizures within 5 days. At that time, the EEG showed right frontal, sharp spike-wave complexes. Carbamazepine (CBZ) was started. Hereafter, she had weekly unprovoked tonic–clonic seizures, gradually increasing in severity and frequency. Around the age of 2 years, she developed myoclonias, myoclonic-astatic seizures, and atypical absences. At 3 years of age, she had tonic–clonic seizures every 2 weeks while under the treatment of VPA, levetiracetam (LEV), and clonazepam (CZP). Subsequent EEGs showed multifocal (poly)spike-wave complexes. An MRI scan of the brain was normal. Her psychomotor development slowed when the seizures started to occur weekly. At the age of 2.5 years, she had a developmental level of an 18-month-old.

The proband’s father had his first seizure, a hemiconvulsion provoked by fever, at the age of 5 months. Hereafter he had generalized tonic–clonic seizures with a duration of around 2 min, and secondary generalized tonic–clonic seizures that started with a scream, deviation of the eyes, and turning his head to the left. He was treated with phenobarbital (PB) and CBZ. The seizure frequency increased after CBZ was withdrawn at the age of 18 months. Myoclonic and atonic seizures appeared and his behavior was hyperactive. EEG showed focal and generalized spike-wave complexes. A cerebral computed tomography (CT) scan and metabolic screening were normal. PB was discontinued; VPA was started and his seizure frequency decreased to one generalized tonic–clonic seizure a year. No information was available on his psychomotor development, but, with additional support, he had followed secondary school education. During adolescence, he was diagnosed with an autism spectrum disorder. He worked in a sheltered workplace. At the age of 22, his seizure frequency increased. Clobazam and lamotrigine were introduced but resulted in a further increase in seizure frequency. At the age of 25, he had two psychotic episodes and during a hospital admission he drowned in the bath during a seizure. His family history was negative for febrile seizures and epilepsy.

Molecular Analysis and Results

Sequence analysis of the SCN1A gene in the first proband revealed a frameshift mutation (c.1537delG; p.Glu513fs). The mutation was also detected in DNA isolated from lymphocytes, hair, saliva, and urine from his father. However, the level of mutant sequence was lower than the level of wild type sequence (Fig. 1), indicating somatic mosaicism for the mutation in the father. The father’s parents were not available for DNA analysis.

Figure 1.   Sequence electropherography showing paternal mosaicism for the c.1537delG SCN1A mutation in family 1 with Dravet syndrome. The arrows indicate the position of the frameshift mutation from DNA derived from the first proband’s lymphocytes and a control sample, and from the father’s DNA derived from different tissue samples. In the father, the level of mutant sequence differs between the tissues and is lower than the level of wild-type sequence as compared with this ratio in the proband’s lymphocytes. This indicates somatic mosaicism due to a postzygotic, de novo mutation in the father (Vadlamudi et al., 2010).

Sequence analysis of the SCN1A gene of the second proband showed a missense mutation (c.2837G>A, p.Arg946His). There was no DNA from the deceased father and the paternal grandparents were not available for analysis.

Discussion

Both the fathers had severe disabling epilepsy at a young age, but a mild form of Dravet syndrome was only considered after Dravet syndrome was diagnosed in their children and confirmed by detection of an SCN1A mutation. The limited information on the childhood epilepsy of the father of the first proband precludes a definite diagnosis, but may be compatible with a diagnosis of borderline Dravet syndrome. The father of the second proband fulfilled the clinical criteria for Dravet syndrome at the age of 18 months. His epilepsy improved on valproate monotherapy. Both fathers had a relatively good cognitive outcome, as reported earlier, for a minority of patients with Dravet syndrome (Buoni et al., 2006; Jansen et al., 2006). Two recent articles reported on adults who themselves had a history of severe epilepsy, who transmitted an SCN1A mutation to their offspring with Dravet syndrome (Depienne et al., 2010; Suls et al., 2010).

The father of our first proband showed somatic mosaicism for the familial SCN1A mutation, which is a likely explanation for his favorable outcome. Somatic mosaicism was reported to be present in one of the parents of at least 7% of Dravet syndrome patients. Parents with a higher level of mosaicism had a more severe epilepsy (Depienne et al., 2010). Because levels of mosaicism in the brain do not necessarily correlate with levels of mosaicism in other body tissues, analysis of DNA samples isolated from lymphocytes may give a false impression of a heterozygous germline mutation being present. The amount and distribution of the SCN1A mutation in the brain may be a major determinant for the severity of the phenotype in individuals with mosaic mutations. The phenotypic variability in Dravet syndrome is probably also influenced by other, as yet unidentified, genetic and environmental factors, for example, the adequacy of the treatment received.

In children with Dravet syndrome, treatment with sodium channel blockers should be avoided (Ceulemans et al., 2004) because of the potential for seizure aggravation, as occurred in our second proband. This may also hold true for adult patients: the father of the second proband, whose seizures in childhood were well controlled on VPA monotherapy, developed refractory seizures again after the introduction of lamotrigine in adulthood.

Neurologists who see adult patients with a history of severe epilepsy at a young age should be aware of the possibility of milder forms of Dravet syndrome and should consider analysis of the SCN1A gene. If the patient has a mutation, the risk of recurrence for Dravet syndrome in their offspring is 50%. Because the phenotype severity in patients with de novo SCN1A mutations may be reduced due to somatic mosaicism in brain tissue, the symptoms of the disease in offspring may be more severe and hard to predict.

Acknowledgments

We thank both families for participating in this study and Jackie Senior for checking the manuscript. N. Verbeek was supported by the NutsOhra Fund (grant no.0801-064).

Disclosure

We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. None of the authors has any conflict of interest to disclose.