Successful treatment of pediatric patient with Dravet syndrome with cenobamate: Case report

Introduction

Dravet syndrome (DS), also known as severe myoclonic epilepsy of infancy, is an epileptic encephalopathy characterized by severe epilepsy accompanied by impaired psychomotor and neurologic development. Onset is in the first year of life in apparently healthy infants. Epilepsy in DS is characterized by the presence of febrile and afebrile clonic or tonic-clonic seizures, as well as myoclonic, atypical absences, and partial seizures later in the course of the disease. Notably, seizures tend to be highly pharmacoresistant. ¹ SUDEP risk is increased in patients with DS, estimated to be 9.3/1000 person-years, almost double the rate of adults with drug-resistant epilepsy, estimated to be 5.1/1000 in one study. ²

The majority of cases are caused by de-novo variants in the SCN1A gene; however, other genes such as GABRG2, PCDH19, SCN1B, and CHD2 may be implicated. ³ Mutations in the SCN1A gene lead to reduced sodium currents in inhibitory interneurons, thus enhancing the excitability of their downstream targets, such as pyramidal neurons. ⁴ Valproic acid (VPA), fenfluramine (FFA), stiripentol (STP), clobazam (CLB), cannabidiol (CBD), and topiramate (TPM), as well as the ketogenic diet are used for treating seizures in DS. Notably, archetypal sodium channel blockers such as carbamazepine (CBZ), oxcarbazepine, lamotrigine (LTG), and phenytoin are to be avoided. ⁵ For example, LTG worsened seizures in up to 80% of patients. ⁶ Even with proper treatment, the outcome for the vast majority of DS patients is still poor, with multiple drug-resistant seizures, moderate-to-severe intellectual disability, behavioral issues, and motor abnormalities. ⁷

Cenobamate (CNB) is a new anti-seizure medication (ASM) that is FDA-approved for focal-onset seizures in adults. ⁸ The off-label use in pediatric patients has been described.^9,10 Moreover, the successful use was recently described in four adult patients with DS. ¹¹ To our knowledge, its use in pediatric DS has not been reported. Here, we describe the successful use of CNB in a pediatric patient with typical DS caused by de-novo mutation in SCN1A with seizures refractory to numerous ASMs.

Methods

In this case report, seizure and treatment-specific data were extracted from electronic medical records available from our institution and through outside hospital records using Care Everywhere.

Case report

The patient was a 9-year-old girl with DS caused by a de-novo variant c.2415+4A->G in the SCN1A gene detected at 3 years of age via an epilepsy gene panel. This variant has not been observed in large population cohorts. In silico prediction indicated that it might damage the natural splice donor site in intron 13 and lead to abnormal gene splicing according to the genetic test report.

She was born at 39 weeks via an emergency C-section due to recurrent decelerations. She had meconium aspiration that led to an otherwise uncomplicated 7-day NICU stay. Her early development was unremarkable (rolling over at 4 months of age, sitting independently at 6 months of age, walking at 12 months of age, and speaking six words at 15 months of age). Her past medical history was notable for asthma. The family history was noncontributory. Her physical exam was unremarkable.

Seizure onset was at 4 months of age. The initial seizures were provoked by fever, were prolonged, and were hemiclonic and generalized tonic-clonic (GTC), lasting up to 20 min. She was initially treated with levetiracetam but continued to have prolonged seizures provoked by fever despite reaching a dose of 80 mg/kg/day. A routine electroencephalogram at 1 year of age was read as normal. A magnetic resonance imaging of the brain with and without contrast at 3 years of age did not demonstrate a structural cause for the patient’s seizures.

As she continued to have prolonged seizures including episodes of status epilepticus refractory to home rectal diazepam, additional medications were trialed. LTG and CBZ were trialed prior to the diagnosis of DS. Maximum doses are not available, and no improvement occurred; caregivers are uncertain if seizures increased. TPM was dosed up to 11.5 mg/kg/day and was ineffective. VPA was efficacious at 34 mg/kg/day but was discontinued due to pancreatitis.

At the time of her initial evaluation at our center, the patient was 9 years of age and continued to have the following seizure types: (1) episodes of activity arrest accompanied by rhythmic blinking lasting up to 8 min, likely atypical absence versus focal with impaired awareness, occurring once every few months, and (2) GTC seizures occurring twice weekly and necessitating the use of rescue diazepam approximately once weekly to every other week. She was on FFA (0.6 mg/kg/day), CBD (18 mg/kg/day), and CLB (0.3 mg/kg/day, max dose 1.8 mg/kg/day, decreased due to behavioral issues with no perceived worsening in seizure control).

Given the potential for inhibition of non-inactivating persistent component of sodium current ¹² and a recent report of the successful use of CNB in adult patients with DS, ¹¹ we initiated CNB at 9 years of age, starting at 6.25 mg daily (0.23 mg/kg/day) and increasing in 12.5 mg increments every 2 weeks. At a dose of 25 mg daily (0.9 mg/kg/day), she was seizure-free for 4 weeks. Further incremental increases were made for seizure recurrences which occurred approximately every 4 weeks, and at the time of her most recent follow-up in January of 2025, she was maintained on CBM at 75 mg/day. Her current GTC seizure frequency of approximately 1 per 4 weeks represents an 87.5% reduction compared to her baseline. Notably, the duration of her GTC seizures has shortened and rescue diazepam has not been used since CNB initiation. No side effects occurred, and the existing anti-seizure medication regimen did not require adjustment. The patient has been taking CNB for 6 months at the time of the last follow-up.

Discussion

We describe the successful use of a relatively low dose of CNB in a pediatric patient with typical DS due to a mutation in the SCN1A gene, refractory to multiple (eight) ASMs, including those that are typically and most successfully used in DS (VPA, CLB, FFA). CNB initiation resulted in an approximately 87.5% GTC seizure reduction at the time of the last follow-up.

CNB is a tetrazole carbamate derivative effective against focal-onset seizures. Its effects have been shown in maximal electroshock models, 6-Hz tests, chemoconvulsant seizure models, and hippocampal kindling models. The mechanism of action is believed to involve the blockade of persistent sodium channel currents and positive allosteric modulation to mediate both phasic and tonic GABA_a currents. Interestingly, the enhancement of tonic GABA_a current is similar to the action of barbiturates and may occur through an action at a site independent of the benzodiazepine binding site on GABA_a receptors. ¹³ The effectiveness of CNB in treating a patient with DS may indicate that: (1) CNB has a greater effect on the sodium channels responsible for excitation (SCN2A, SCN3A, SCN8A) rather than inhibition (SCN1A); (2) its effect on GABA_a modulation outweighs its inhibitory effect on SCN1A; (3) a combination of the two; or (4) another unique mechanism of action different from typical anti-seizure medications used in this patient.

From a practical standpoint, the initiation of CNB remains challenging in pediatric patients given its off-label use with the resultant difficulty in obtaining insurance authorization. Additional practical considerations include potential interactions with other GABA_a-positive allosteric modulators such as CLB and sodium channel blockers. The latter interaction is not significant in patients with DS as typical sodium channel blockers are contraindicated. Interaction with CLB is important in patients with DS, as CLB is a mainstay of therapy for this condition. The interaction is likely due to both acting as positive allosteric modulators of GABA_a and through CNB inhibition of CYP2C19, leading to an increase in the plasma concentration of the active metabolite of CLB. ¹⁴ In our patient, this interaction was not clinically significant as she was already on a low dose of CLB and was possibly acclimated to higher concentrations of norclobazam due to concomitant CBD use. In our experience with CNB use in pediatric patients, a reduction of daily CLB or complete discontinuation may be necessary.

International consensus on diagnosis and management of DS suggests the use of VPA as the first line, FFA, STP, or CLB as the second line, CBD pharmaceutical grade as the third line, and TPM and ketogenic diet as the fourth line medications in the management of DS. ⁵ Recent studies show data on efficacies of some of these medications. For example, FFA at 0.7 mg/kg/day led to a 62.3% median reduction in seizure frequency when compared to placebo ¹⁵ ; percent change from baseline in seizure frequency was higher on STP when added to CLB or VPA (−69%) than on placebo (+7%), ¹⁶ percent decrease in seizure frequency was 22.8 in add-on CBD group when compared to placebo group. ¹⁷

The limitations of the current report include the number of patients described, as well as a relatively short follow-up period. This did not allow us to judge motor and cognitive improvements with certainty. This important aspect should be explored in future works. In addition, no functional studies have been done to classify the variant according to gain or loss of function, however, the lack of response to sodium channel blockers and a clinical course consistent with DS may suggest a loss of function variant in this patient.

Given the relative lack of pharmacologic options available for the treatment of refractory epilepsy in general and DS in particular, as well as the high risk of SUDEP in DS, we felt that any available evidence should be shared as soon as possible. However, additional data, including studies with a higher number of pediatric and adult patients with DS, longer follow-up, and nuances of patient selection (e.g., by the lower limit of age or certain variants that are likely to respond better to CNB), are needed.

Conclusion

CNB was an effective add-on medication in controlling GTC seizures in this pediatric patient with DS. No side effects or interactions with current anti-seizure medications were reported. Further studies on the generalizability of this finding in this age group and by mutation subtype are needed.