Summary Basis of Decision for Diacomit

Clinical Pharmacology

Dravet syndrome is a rare and serious form of epilepsy that is very debilitating for affected children. The prognosis is unfavorable for development, epilepsy and survival. Currently, there is no approved pharmacological treatment available in Canada for Dravet syndrome and the resulting seizures are difficult to control with current available antiepileptic drugs.

The precise mechanism of action by which Diacomit exerts its antiepileptic effect in humans is unknown. Diacomit inhibits several cytochrome P450 (CYP) isoenzymes (CYP2C19, CYP2D6 and CYP3A4) which increase the plasma concentration of concomitantly administered antiepileptic drugs and their metabolites, notably clobazam and its active metabolite norclobazam. Diacomit differentially acts on the metabolism of clobazam and norclobazam; it increases the clobazam plasma concentration by 2-3 fold and norclobazam by 5-fold. In addition, it is believed that Diacomit has a direct effect on gamma-aminobutyric acid (GABA) receptors.

Pharmacokinetics

The following pharmacokinetic properties of Diacomit have been reported from studies in adult healthy volunteers and adult patients.

Absorption / Bioavailability

Diacomit is quickly absorbed, with a time to peak plasma concentration of about 1.5 hours. The absolute bioavailability of Diacomit is not known since an intravenous (IV) formulation is not available for testing. It is well absorbed by the oral route since the majority of an oral dose is excreted in urine.

Distribution

Diacomit binds extensively to circulating plasma proteins (about 99%).

Elimination

Systemic exposure to Diacomit increases markedly compared to dose proportionality. Plasma clearance decreases markedly at higher doses; it falls from approximately 40 L/kg/day at the dose of 600 mg/day to about 8 L/kg/day at the dose of 2,400 mg. Clearance is decreased after repeated administration of Diacomit, probably due to inhibition of the cytochrome P450 (CYP) isoenzymes responsible for its metabolism. The elimination half-life ranges from 4.5 hours to 13 hours and increases with dose.

A pediatric population pharmacokinetic study was conducted in 35 children with Dravet syndrome (median age 7.3 years) treated with Diacomit, valproate and clobazam. In this study, results showed that the elimination half-life increased from 8.5 hours (for 10 kg) to 23.5 hours (for 60 kg).

Metabolism

Diacomit is extensively metabolized, 13 different metabolites having been found in urine. The main metabolic processes are demethylenation and glucuronidation, although precise identification of the enzymes involved has not yet been identified.

On the basis of in vitro studies, the main liver CYP450 isoenzymes involved in phase 1 metabolism are considered to be CYP1A2, CYP2C19 and CYP3A4. Diacomit may therefore markedly increase the plasma concentrations of concomitantly administered drugs metabolized by these enzymes, increasing the risk of adverse events.

Excretion

Diacomit metabolites are excreted mainly via the kidney. Urinary metabolites of Diacomit accounted collectively for the majority (73%) of an oral acute dose whereas a further 13-24% was recovered in feces as unchanged drug.

Bioequivalence

Relative bioavailability between the capsules and powder for suspension formulations has been studied in healthy male volunteers after a 1,000 mg single oral administration. The powder for suspension formulation has a slightly higher maximum plasma concentration (C_max) than the capsule. Clinical supervision is therefore recommended if switching between the Diacomit capsule and powder for suspension formulations.

Special Populations and Condition

During pregnancy, antiepileptic drugs can cause fetal harm. Data from pregnancy registries indicate that infants exposed to antiepileptic drugs in utero have an increased risk for malformations. The prevalence of malformations is two- to three- times greater than the rate of approximately 3% in the general population. Therefore, Diacomit should not be used during pregnancy unless the potential benefit justifies the potential risk to the fetus. Furthermore, due to the absence of human studies on excretion in breast milk, and given that stiripentol (Diacomit) passes freely from plasma into milk in the goat, breast-feeding is not recommended for women treated with Diacomit.

The pharmacokinetics of Diacomit has not been evaluated in elderly patients. As the drug is mainly metabolized by the liver, caution should be used when considering using the drug in elderly people.

There has been no formal study of the pharmacokinetics of Diacomit in patients with hepatic or renal impairment. As the drug is mainly metabolized and excreted respectively through the liver and kidney, caution should be used when administering Diacomit to individuals with renal or hepatic impairment.

For further details, please refer to the Diacomit Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

The clinical efficacy of Diacomit, as an add-on therapy to valproate and clobazam, in children presenting with Dravet syndrome, was primarily demonstrated in two randomized, double-blind, placebo-controlled studies: STICLO France and STICLO Italy (hereafter referred to as the STICLO studies). In addition, other supportive studies conducted throughout the development of Diacomit were also reviewed as part of Health Canada's approval process.

Both STICLO studies were similar in design and compared Diacomit or placebo as an add-on therapy to antiepileptic treatment with valproate and clobazam in pediatric Dravet patients. The protocol specified that to be enrolled, patients had to present with Dravet syndrome according to the diagnostic criteria established by Dravet (1982) and included in the International League Against Epilepsy (ILAE) classification of epilepsy (ILAE, 1989), be aged 3 to 18 years (maximum weight: 60 kg), and have at least 4 generalized clonic or tonic-clonic seizures per month despite optimized therapy.

Eligible patients were initially included in a 1-month baseline period during which they continued to receive their antiepileptic treatment. During the baseline period, the daily dose of clobazam was adjusted so it was not greater than 0.5 mg/kg/day (administered in divided doses, twice daily). Following this 1-month baseline, patients were then randomly allocated to receive either Diacomit (50 mg/kg/day) or a placebo as an add-on to their current antiepileptic treatment. Patients were then treated double-blind for an 8-week period. The frequency of generalized clonic or tonic-clonic seizures during the study was then recorded by the parents or the caregivers, using a diary.

Upon completion of the double-blind comparison period, all patients (including those randomized to placebo) were offered the option of enrolling in another study, STILON, and received open-label Diacomit for 1 month, with the possibility of continuing open-label Diacomit treatment if there was therapeutic benefit.

During the double-blind period, the dose of Diacomit was fixed at 50 mg/kg/day, administered in 2 or 3 divided doses using 250 mg or 500 mg capsules. During the open-label follow-on period, the dose of Diacomit could be increased to a maximum of 100 mg/kg/day.

The primary efficacy endpoint for this study was response rate. A responder was defined as a patient who experienced a ≥50% decrease in the frequency of generalized clonic or tonic-clonic seizures during the double-blind treatment compared to baseline. Patients were also evaluated on several secondary endpoints, notably mean change from baseline in frequency of generalized clonic or tonic-clonic seizures. Note that although patients with Dravet syndrome have several different types of seizures, only generalized clonic or tonic-clonic seizures were recorded, as other seizure types can be difficult to recognize as seizures by the parents or the caregivers.

A total of 42 patients were enrolled in the STICLO France study. Twenty-two (22) patients received Diacomit and 20 patients received a placebo. One patient in the Diacomit treatment group was considered non-evaluable because of non-compliance with treatment and non-completion of the patient diary; therefore leaving 21 evaluable patients in the Diacomit group. A total of 23 patients were enrolled in the STICLO Italy study. Twelve (12) patients received Diacomit and 11 patients received placebo. In both STICLO studies, the demographic and baseline clinical characteristics were similar for the Diacomit and placebo groups.

Based on the mean number of seizures at baseline, patients enrolled in the STICLO Italy study appeared to have had more severe disease. The number of seizures during the baseline period ranged from 2 to 101 in the STICLO Italy study, and from 4 to 76 in the STICLO France study. In both STICLO studies, the types of seizures [that is (i.e.), unilateral or bilateral tonic-clonic, atypical absences, myoclonus, and others] were also comparable among treatment groups. The mean daily doses of concomitant antiepileptic drugs were similar in both treatment groups and among both STICLO studies.

In the STICLO France study, 5 patients withdrew from the study during the double-blind period [1 patient in the Diacomit group (status epilepticus) and 4 patients in the placebo group (status epilepticus: 1; no improvement: 2; drowsiness and motor deficit: 1]. In the STICLO Italy study, 3 patients withdrew during the double-blind period [1 patient in the Diacomit group (drowsiness; balance impaired) and 2 patients in the placebo group (worsening seizures; lack of improvement)].

The primary endpoint was met for both STICLO studies. Study results demonstrated that Diacomit was significantly more efficacious than placebo, as judged by the number of responders (≥50% decrease in the frequency of generalized clonic or tonic-clonic seizures). In STICLO France, 15 of 21 (71%) patients on Diacomit vs. 1 out of 20 (5%) patients on placebo were responders (p <0.00002). In STICLO Italy, 8 of 12 (67%) patients on Diacomit vs. 1 out of 11 (9%) patients on placebo were responders (p = 0.009).

Diacomit was also superior to placebo as judged by reduction in mean frequency of generalized clonic or tonic-clonic seizures in STICLO France (p <0.0001) and in STICLO Italy (p = 0.0018). Nearly half the patients (45%) in STICLO France and approximately one third of patients (27%) in STICLO Italy became free of generalized clonic or tonic-clonic seizures on Diacomit.

The dose of Diacomit was similar in both STICLO studies with the daily dose of Diacomit ranging from 43.1 mg/kg to 58.3 mg/kg. The corresponding mean [± Standard Deviation (SD)] minimal plasma concentrations (C_min) at steady state as measured at the end of the 8-week treatment period were 10.0 ± 3.6 mg/L (range 6.0-18.8 mg/L) in STICLO France and 10.2 ± 2.98 mg/L (range 5.70-14.0 mg/L) in STICLO Italy.

Despite favourable results observed for both STICLO studies, Health Canada expressed concern regarding to the study design used for both studies as the studies did not use the maximum tolerated dose of clobazam as adjunctive therapy. Treatment with clobazam was considered sub-optimal and may have exaggerated the benefit of Diacomit over placebo. Consequently, a Notice of Deficiency (NOD) was issued to the sponsor on October 18, 2011 expressing this concern. In response to the NOD, the sponsor provided an acceptable explanation for the study design.

In brief, the sponsor highlighted that the enrolled patients had been on higher doses of clobazam and many other drugs prior to enrolling in the STICLO studies. A clobazam dose adjustment was required at baseline for safety/tolerability reasons as Diacomit inhibits the metabolism of clobazam, thereby raising the plasma concentrations of both clobazam and its active metabolite norclobazam 2 to 3 times and 5 times, respectively. Patients taking clobazam + Diacomit had more side-effects than patients only on clobazam. Additionally, as Diacomit inhibits the metabolism of clobazam and norclobazam at different rates, a comparison between clobazam alone and clobazam + Diacomit plasma levels in treated patients cannot be made because the resultant plasma concentrations of clobazam and its metabolite norclobazam are not the same in both groups of patients.

Secondly, any attempt to conduct a randomized placebo-controlled study (as suggested by Health Canada) to evaluate the efficacy of Diacomit in combination with clobazam/valproate at usual doses in comparison to clobazam/valproate at maximal tolerated doses in pediatric patients with Dravet syndrome was essentially unfeasible. It would take more than 15 years to enroll a sufficient number of patients with Dravet syndrome to provide the confirmatory efficacy data required.

In light of this explanation, although increased concentrations of clobazam and norclobazam could possibly contribute to the efficacy of Diacomit, it seemed unlikely that the efficacy of Diacomit could be entirely attributed to these increases alone. If this were the case, increasing the dose of clobazam (without adding-on Diacomit) would be sufficient to obtain a significant reduction in seizure frequency. However, open-label observations showed that increasing the dose of clobazam up to 3-times the usual dose did not lead to increased efficacy. Health Canada therefore came to the conclusion that both STICLO studies did in fact demonstrate the antiepileptic efficacy of Diacomit as an add-on therapy to valproate and clobazam for the treatment of Dravet syndrome. A similar conclusion was also reached by the European Medicines Agency (EMA). Therefore, the sponsor's response to the NOD was accepted on March 2, 2012 and a Notice of Compliance was issued for Diacomit on December 21, 2012.

During the original filing of this New Drug Submission, the sponsor initially sought approval for the following indication: Diacomit is indicated as an adjunctive therapy for the management of generalized tonic-clonic and clonic seizures in patients (infants over 6 months of age; children and adults) with Dravet syndrome. Health Canada has revised this indication to clearly state that Diacomit is only indicated for use in conjunction with clobazam and valproate. Furthermore, the initially proposed statement that Diacomit was indicated in infants over 6 months of age has been removed since the two STICLO studies were conducted in children over 3 years of age. The approved indication therefore reads as follows:

Diacomit (stiripentol) is indicated for use in conjunction with clobazam and valproate as adjunctive therapy of refractory generalized tonic-clonic seizures in patients with severe myoclonic epilepsy in infancy (SMEI, Dravet syndrome) whose seizures are not adequately controlled with clobazam and valproate alone. Diacomit when used in combination with clobazam and valproate was demonstrated to be effective and safe in patients 3 years of age or older with SMEI. The clinical decision for use of Diacomit in children with SMEI less than 3 years of age needs to be made on an individual patient basis taking into consideration the potential clinical benefits and risks. In this younger group of patients, adjunctive therapy with Diacomit should only be started when the diagnosis of SMEI has been clinically confirmed. Data are limited about the use of Diacomit under 12 months of age and use in this age group should be under the close supervision of a doctor. Diacomit is not effective on other forms of epilepsy. Diacomit also has not been studied in Dravet syndrome patients over 65 years of age. The possibility of age-associated hepatic and renal function abnormalities should be considered when using Diacomit in patients over 65 years of age.

For more information, refer to the Diacomit Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Safety

The safety and tolerability of Diacomit were evaluated in 42 healthy volunteers, 272 Dravet syndrome patients, and 551 patients with other forms of epilepsies. Overall, Diacomit was demonstrated to be generally safe and well-tolerated.

Across all studies, the majority of adverse events (AEs) reported were mild to moderate in severity and could be categorized as either neurological (sleepiness/drowsiness/somnolence) or gastrointestinal (anorexia, loss of appetite, nausea, vomiting, and loss of weight) in origin. Given the frequency of gastrointestinal adverse reactions to treatment with Diacomit and valproate, the growth rate of children treated with these antiepileptic drugs should be carefully monitored. In all studies, no major differences were noted in either the type or incidence of AEs as reported by Dravet syndrome patients. For studies that enrolled both Dravet syndrome and non-Dravet syndrome patients, there was no noted difference in the AE profile between patients with Dravet syndrome and patients with other types of epilepsies.

Across all studies, 2 patients presented with a skin rash. In one case, treatment with Diacomit was maintained, and the skin rash recovered spontaneously. In the other case, treatment with Diacomit was suspended, the patient recovered, and Diacomit treatment was resumed uneventfully.

In both STICLO studies, the incidence of adverse events (AE) was much higher in the Diacomit group in comparison to the placebo group, or to the baseline period (clobazam and valproate combined treatment). Although many of the treatment-emergent events were consistent with the safety profile characterized for clobazam and valproate, it cannot be clearly determined if some of the AEs observed in both STICLO studies were caused by Diacomit directly or perhaps indirectly through increased plasma concentrations of clobazam/valproate; given a decrease in AEs was seen in some patients following dose reduction of the concomitant antiepileptic drugs (clobazam/valproate).

For both STICLO studies, 2 of 33 (6%) patients randomized to Diacomit and 2 of 31 (6%) patients randomized to placebo were withdrawn from the studies for AEs. In patients treated with Diacomit, 1 patient withdrew due to status epilepticus and 1 patient withdrew for drowsiness/balance impaired. In patients treated with placebo, 1 patient withdrew due to status epilepticus and 1 patient withdrew for drowsiness/motor deficit.

In the STICLO France study, 2 patients in the Diacomit group (status epilepticus and urticaria) and 2 patients in the placebo group (status epilepticus and repeated seizures) had Serious Adverse Events (SAE). In the STICLO Italy study, there were no reports of SAEs. In the long-term STILON study, SAEs were reported in 18 of 45 (40%) of Dravet syndrome patients. It should be noted, however, that most of these SAEs did not meet the criterion of "immediately life-threatening," and most were judged to be unrelated to Diacomit treatment.

Diacomit (stiripentol) has been shown to inhibit several cytochrome P450 (CYP) enzymes notably CYP2C19, CYP3A4, CYP1A2, CYP2C8 and CYP2D6. Stiripentol may therefore markedly increase the plasma concentrations of concomitantly administered drugs metabolized by these enzymes, increasing the risk of AEs.

Among antiepileptic drugs used to manage Dravet syndrome, stiripentol increased the plasma concentrations of clobazam by a factor of 2 or 3, and that of its metabolite (norclobazam) by a factor of 5; thereby usually resulting in a clobazam dose reduction. Furthermore, inhibition of the CYP450 isoenzymes CYP2C19 and CYP3A4 may lead to drug interactions with phenobarbital, primidone, phenytoin, carbamazepine, clobazam, diazepam, ethosuximide, and tiagabine by inhibiting the hepatic metabolism of these drugs. As a consequence, the plasma concentrations of these antiepileptic drugs may increase with a risk of overdose. Clinical monitoring of plasma concentrations of concomitantly administered anticonvulsants with Diacomit and possible dose adjustments are therefore recommended.

Co-administration with CYP3A4 substrates that have a narrow therapeutic range (dihydroergotamine, ergotamine, cyclosporine, sirolimus, tacrolimus, quinidine, and fentanyl) should be considered on an individual basis, taking into consideration the potential clinical benefits and risks. As a result, a Black Box Warning identifying this risk has been included under the serious warnings and precautions section of the Diacomit Product Monograph.

Also included as a Black Box Warning in the Diacomit Product Monograph is a statement that rare episodes of delirium and hallucinations have been reported in adult patients taking Diacomit. Therefore, patients with a past history of psychoses in the form of episodes of delirium should be monitored closely when prescribed Diacomit.

Neutropenia may be associated with the administration of Diacomit, clobazam and valproate. Blood counts should therefore be assessed prior to starting treatment with Diacomit and should be checked every 6 months or as clinically indicated.

Liver toxicity has also been observed with valproate, clobazam, and Diacomit. Liver function should also be assessed prior to starting treatment with Diacomit and should be checked every 6 months or as clinically indicated.

For more information, refer to the Diacomit Product Monograph, approved by Health Canada and available through the Drug Product Database.