Persistent Hypersomnolence Following Clobazam in a Child With Epilepsy and Undiagnosed CYP2C19 Polymorphism

Abstract

We describe an 11-year-old female who presented with severe hypersomnolence after receiving 1 week of modest doses of clobazam (CLB). In reviewing the above case, we considered that the hypersomnolence could be related to a pharmacodynamic, pharmacokinetic, or pharmacogenomic issue associated with CLB or to a combination of these factors. Although serum concentrations of CLB and its active metabolite are sensitive to factors that affect cytochrome-dependent metabolism, drug-drug interactions were omitted as a cause of the hypersomnolence. Subsequent DNA analysis of the cytochrome P450 2C19 gene revealed the patient as *2/*2 genotype with poor metabolizer enzyme activity. Because genetic testing of all patients treated with CLB is currently not practical, CLB dose/concentration ratios and pharmacokinetic drug-drug interaction impact models may be indicated. Genetic testing should be considered when an adverse effect suggests the possibility of a polymorphism important to drug metabolism.

Introduction

Clobazam (CLB), a 1,5-benzodiazepine, was synthesized in 1966 with hopes of improving the efficacy and decreasing the sedative effects associated with the 1,4-benzodiazepines. Clobazam was approved in the United States in 2011 for use as an adjunctive therapy for Lennox-Gastaut syndrome in patients ≥2 years of age.¹ Worldwide, it is also used as an adjunctive therapy for refractory epilepsy and is commonly used in other epileptic syndromes.² Individuals with multiple intractable seizure types may require CLB in combination with many different antiepileptic drugs (AEDs), which places these patients at an increased risk for drug-drug interactions. Because CLB is a substrate for cytochrome P450 (CYP)2C19, the influence of genetic polymorphisms on the metabolism of CLB and subsequent efficacy and toxicity must also be considered. We describe an 11-year-old Caucasian female who was receiving multiple AEDs for intractable, partial seizures. She presented with recurrent hypersomnolence after receiving 1 week of CLB therapy. Subsequently, she was found to have a *2/*2 genotype and poor CYP2C19 metabolizer enzyme activity, which significantly contributed to the adverse effect.

Case

An 11-year-old, 30-kg Caucasian female was admitted to our hospital for evaluation of intractable, symptomatic partial seizures and a recent episode of hypersomnolence. Her past medical history was significant for anoxic brain injury due to a near-drowning 8 years prior to admission that resulted in symptomatic epilepsy, spastic quadriplegia (managed with a baclofen pump), mental impairment, gastrostomy tube, and tracheostomy dependence.

Three months prior to arrival at our institution, the child experienced a tonic-clonic seizure that stopped within 2 minutes of onset following in-home administration of oral lorazepam 2 mg (0.07 mg/kg). Because of prolonged postictal sedation, she was taken to an outside hospital. At this time her seizures were chronically managed with oral levetiracetam 1200 mg twice a day (80 mg/kg/day) and lorazepam 1 mg at bedtime (0.03 mg/kg/dose). The mother and in-home nurse deny any missed doses of AEDs or the presence of an acute illness. At the time of admission, routine laboratory studies were unremarkable and urine was negative for signs of infection. Video electroencephalogram (EEG) captured brief tonic seizures that were characterized by upward eye deviation and, in one event, slight raising of the left hand. Subclinical status epilepticus was ruled out as a cause of the postictal sedation. Routine home lorazepam was discontinued, and she was started on oral CLB (Onfi, Lundbeck, Deerfield, IL) at 5 mg twice a day (0.3 mg/kg/day) for improved seizure control, and a temporary bridge of oral clonazepam 0.02 mg/kg/day divided three times a day was initiated.

The patient's seizures responded, but her dose of CLB was progressively increased in order to titrate her onto a normal maintenance dose. Hence, oral CLB was increased the following day to 7.5 mg twice a day (0.5 mg/kg/day). Two days later the dose was again increased to 10 mg twice a day (0.7 mg/kg/day), and the patient was discharged. Home medications were all given orally and included CLB 10 mg twice a day (0.7 mg/kg/day), levetiracetam 1200 mg twice a day (80 mg/kg/day), lorazepam 1 mg at bedtime (0.03 mg/kg/day) and 2 mg every 4–6 hours as needed for anxiety, and clonidine 0.2 mg at bedtime. She was also on baclofen 1051 mcg/day by pump, oral loratadine 10 mg/day, and oral melatonin 3 mg at bedtime. Glycopyrrolate was scheduled at 1 mg in the morning, 1.5 mg at noon, and 1 mg at bedtime and was given orally as needed.

Six days following discharge, the child was readmitted for evaluation of persistent hypersomnolence. Since the patient's EEG and infectious workups were negative and the baclofen pump was functioning correctly, the neurology team suspected oversedation from interacting medications (Table). She was monitored overnight while the clonidine dose was weaned. She did not have any vital sign instability or signs of storming; however, the sedation did not significantly improve. Lorazepam was subsequently held on the third day of hospitalization, but there was no major improvement in her neurologic status. Because her seizures were controlled and her EEG was normal, a decision was made to wean her from the CLB. The dose was reduced from 10 mg twice a day to 5 mg twice a day (0.3 mg/kg/day) and her mental status began to improve. Serum concentrations of CLB and N-desmethylclobazam (NCLB) concentrations were not measured. Clobazam was discontinued on the sixth day of hospitalization, and oral valproic acid was initiated (i.e., 20 mg/kg load followed by 200 mg three times a day [20 mg/kg/day]). Previous medications were continued and she was discharged.

Table.

Drug-Drug Interactions Between CLB and Other Anticonvulsants Based on Inducers and Inhibitors of Substrates of CLB

CLB Substrate	Inducer	Inhibitor
CYP3A4 (major CLB)	Carbamazepine (strong); eslicarbazepine (moderate); felbamate (weak); oxcarbazepine (weak); phenobarbital (strong); phenytoin (strong); rufinamide (weak); topiramate (weak); valproate (weak)*	Valproate (weak); cannabidiol (strong)
CYP2C19 (minor CLB, major NCLB)	Carbamazepine (moderate); eslicarbazepine (moderate),* felbamate (weak); phenytoin (strong); valproate (weak)	Cannabidiol (strong); felbamate; lacosamide*; oxcarbazepine (weak); stiripentol (moderate); topiramate (weak); valproate (weak)
CYP2B6 (minor CLB)	Phenobarbital (weak); phenytoin (weak); cannabidiol (weak)
P-glycoprotein	Carbamazepine (weak); phenobarbital (strong); phenytoin (weak); cannabidiol (weak)	Stiripentol

^* In vitro.

The patient had improved somewhat after discontinuing CLB; however, her mental status had not returned to baseline. The in-home nurse also reported bradycardia, oxygen desaturation, and increased somnolence approximately 1 hour after each valproic acid dose. The physician recommended weaning the patient off valproic acid and stopping other non-required sedating medications (e.g., diphenhydramine, loratadine, clonidine, and melatonin).

Although sedation resolved about 10 days after discontinuation of CLB, the mother requested a second opinion of her daughter's seizure management and persistent sedation; hence, she was seen at our facility about 3 months later. At this time, the child's seizures were fairly well controlled on levetiracetam 1200 mg twice a day (80 mg/kg/day), and her anxiety was managed with lorazepam 2 mg (0.07 mg/kg/dose) every 4 hours as needed. A levetiracetam trough serum concentration was 17.6 mg/L (reference range: 15–45 mg/L). Routine laboratory studies were unremarkable. The mother stated that she only saw one brief seizure every 2 to 3 weeks, from which the patient recovered spontaneously or following a low dose of lorazepam. Genetic studies were performed in order to determine if the patient's sedation could be explained by a genetic polymorphism of the CYP2C19 enzymes. Results revealed she had a genetic variant at *2/*2 genotype with poor metabolizer enzyme activity for CYP2C19.

Discussion

Somnolence is the most common side effect associated with all benzodiazepines. It has been reported¹ to occur in about 26% to 32% of patients following initiation of CLB therapy and is one of the most common side effects in children. The patient in our case had hypersomnolence that was likely due to CLB per the Naranjo Adverse Drug Reaction Probability Scale (score = 7; Supplemental Table ^{(16.2KB, pdf)} ).³ In reviewing and assessing reasons for persistent hypersomnolence in this child, we considered pharmacodynamic, pharmacokinetic, or pharmacogenomic issues associated with CLB or a combination of these issues.

Pharmacodynamic. Like other benzodiazepines, CLB acts mainly through allosteric activation of gamma-aminobutyric acid type A (GABA_A) receptors. While classic 1,4-benzodiazepines (e.g., diazepam) are frequently non-selective, full agonists of GABA_A, CLB is thought to be a selective, partial agonist⁴; hence, CLB appears to exhibit better selectivity toward those GABA_A subunits (i.e., o2; a2/3b1/2g2) responsible for anxiolytic and anticonvulsant effects, but has less selectivity for those subunits (i.e., o1; a1b1/2g2) involved in sedation.^5,6 This may explain, in part, the lack of hypersomnolence in our patient with lorazepam.

In clinical trials, somnolence was reported¹ at all effective doses of CLB, was generally dose-related, and began within the first month of treatment. Subsequent doses in our patient were also consistent with recommendations; however, the rate of titration between each increase in dose may have been too rapid for the patient to tolerate. This is certainly true for someone with undiagnosed CYP2C19 polymorphism.

One might also consider that the hypersomnolence was due to other medications known to cause sedation (i.e., diphenhydramine, loratadine, clonidine, and melatonin). She had not previously experienced excessive sedation on these drugs, but the sedation could have been due to a synergistic effect when CLB was added. She also had intermittent sedation after each dose of oral valproate. Given that valproate was a recent addition, this sedation was an expected side effect. Valproate is also a minor substrate for CYP2C19; hence, decreased metabolism in a person with a polymorphism could explain the sedation. Sedation did resolve when CLB and other non-required sedating medications were stopped.

Pharmacokinetics. Hypersomnolence in our patient might also be related to altered metabolism due to a drug-disease or drug-drug interaction. Clobazam is metabolized in 2 steps via hepatic CYP isoenzymes.^2,7 During the first step, more than 70% of CLB is demethylated by CYP3A4 and, to a lesser extent, by CYP2C19 to yield a pharmacologically active metabolite (NCLB; Figure). N-desmethylclobazam is subsequently hydroxylated by CYP2C19 to the inactive 4′-hydroxydesmethylclobazam metabolite. Even though NCLB is less active than CLB (represents about 20% of its activity), at steady state its serum concentrations are typically 5- to 10-fold higher; hence, NCLB contributes significantly to the activity and adverse effect profile of CLB.

Figure.

Metabolic pathways and CYP isozymes for CLB (A) and lorazepam (B).

Organ Dysfunction. N-desmethylclobazam and its metabolites account for 94% of the total drug-related components in urine, while unchanged CLB accounts for 2%.^2,8 Even in the presence of renal dysfunction, the manufacturer does not recommend dosage adjustment in patients with mild or moderate impairment.¹ Roberts et al⁹ found that CLB and NCLB concentrations did not reach “toxic” values in adults with end-stage renal disease. and Tolbert et al¹⁰ reported that the formation of NCLB is elimination-rate limited and that the total apparent clearance of CLB is unaffected by hepatic impairment. However, the manufacturer recommends dosage adjustment in patients with mild to moderate hepatic impairment.¹ Our patient had normal renal and hepatic function; hence, organ impairment was eliminated as an etiology for the hypersomnolence.

Drug-Drug Interactions. CLB is demethylated by CYP3A4 to yield a pharmacologically active metabolite (NCLB), which is subsequently hydroxylated by CYP2C19 to form an inactive moiety (Figure). Antiepileptic drugs that change the activity of CYP3A4 or CYP2C19 have the potential to produce changes in CLB and/or NCLB concentrations as well as the ratio of NCLB to CLB.^1,2,8 These changes can cause a loss of therapeutic effect (i.e., induction of CYP3A4 and/or CYP2C19), produce adverse effects (i.e., inhibition of CYP2C19 or CYP3A4), and complicate therapeutic drug monitoring. Clobazam is used in many difficult-to-treat epilepsies; therefore, these individuals may be on numerous AEDs that have the potential to cause drug-drug interactions.²

Despite changes in CLB pharmacokinetics due to CYP3A4 or CYP2C19 inhibitors or inducers, some have suggested^7,11–14 that drug-drug interactions may not be clinically important. Authors hypothesized this may be due to the wide therapeutic index of CLB, poor correlation between serum concentrations and efficacy, or the failure of single dose studies in healthy volunteers or simulated drug-drug interaction studies to determine clinical relevance. Studies^13,15–23 involving chronic administration of CLB have found that drug-drug interactions may be clinically relevant and may necessitate adjustment of therapy.

Inducers. CYP3A4 is the most abundant hepatic isoenzyme accounting for the metabolism of more than 50% of all medications.²⁴ Potent non-AED inducers of CYP3A4 include rifampin, St John's Wort, and glucocorticoids.²⁴ Antiepileptic drugs that are strong, moderate, and weak inducers of CYP3A4 are noted in the Table.²⁴

Studies^{13,17,19–23} have shown a significant decrease in CLB and increase in NCLB serum concentrations following concurrent administration of an AED that induces CYP3A4 or CYP2C19 isozymes. Yamamoto et al²² reported that even at low serum concentrations, carbamazepine, phenobarbital, and phenytoin induced CYP3A4 but that phenytoin was the strongest inducer of CLB metabolism.

Theis et al¹³ and Contin et al¹⁹ both noted that AEDs that are CYP3A4 inducers decrease CLB and increase NCLB serum concentrations. Others^17,21,23 found that patients receiving CYP3A4 inducers had lower CLB concentration-to-dose ratios, an increase in NCLB serum concentrations, and an increase in NCLB/CLB ratio compared with those receiving non-inducing anticonvulsants.

Conversely, Walzer et al¹² used an integrated predictive population pharmacokinetic model of data from several studies. They found a non-significant increase in NCLB formation in those given CYP3A4 inducers and a non-significant increase in NCLB elimination in those given CYP2C19 inducers. Tolbert et al¹¹ came to the same conclusions; however, it appears that they used the same data set as did Walzer et al.¹²

Although valproate induces CYP3A4 in vitro,²⁵ its influence on CLB metabolism is unclear. Sennoune et al¹⁷ and Burns et al²¹ reported that the mean CLB/dose ratio in those given CLB monotherapy was not significantly different from that of patients taking valproate concomitantly. Conversely, Yamamoto et al²² demonstrated that concomitant use of valproate led to a marked decrease in CLB/dose ratio in both adult and pediatric patients. A possible explanation for this difference is that valproate might have increased the expression of P-glycoprotein, resulting in a decrease of the CLB-to-dose ratio. Concomitant use of topiramate or felbamate, both weak inducers of CYP3A4, did not influence the metabolism of CLB.^18,19 Although carbamazepine, phenobarbital, phenytoin, and valproic acid induce CYP2C19, this isozyme accounts for only 1.4% of conversion of CLB to NCLB; hence, induction of this isozyme is clinically unimportant.²⁶

Inhibitors. Drugs that inhibit CYP3A4 can block the conversion of CLB to NCLB, which may result in elevated concentrations of CLB and a decrease in the NCLB/CLB ratio. Strong inhibitors of CYP3A4 include clarithromycin, telithromycin, nefazodone, itraconazole, ketoconazole, and protease inhibitors (i.e., atazanavir, darunavir, indinavir, lopinavir, nelfinavir, ritonavir, saquinavir, and tipranavir).²⁵ Moderate inhibitors include amiodarone, erythromycin, fluconazole, miconazole, diltiazem, verapamil, delavirdine, amprenavir, fosamprenavir, and conivaptan, while cimetidine is a weak inhibitor.²⁵ The only commercially available AEDs that inhibit CYP3A4 are valproate, which is a weak inhibitor, and cannabidiol (CBD), which is a strong inhibitor (Table).²⁷

Walzer et al¹² evaluated the impact of ketoconazole, a known inhibitor of CYP3A4, on CLB in 36 healthy volunteers. Although they found a 54% increase in CLB area under the concentration curve (AUC), they conclude that there were no clinically important interactions between CLB and drugs that inhibit CYP3A4. Cannabidiol is also a strong inhibitor of CYP3A4. Geffrey et al²⁷ evaluated the impact of CBD and found a mean increase in serum CLB concentration of 60% ± 80% (95% CI, −2% to 91%) after 4 weeks of concomitant therapy.

Drugs that inhibit CYP2C19 can cause an accumulation of the active metabolite NCLB (Figure). Significant non-AED inhibitors of CYP2C19 include cimetidine, clarithromycin, diltiazem, erythromycin, fluoxetine, fluvoxamine, itraconazole, ketoconazole, lansoprazole, omeprazole, paroxetine, ritonavir, ticlopidine, and verapamil.²⁶ Oxcarbazepine and CBD are strong inhibitors of CYP2C19, while felbamate is a moderate inhibitor, and eslicarbazepine, topiramate, and valproate are weak inhibitors.²⁶

Walzer et al¹² studied the effect of omeprazole, a CYP2C19 inhibitor, in 36 healthy volunteers given a single oral dose of CLB. They reported that omeprazole had no effect on CLB concentrations. They did note an increase in NCLB's AUC (36%) and maximum serum NCLB concentration (15%), but they concluded that there should be no clinically meaningful drug-drug interactions between CLB and drugs metabolized by CYP2C19.

Several studies^13,27,28 have reported side effects (e.g., sedation) from CLB due to drug-drug interactions. Parman and Holmes²⁹ also described CLB-associated hypothermia due to supratherapeutic serum concentrations of CLB that resulted from the combination of CLB with 2 CYP2C19 inhibitors (i.e., oxcarbazepine and omeprazole).

Burns et al²¹ reported that patients chronically receiving AEDs that inhibit CYP2C19 had a high mean NCLB serum concentration and/or a total CLB concentration-to-dose ratio when compared with a neutral group. Contin et al¹⁹ also found that felbamate caused a higher concentration-to-dose ratio of NCLB and NCLB/CLB. Conversely, Renfroe et al¹⁴ performed a population pharmacokinetic analysis of 7 trials in patients given CLB with either valproic acid, felbamate, or oxcarbazepine and concluded that stable dosages of these AEDs did not affect CLB. Russell et al²³ found that AEDs that inhibited CYP2C19 generally increased NCLB/CLB concentration ratio, NCLB concentration/dose ratio, and CLB + NCLB concentration/dose ratio compared with a group of concentration sets in patients receiving only “neutral” AEDs.

In 2018 the Food and Drug Administration approved a plant-based oral formulation of CBD for Dravert and Lennox-Gastaut syndromes in patients ≥2 years of age. Cannabidiol is a potent inhibitor of CYP2C19.^28,30–33 Geffrey et al²⁷ evaluated the effect of CBD on CLB in 13 patients (4–19 years) with refractory epilepsy. The mean NCLB concentrations increased by 300% to 500% (a 3-fold increase) at 4 weeks. Importantly, 9 patients had a >50% decrease in seizures, corresponding to a responder rate of 70%, and 77% and 10 patients required a decrease in CLB dose. Concurrent CLB and CBD in healthy volunteers also increased the maximum serum concentration (Cp_max) and AUC of NCLB by approximately a 3-fold measure.³² Conversely, Morrison et al³⁴ found little effect of CBD on both the Cp_max and AUC of CLB and NCLB in patients ≥2 years of age.

At the time our patient was begun on CLB she was chronically receiving levetiracetam, lorazepam, baclofen, loratadine, melatonin, and glycopyrrolate. None of these induce or inhibit CYP3A4 or CYP2C19; hence, a drug-drug interaction did not contribute to our patient's hypersomnolence.

Pharmacogenomics. Knowledge of genetic polymorphism has significantly contributed to our understanding of interindividual and interethnic differences in the pharmacokinetics of drugs metabolized by CYC2C19.³⁵ Approximately 2% of Caucasians, 4% of African Americans, and 15% of Asians are CYP2C19 poor metabolizers (PMs).³⁶ To date, several genetic variant alleles of CYP2C19 have been identified, with CYP2C19*2 and CYP2C19*3 being the most common non-functional alleles responsible for most PM phenotypes.³⁷ Individuals carrying 1 or 2 copies of the defective CYP2C19*2 allele will have normal CLB pharmacokinetics but may develop markedly elevated serum NCLB concentrations and be at risk for adverse effects.

Several case reports^38–43 have highlighted the dangers of CYP2C19 polymorphisms with CLB therapy. Boels et al³⁸ described an adult admitted after a voluntary CLB overdose. Altered drug elimination was suspected when the NCLB serum concentrations continued to be elevated 30 days after discontinuation of CLB. Testing revealed the patient to have a homozygous mutated allele of CYP2C19*2. Because the patient had one functional gene, 100 mg of phenobarbital was given for 10 days in order to induce CYP2C19 metabolism of NCLB.

Contin et al³⁷ described 2 children on CLB with extremely high serum NCLB concentrations. Patient 1 was homozygous for CYP2C19*2. CLB was discontinued, but NCLB concentrations continued to be measurable for months. Patient 2 had only one copy of the same mutation but experienced several adverse effects, including somnolence. Parmeggiani et al³⁹ described a 10-year-old given CLB monotherapy for epilepsy. She presented with a 7-kg weight gain, severe somnolence, and both waking and sleeping enuresis due to elevated NCLB concentrations. She was subsequently found to have one copy of the CYP2C19*2 mutation.

Seo et al⁴⁰ retrospectively investigated the association between CYP2C19 genotypes and CLB adverse effects and found that 37.3% of Japanese subjects were extensive metabolizers (EMs) and 22.7% were PMs. The incidence of adverse effects (e.g., drowsiness, dizziness) was significantly higher in PMs (64%) than in EMs (39%). The NCLB concentration was 1.5-fold greater in those with at least one variant, and NCLB concentration increased with the number of CYP2C19-defective alleles. Saruwatari et al⁴¹ evaluated the impact of genotypes on the pharmacokinetics of CLB and NCLB in patients with CYP2C19*2 (27.6%) and *3 (12.9%) polymorphisms. Using 128 steady-state serum CLB and NCLB concentrations they found that the clearance of NCLB was significantly lower (84.9%) in PMs. Yamamoto et al⁴² also found a significant difference in the mean ratios of serum CLB and NCLB concentrations to the CLB dose in EM (3.1) and PM (21.6) patients.

Interestingly, patients with a defect in the CYP2C19 gene may have different rates in terms of seizure control. Hashi et al⁴³ found that seizure frequency was significantly decreased in PMs compared with EMs. CLB serum concentrations did not correlate with seizure reduction, but median NCLB serum concentrations were significantly higher in patients with excellent seizure control.

The child described in our case received CLB prior to identification of her CYP2C19 polymorphism. We believe that her genetic variant at *2/*2 genotype was the major factor that contributed to her persistent hypersomnolence. Although she was receiving lorazepam prior to the addition of CLB, one would not expect her PM status to affect lorazepam clearance because lorazepam is metabolized by glucuronidation and not CYP isozymes (Figure).

Conclusions and Recommendations

Assessing the causes of mental status changes in a patient should exclude ongoing electroencephalographic seizures via prolonged EEG and the presentation of other etiologies (e.g., subacute hypothyroid, infection, metabolic). As polytherapy is used frequently in patients with intractable epilepsy it is important to consider the contribution of drug-drug interactions, both prescription and non-prescription (including herbals and CBD), in any patient who exhibits loss of seizure control or the onset of adverse effects. This is especially true for patients on CLB who receive CYP3A4 inducers or CYP2C19 inhibitors.

When patients present with possible adverse effects to CLB it is imperative that the initial dosage is appropriate for age, organ function, and metabolizer status (if known) and that the rate of titration to a maintenance dose is consistent with recommendations. If a patient is known to be a CYP2C19 PM, the manufacturer recommends an initial dose of 5 mg once daily for ≥1 week. Subsequent increases are based on patient tolerability and response, but generally the dose is increased to 5 mg twice daily for ≥1 week, followed by an increase to 10 mg twice daily for ≥1 week.

Current recommendations do not include the use of therapeutic drug monitoring to ensure patients do not suffer from CLB or NCLB toxicity. In an earlier study²³ we found that therapeutic drug monitoring of CLB and NCLB with calculation of various Cp/Cp and Cp/dose ratios may help differentiate between drug-drug interactions and PM status. Measurement of serum NCLB concentration alone may also be clinically useful for identifying individuals with a PM phenotype. In fact, Yamamoto et al⁴² noted that when the cut-off value of the Cp/dose ratio for NCLB was set as 10.0 (mg/L)/(mg/kg) the sensitivity and specificity for predicting CYP2C19 PM status were 94.4% and 95.7%, respectively.

Current recommendations do not include the use of genetic testing prior to beginning CLB; however, knowledge of a patient's CYP2C19 genotypes may prove helpful in achieving therapeutic serum CLB and NCLB concentrations and in avoiding unnecessary adverse effects due to NCLB. Unlike genetic testing that is used to ascertain a possible etiological diagnosis (e.g., epilepsy gene panels), genetic testing to assay for CYP isoenzyme function is easily obtained through standard laboratories (e.g., LabCore, Quest). Typically it does not require prior authorization, and results are usually returned in about 5 to 7 days.

In order to safely and effectively use CLB, clinicians should have a working knowledge of the complex metabolism of CLB and the importance of the NCLB metabolite in therapy. They should be able to appropriately monitor and interpret CLB and NCLB serum concentrations and should also be aware of the potential for drug-drug interactions in patients receiving CYP3A4 inducers and CYP2C19 inhibitors. An understanding of the impact of CYP2C19 genotype status on CLB clearance is important in differentiating the response to CLB. Future research should focus on the development of impact models that prevent adverse effects in those with unidentified genetic polymorphisms and on targeting which patients may benefit from CYP2C19 pharmacogenomic testing. When CLB dosing is appropriate, serum concentrations are monitored, and CYP2C19 status is known, even patients with a history of CLB-induced adverse effects can have their dose modified and continue on CLB without significant central nervous system adverse effects.

ABBREVIATIONS

AED

antiepileptic drugs

AUC

area under the curve

CBD

cannabidiol

CLB

clobazam

Cp_max

maximum serum concentration

CYP

cytochrome P450

EEG

electroencephalogram

EM

extensive metabolizer

GABA_A

gamma-aminobutyric acid type A

NCLB

N-desmethylclobazam

PM

poor metabolizer