E

Att

HA

a

b

Pc

RA

1

h0

pilepsy Research (2014) 108, 1204—1211

j ourna l h om epa ge: www.elsev ier .com/ locate /ep i lepsyres

ssessment of pharmacokinetics andolerability of intranasal diazepam relativeo rectal gel in healthy adults

erbert R. Henney III a, Michael R. Sperlingb,drian L. Rabinowicza,∗, Gary Breamc, Enrique J. Carrazanaa

Acorda Therapeutics, Inc., 420 Saw Mill River Road, Ardsley, NY 10502, USADepartment of Neurology, Thomas Jefferson University, 901 Walnut Street, Suite 400,hiladelphia, PA 19107, USAAerial BioPharma, LLC, 9001 Aerial Center Parkway, Suite 110, Morrisville, NC 27560, USA

eceived 14 January 2014; received in revised form 21 March 2014; accepted 28 April 2014vailable online 13 May 2014

KEYWORDSDiazepam;Intranasal;Pharmacokinetics;Seizures;Adults

Summary Diazepam rectal gel (RG) is currently the only approved rescue therapy for out-patient management of seizure clusters in the United States. There is an unmet medical needfor an alternative rescue therapy for seizure clusters that is effective, and more convenient toadminister with a socially acceptable method of delivery. An intranasal diazepam formulationhas been developed, and this study evaluates the tolerability and bioavailability of diazepamnasal spray (NS) relative to an equivalent dose of diazepam-RG in healthy adults. Twenty-four healthy adults were enrolled in a phase 1, open-label, 3-period crossover study. Plasmadiazepam and metabolite concentrations were measured by serial sampling. Dose proportion-ality for 5- and 20-mg intranasal doses and the bioavailability of 20 mg diazepam-NS relative to20 mg diazepam-RG were assessed by maximum plasma concentration (Cmax) and systemic expo-sure parameters (AUC0—∞ and AUC0—24). The mean Cmax values for 20 mg diazepam-NS and 20 mgdiazepam-RG were 378 ± 106 and 328 ± 152 ng/mL, achieved at 1.0 and 1.5 h, respectively. Sub-jects administered intranasal and rectal gel formulations experienced nasal and rectal leakage,respectively. Diazepam absorption following intranasal administration was consistent but 3 sub-

jects with diazepam-RG had low plasma drug levels at the earliest assessment of 5 min, due topoor retention, and were excluded from analysis. Excluding them, the treatment ratios (20 mgdiazepam-NS:20 mg diazepam-RG) and 90% confidence intervals for diazepam Cmax and AUC0—24

were 0.98 (0.85—1.14) and 0.89was comparable between the tthe lowest and highest dose-st

∗ Corresponding author at: Clinical Development and Medical Affairs,0502, USA. Tel.: +1 914 326 5138.

E-mail address: arabinowicz@acorda.com (A.L. Rabinowicz).

ttp://dx.doi.org/10.1016/j.eplepsyres.2014.04.007920-1211/© 2014 Published by Elsevier B.V.

(0.80—0.98), respectively, suggesting that the bioavailabilitywo formulations. Dose proportionality was observed betweenrengths of intranasal formulation. Both intranasal and rectal

Acorda Therapeutics, Inc., 420 Saw Mill River Road, Ardsley, NY

Pharmacokinetics of intranasal diazepam 1205

treatments were well tolerated with mild to moderate adverse events. Results suggest thata single-dose of 20 mg diazepam-NS is tolerable and comparable in bioavailability to that ofdiazepam-RG. The intranasal formulation may provide caregivers and patients with a moresocially acceptable and convenient alternative rescue therapy in the acute treatment of seizureclusters.© 2014 Published by Elsevier B.V.

atsap

M

S

T3isfieiStwwtpbwpdt

ioleDcwBs

S

T1were supplied to the study site as microemulsion by UPMPharmaceuticals (Baltimore, MD). To administer the drug,a single-use Pfeiffer bi-dose (Aptar Pharma, Congers, NY)

Introduction

Despite the availability of anti-epilepsy drugs (ASDs),approximately a third of patients with epilepsy are treat-ment resistant (Friedman and French, 2012) and seizureclusters occur in some individuals with treatment-resistantepilepsy (Sillanpää and Schmidt, 2008). Clusters, alsoreported as acute repetitive seizures, are characterized bymultiple seizures occurring over a relatively brief periodof time (Cereghino, 2007). One operational definition ofseizure cluster is the occurrence of 3 or more seizures withina 24 h period with return to baseline between events (Haut,2006).

Seizure clusters commonly occur outside the hospital,and benzodiazepines, particularly diazepam, are the first-line therapy in the outpatient setting (Pellock, 2007; Chenand Wasterlain, 2006; Holsti et al., 2007). A rectal gelformulation of diazepam (diazepam-RG) is currently theonly FDA-approved rescue therapy for the out-of-hospitalmanagement of patients with refractory epilepsy who areon stable regimens of AEDs but require intermittent useof diazepam to control bouts of increased seizure activ-ity (Diazepam rectal gel prescribing information, 2005).Previous studies have reported that home treatment withdiazepam-RG was associated with greater seizure control,improved quality of life and fewer emergency room visits(Kriel et al., 1991, 1999). Although diazepam-RG may alle-viate the need for treatment in an acute care facility,thereby reducing the time between seizure onset and medi-cal intervention, rectal administration still requires deliveryby trained caregivers and can be technically difficult toadminister during a seizure. In addition, rectal adminis-tration is inconvenient (Tatum, 2002) and can be sociallyembarrassing due to lack of personal privacy (Bhattacharyyaet al., 2006). Hence, there is an unmet need for an alterna-tive therapy that is both socially acceptable and convenientto administer in an outpatient setting.

Intranasal delivery provides a convenient and sociallyacceptable delivery route (Wilson et al., 2004; de Haanet al., 2010; Holsti et al., 2010; Kyrkou et al., 2006). Adiazepam nasal spray formulation (diazepam-NS) has beendeveloped as an alternative to diazepam-RG to controlincreased bouts of seizure activity in patients with epilepsywho are on stable regimens of AEDs (Bream et al., 2013).The intranasal diazepam formulation is a single-use devicewith strengths based on the dosing schedule of the refer-enced product, diazepam-RG. The device consists of twointranasal sprays with each spray of approximately 100 �L

to be administered per nostril.

The objective of the study reported here was to deter-mine the bioavailability of a single dose of diazepam-NSformulation relative to diazepam-RG in healthy individuals

nwie

nd to evaluate the pharmacokinetic (PK) linearity betweenhe minimum and maximum dose-strengths of the nasalpray formulation in healthy adults. The study alsossessed the tolerability of diazepam-NS formulation in thisopulation.

ethods

tudy design

his was a phase 1, single-center, open-label, single-dose,-period, randomized, crossover study with 14 days separat-ng the dose administrations. Following a 28-day screening,ubjects who met the eligibility criteria received 1 of theollowing treatments during each dosing period in a random-zed order: a 5-mg dose as one 2.5-mg spray of 100 �L inach nostril, a 20-mg dose as one 10-mg spray of 100 �Ln each nostril, or a single 20-mg dose of diazepam-RG.ubjects were required to fast for 10 h prior to dosing ando remain fasting for 2 h after administration. Diazepam-NSas administered in a reclined position at approximately 45◦

ith the subject’s head facing straight upward. No prepara-ion such as blowing or clearing of the nose was performedrior to dosing, and subjects were asked to refrain fromlowing their nose for 4 h after administration. Diazepam-RGas administered in the lateral decubitus position withoutreparation or bowel cleansing. Subjects were kept in theosing position for 30 min with no restriction in movementhereafter.

This study was conducted in accordance with Good Clin-cal Practice requirements described in the current versionf the International Conference on Harmonization guide-ines and all applicable regulations, and compliant with thethical principles described in the current version of theeclaration of Helsinki. Prior to study initiation, the proto-ol, informed consent form, and all other study documentsere reviewed and approved by an Institutional Reviewoard. Written informed consent was obtained from allubjects.

tudy drugs

wo concentrations of intranasal formulations (2.5% and0% diazepam for 5 mg and 20 mg doses, respectively)

asal spray device was used, capable of two actuationsith one spray per nostril. The inactive ingredients of the

ntranasal formulation included diethylene glycol monoethylther, propylene glycol monocaprylate, methyl laurate,

1

NdNt

S

Mbmmtwcanftomo

D

Btt11adD1opp3

P

NPeptc((htfo

tRCvsatf

lta(ldna

lr>tmqoo

T

Seetmisa(iotomnosrtias

msdartC

R

Aysp

206

-methyl-2-pyrrolidone, ethanol, and sterile water. Theiazepam rectal gel formulation (Valeant Pharmaceuticals,orth America) was supplied by the study site and adminis-ered per the instructions provided in the package insert.

ubjects

en and women 18—50 years of age, inclusive, weighingetween 88 and 111 kg with general good health based onedical history, physical examination, and screening assess-ents were eligible. The weight range was chosen based on

he diazepam-RG dosing information so that each subjectould receive a 20-mg dose. Subjects with the followingriteria were excluded: pregnant or lactating, history ofllergy or hypersensitivity to diazepam, history of any ill-ess that may affect the PK of diazepam, abnormal liverunction tests, drug or alcohol abuse, use of tobacco or nico-ine products within 90 days prior to study entry, or usef any medications other than vitamin supplements, hor-onal birth control, or replacement therapy within 1 week

f randomization and throughout study participation.

rug assay

lood sampling for the PK measurements of diazepam andhe desmethyl metabolite was performed pre-dose and athe following times after each dose administration: 5, 10,5, 30, and 45 min, and 1, 1.5, 2, 4, 6, 9, 12, 24, 48, 96,44, 192, and 240 h. Plasma diazepam concentrations weressayed using a validated liquid chromatography with tan-em mass spectrometry assay (LC—MS/MS; Tandem Labs,urham, NC). The assay had a lower limit of quantitation of

ng/mL and was validated for concentrations in the rangef 1—500 ng/mL based on the analysis of 0.2 mL of humanlasma. Based on the inter-run quality control data, therecision and the accuracy of the assay for diazepam were—5.9% and 2.4—6.8%, respectively.

harmacokinetic analysis

oncompartmental analyses (WinNolin® Phoenix version 6.1,harsight Corporation, Mountain View, CA) were used tostimate the following PK parameters from plasma sam-les: maximum observed plasma concentration (Cmax); timeo maximum concentration (Tmax); area under the plasmaoncentration—time curves for time 0—24 h post-doseAUC0—24), to time of last measurable plasma concentrationAUClast), and extrapolated to infinity (AUC∞); eliminationalf-life (t1/2); apparent systemic clearance (CL/F). Concen-rations below the limit of quantification were assigned as 0or computation of descriptive statistics and for constructionf mean concentration—time profiles.

The relative bioavailability of diazepam with respecto 20 mg diazepam-NS (test product) and 20 mg diazepam-G (reference product) was evaluated by comparing themax, AUC0—24, AUClast, and AUC∞ using an analysis ofariance (ANOVA) model with sequence, subject within

equence, treatment, and period as the classification vari-bles using the original data and the natural logarithm ofhe data. Confidence intervals (CIs) (90%) were constructedor the treatment ratios (test to reference) using the

wwuf

H.R. Henney III et al.

og-transformed data and the two 1-sided test procedureo determine whether the CIs were within the 80—125%cceptance range for the two formulations to be comparableGuidance for industry [FDA], 2014). The pharmacokineticinearity between the 5-mg and 20-mg dose strengths ofiazepam-NS formulation was assessed by comparing theatural log-transformed and dose normalized Cmax, AUC0—24,nd AUC∞ and the calculated CL/F values.

Subjects were observed for visual signs of rectal or nasaleakage at 5, 15, 30, 45, and 60 min after administration. Forectal doses, volume of loss was estimated as none, ≤1 mL,1 to ≤3 mL, or >3—4 mL, and at the 45 and 60-min checks,he maximum amount of leakage since the previous assess-ent was noted. For intranasal doses, nasal leakage wasuantified as none or observed and included a descriptionf the amount of leakage; occurrence of sneezing and timef events were also noted.

olerability

afety and tolerability assessments included treatment-mergent adverse events (TEAEs), clinical laboratoryvaluations, vital signs, pulse oximetry, physical examina-ions, and electrocardiograms. The TEAEs were rated asild (transient, required no special treatment, and did not

nterfere with subject’s daily activities), moderate (requiredimple therapeutic measures, low degree of inconvenience,nd may have interfered with daily activities) or severerequired systemic drug therapy or other treatment andnterrupted daily activities). Independent of the assessmentf TEAEs, a nasal and pharyngeal irritation or inflamma-ion assessment tool was used to determine the proportionf subjects with nasal and pharyngeal symptoms by treat-ent at pre-dose and 30 min, 1, 8, and 24 h post-dose. The

asal mucosa was inspected for the presence or absencef redness, edema, congestion, runny nose, or for otherigns or symptoms. Similarly, the pharynx was assessed foredness, edema, post-nasal drip, or for other signs or symp-oms. Subjects were also asked to report any incident ofrritation or inflammation in between the evaluation time,nd the reported discomfort was rated as mild, moderate, orevere.

To assess the level of alertness following diazepam treat-ent, a 4-level categorical scale was used that rated

ubjects by treatment and time point as being either alert,rowsy, sleeping but arousable, or sleeping not arous-ble. The assessment of treatment-emergent suicidality, asequired by regulatory authorities for all central nervous sys-em compounds in development, was evaluated using theolumbia-Suicide Severity Rating Scale (C-SSRS).

esults

total of 24 subjects (20 men and 4 women; mean age 34ears, range 21—46 years; mean weight 96.4 ± 4.98 kg) werecreened and all met the eligibility criteria for study partici-ation. All 24 subjects completed dosing periods 1 and 2 and

ere included in the safety and PK analyses. Four subjectsere not dosed in period 3, 2 withdrew consent for reasonsnrelated to adverse events (AEs), and 2 were discontinuedor positive urine drug screens.

Pharmacokinetics of intranasal diazepam 1207

-100

0

100

200

300

400

500

600

700

800

0 4 8 12 16 20 24

Me

an

dia

ze

pa

m (

ng

/mL

)

Time (h)

5 mg diazepam-NS 20 mg diazepam-RG 20 mg diazepam-NS

-100

0

100

200

300

400

500

600

700

0 4 8 12 16 20 24

Me

an

dia

ze

pa

m (

ng

/mL

)

Time (h)

5 mg diazepam-NS 20 mg diazepam-RG 20 mg diazepam-NS

A

B

Figure 1 Mean plasma diazepam concentration—time profile by treatment: (A) after excluding 3 diazepam-RG subjects with low

2p

pfRratCc3vf

N

bioavailability, and (B) including all subjects.

Pharmacokinetics

Concentration—time profiles for 5-mg and 20-mg diazepam-NS formulations showed that the Cmax of 96.3 ± 28 and350 ± 103 were reached at a median Tmax of 1.0 h, whereasfor 20 mg diazepam-RG, the Cmax of 352 ± 93 ng/mL wasreached at 1.5 h after administration (Fig. 1). Assessmentfor leakage at 5—60 min after administration of diazepam-RG showed varying degree of leakage in 7 subjects (33%).However, only 3 subjects (13%) experienced poor retentionat the earliest assessment of 5 min, with leakage varyingin estimated quantity from less than 1 mL to 3 mL. Thesesubjects had lower diazepam exposures with mean Cmax lessthan 10% measured in those with good rectal gel retention.Nasal leakage was noted in 15 subjects (65%) at various timepoints from 5 to 60 min after dosing with 20 mg diazepam-

NS. Four subjects sneezed within 2—36 min of dosing and3 had nasal leakage. However, regardless of nasal leakageor sneezing, the absorption of diazepam was consistent,as the coefficients of variation for Cmax and AUC0—24 with

rewC

0 mg diazepam-NS were 28% and 27%, respectively, com-ared with 46% and 43% with rectal gel.

Table 1 describes the summary statistics of noncom-artmental PK parameters by dosing. The mean ± SD Cmax

or 5-mg and 20-mg diazepam-NS and 20-mg diazepam-G groups were 108 ± 31, 378 ± 106, 328 ± 152 ng/mL,espectively; AUC0—24 values were 823 ± 285, 2720 ± 738,nd 2660 ± 1146 ng h/mL, respectively. After excludinghose subjects with rectal diazepam leakage, the meanmax and AUC0—24 values for 20 mg diazepam-RG wereomparable to 20 mg diazepam-NS (375 ± 97 ng/mL and015 ± 710 ng h/mL, respectively). The estimated mean t1/2

alues for diazepam varied from 96 to 108 h with a rangerom 43.8 to 243 h.

To evaluate the bioavailability of the 20-mg diazepam-S formulation relative to 20 mg diazepam-RG, treatment

atios and 90% CIs of natural log-transformed PK param-ters were compared. Comparative bioavailability data,hen including all subjects, indicated that the ratios ofmax and AUC0—24 for 20 mg diazepam-NS exceeded 100%

1208 H.R. Henney III et al.

Table 1 Diazepam noncompartmental PK parameters by treatment.

Group N Cmax (ng/mL)Mean (SD)

Tmax (h)Median

AUC0—24 (ng h/mL)Mean (SD)

5 mg diazepam-NS 22 108 (30.5) 1.0 823 (285)20 mg diazepam-NS 23 378 (106) 1.0 2720 (738)20 mg diazepam-RG 23 328 (152) 1.5 2660 (1146)20 mg diazepam-RG (subpopulation) 20a 375 (96.8) 1.25 3015 (710)

a Excluding subjects with poor retention.

Table 2 Bioavailability analysis between 20 mg diazepam-NS and 20 mg diazepam-RG.

PK parameters Geometric meana Ratio (%)b

(test/ref)90% CI Power

20 mg diazepam-NS(test) N = 23

20 mg diazepam-RG(ref) N = 20

Lower Upper

Ln(Cmax) 361.143 366.801 98.45 85.30 113.64 0.8260Ln(AUC0—24) 2631.576 2972.011 88.55 80.23 97.72 0.9790Ln(AUClast) 9172.612 10,868.107 84.40 75.44 94.42 0.9482Ln(AUC∞) 11,118.795 13,378.723 83.11 75.34 91.68 0.9796

Ln, natural logarithm.a Geometric mean based on least-squares mean of ln-transformed parameter values.

ar9sgqefwabotc

aldhsdmthCttomsm

to

wcvNcpd1t1l

T

Aersiamtiwaaaww

b Ratio (%) = geometric mean (test)/geometric mean (ref).

nd the 90% CIs were outside of the standard acceptanceange (Cmax: 145.16, 90% CI: 98.73—213.43; AUC0—24: 119.76,0% CI: 88.77—161.58). This result is due to inclusion ofubjects with low diazepam exposure in the diazepam-RGroup that altered the distribution of the data and conse-uently the ratios. When the 3 subjects with low diazepamxposures following rectal administration were excludedrom the analysis, the ratios for Cmax and AUC0—24 wereithin the 90% CI acceptance range, whereas for AUClast

nd AUC∞, ratios were outside the lower end of the rangey 5% (Table 2). These results show that the bioavailabilityf 20 mg diazepam administered intranasally is comparableo an equivalent dose given rectally when the latter is notompromised by leakage.

To determine whether linearity exists between the lowestnd highest dose-strengths of diazepam intranasal formu-ation, dose-normalized PK parameters of 5-mg and 20-mgoses were compared (Table 3). Due to the observed longalf-life of diazepam, which ranged from 43.8 to 243 h acrossubjects and treatments, a carry-over of residual plasmaiazepam was observed when a 20-mg dose of intranasal for-ulation or rectal gel preceded a 5-mg dose, especially in

hose subjects with diazepam half-life greater than 80 h. Asigher pre-dose diazepam concentrations would impact themax and AUC values, the analysis utilized corrected concen-rations by subtracting residual diazepam using the averageerminal-phase rate constant for each subject. Comparisonf dose-normalized PK parameters showed that the treat-ent ratios and the respective 90% CIs were within the

tandard acceptance range, thus demonstrating that the 5-

g and 20-mg strengths were dose-proportional.To address the potential impact of the route of adminis-

ration on the metabolism of diazepam, the PK parametersf desmethyl diazepam obtained from dosing period 1

tRch

ere compared for all 3 treatments. The results indi-ated that the metabolite concentrations accumulatedery slowly over time after dosing with 20 mg diazepam-S and rectal gel treatments with nearly superimposableoncentration—time profiles (Fig. 2). The highest meanlasma metabolite concentrations for 5 mg and 20 mgiazepam-NS and diazepam-RG were 9.9 ± 3.1 ng/mL at44 h and 37.3 ± 13 and 35.5 ± 14.5 ng/mL at 96 h, respec-ively, indicating that the metabolite Cmax values were about0% of those of diazepam following each treatment, regard-ess of route of administration and dose.

olerability

ll subjects reported at least 1 TEAE (Table 4). All thesevents were considered mild or moderate in severity andesolved without treatment. There were no serious AEs orevere TEAEs. There were no clinically significant changesn vital signs, respiratory function, or electrocardiograms inny of the treatment groups. Increased lacrimation was theost commonly reported TEAE by 78—82% of the subjects

reated with 5-mg and 20-mg diazepam-NS doses, reflect-ng the local effects of intranasal administration. This TEAE,hich occurred almost immediately after dosing, was mildnd was of short duration; 47% were resolved within 5 minfter administration, 92% were resolved within 45 min, andll were resolved within 3 h of onset. Other TEAEs thatere related to local effects of intranasal administrationere nasal symptoms reported in 9% to 44% in each of

he 2 diazepam-NS groups compared with 4% for diazepam-G. These included rhinorrhea, nasal inflammation, nasalongestion, nasal discomfort, nasal edema, paranasal sinusypersecretion, dysgeusia, and sneezing. Pharyngeal TEAEs

Pharmacokinetics of intranasal diazepam 1209

Table 3 Dose-proportionality analysis between minimum and maximum dose-strengths of diazepam intranasal formulations.

Dependent variable Geometric meana Ratio (%)b (20 mg/5 mg) 90% CI Power

20 mg 5 mg Lower Upper

Ln(Cmax/dose) 17.9529 19.6278 91.47 82.02 102.01 0.9569Ln(AUC0—24/dose) 143.5121 137.3799 94.25 87.70 101.29 0.9991Ln(AUC∞/dose) 539.8293 556.9134 96.93 88.74 105.89 0.9918Ln(CL/F) 1.8524 1.7956 103.16 94.44 112.69 0.9918

a Geometric mean based on least-squares mean of ln-transformed parameter values.b Ratio (%) = geometric mean (20 mg)/geometric mean (5 mg).

-5

0

5

10

15

20

25

30

35

40

45

0 4 8 12 16 20 24

Me

an

de

sm

eth

yl

dia

zep

am

(n

g/m

L)

Time (h)

5 mg diazepam-NS 20 mg diazepam-RG 20 mg diazepam-NS

Figure 2 Mean desmethyl diazepam concentration—time profiles for dosing period 1 after excluding data from 3 diazepam-RGsubjects with low bioavailability.

Table 4 TEAEs reported by ≥5% of subjects in any treatment group.

TEAE (preferred term) Diazepam-NS Diazepam-RG

5 mg (N = 22) N (%) 20 mg (N = 23) N (%) 20 mg (N = 23) N (%)

Subjects with one or more TEAE 21 (96) 23 (100) 17 (74)Lacrimation increased 18 (82) 18 (78) 0Somnolence 5 (23) 12 (52) 14 (61)Rhinorrhea 7 (32) 10 (44) 1 (4)Nasal inflammation 2 (9) 9 (39) 1 (4)Dizziness 1 (5) 4 (17) 5 (22)Nasal congestion 4 (18) 4 (17) 0Nasal discomfort 4 (18) 2 (9) 0Paranasal sinus hypersecretion 0 4 (17) 0Dysgeusia 1 (5) 2 (9) 0Fatigue 1 (5) 0 3 (13)Feeling abnormal 0 1 (4) 2 (9)Nasal edema 0 3 (13) 0Sneezing 2 (9) 2 (9) 0Euphoric mood 0 0 2 (9)Eye irritation 2 (9) 0 0

1

sldn26r

tiaotdfsob

wseC

D

Tdwtcatw

kbftAtivrePatw7taeolsslo3t

toswlll

avtctpthtfitkwteceortstwsrrwpdrsDttmo1

Ntsaintaoeea

210

uch as dry throat (4.5%) and throat irritation (4.5%) wereess common and were reported only in the diazepam-NSose groups. Systemic TEAEs such as somnolence and dizzi-ess were dose-related and were reported by the 5-mg and0-mg diazepam-NS and diazepam-RG groups: 23%, 52%, and1% for somnolence and 5%, 17%, and 22% for dizziness,espectively.

The nasal and pharyngeal assessments indicated thathe incidence of nasal symptoms were dose-related to thentranasal doses, as 32% treated with 5 mg diazepam-NSnd 48% with 20 mg diazepam-NS experienced symptomsf nasal redness, congestion, or runny nose. The symp-oms peaked at 30—60 min after administration of intranasaloses and returned to baseline levels by 8 h. Nasal discom-ort, described as mild, was less common and occurred in 4ubjects during the 24 h observation period. Pharyngeal signsr symptoms were also less common and were not reportedy more than 9% in either of the diazepam-NS groups.

Alertness evaluation indicated that decrease in alertnessas dose-related and that by 24 h after administration all

ubjects were alert. There were no indications of treatment-mergent suicidal ideation or behaviors, as evaluated by-SSRS.

iscussion

he results of this study showed that nasally administerediazepam was well tolerated with mild local effects resolvedithout treatment, and that the rate and extent of absorp-

ion of diazepam from the 20-mg intranasal formulation isomparable to that of 20 mg of the rectal gel. This studylso demonstrated that the lowest and highest strengths ofhe intranasal diazepam formulation are dose proportionalhen assessed using the dose-normalized PK parameters.

Large inter-patient variability in diazepam PK is wellnown. A study (Ivaturi et al., 2013) that compared theioavailability of two investigational diazepam intranasalormulations with rectal gel in healthy adults showed thathe Cmax values and systemic exposure were comparable.lthough the nasal formulations were rapidly absorbed inheir study (median Tmax: 0.75 h) with 70—90% bioavailabil-ty relative to rectal gel, the authors noted considerableariability between subjects in Cmax and exposure, possiblyelated to incomplete nasal dosing in some subjects (Ivaturit al., 2013). The same study also discussed variability inK following diazepam-RG administration. On the basis of

pooled analysis of healthy subjects, the study indicatedhat the between-subject PK variability in those treatedith diazepam-RG was substantial and ranged from 30% to0% (Ivaturi et al., 2013). The variability in the absorp-ion of diazepam when administered as rectal gel has beenttributed to rectal contents as well as to leakage (Lamsont al., 2011). Although in the current study rectal leakagef diazepam was observed in 7 of the 23 (30%) subjects,eakage at the earliest assessment of 5 min was noted in 3ubjects (13%). Notably, the diazepam exposure in these 3ubjects was substantially low due to poor retention with

evels less than 10% of the systemic exposure measured inther subjects from the diazepam-RG group (mean Cmax:75 ± 97 ng/mL). Nasal leakage was noted in 15 subjects upo 1 h after dosing with the 20-mg nasal formulation and 3 of

tcet

H.R. Henney III et al.

hese subjects sneezed within 36 min of dosing. Regardlessf nasal leakage or sneezing, the diazepam exposure in theseubjects was comparable to the diazepam-RG group subjectsho did not experience low drug exposure due to early rectal

eakage; this suggests that nasal leakage or sneezing fol-owing intranasal formulation may not have resulted in drugoss.

Rectal administration of medications is generally vieweds inconvenient and socially embarrassing. However, a sur-ey of adult patients with resistant epilepsy indicated thathe majority (93%) would prefer a rectal medication thatould be used outside the hospital than to be transported tohe emergency room (Tatum, 2002). Despite indicating thisreference, nearly 50% of those surveyed were of the viewhat the use of rectal medication is equally embarrassing toaving a seizure, suggesting that embarrassment is a barriero the use of rectal medication for seizure clusters. Thesendings highlight the need for an alternative rescue medica-ion in the treatment of seizure clusters. In this regard, theey attributes of a rescue medication for seizure clustersould be the time to reach the plasma diazepam concen-

ration to exert anticonvulsant effect and to sustain thexposure levels for sufficient duration to maintain seizureontrol. A number of earlier studies have attempted tostablish an association between the anticonvulsant effectf diazepam and its PK. In a study by Milligan et al. (1981),ectal administration of diazepam was highly effective inerminating absence status as evidenced by reduction ofpike-wave activity in the electroencephalogram. Notably,he authors stated that the therapeutic effect of diazepamas observed at serum levels below 150 ng/mL. In their

ubsequent studies, Milligan et al. showed a significanteduction in spike frequency (Milligan et al., 1982) andecurrent seizures (Milligan et al., 1984) when patientsere treated with rectal diazepam suppositories com-ared with placebo, which corresponded to mean serumiazepam levels of 210 ± 125 ng/mL and 190 ± 73 ng/mL,espectively. However, no correlation was found betweenerum diazepam concentration and anticonvulsant effect.espite the lack of consensus on plasma diazepam concen-ration required for anticonvulsant effect, Dreifuss et al. inhe original report of diazepam-RG designed their study toaintain diazepam concentrations at 150—300 ng/mL based

n data from intravenous administration (Dreifuss et al.,998).

In the current study, the median Tmax for 20 mg diazepam-S was 1 h with comparable Cmax and systemic exposure tohat of 20 mg diazepam-RG. These findings suggest that aingle-dose of 20 mg diazepam-NS could potentially providedequate seizure control for those patients experiencingncreased seizure activity and may thereby alleviate theeed for costly emergency medical care. Considering thathe purpose of this study was to assess the relative bioavail-bility of diazepam-NS formulation in the acute treatmentf seizures, such as in seizure cluster, Cmax and AUC0—24

stimates were considered clinically more relevant thanxposures derived using AUC∞. With the use of bioequiv-lence methodologies as measured by Cmax and AUC0—24,

he bioavailability of 20 mg diazepam-NS was found to beomparable to 20 mg diazepam-RG. Although the estimatedquivalence for AUClast and AUC∞ were similar, the lowerails of the respective CIs fell outside the accepted range

2

D

F

2

HH

H

I

K

K

K

L

M

M

M

P

S

Pharmacokinetics of intranasal diazepam

by 5%. However, these differences should not be consid-ered as clinically important, as these exposure estimates donot affect the Cmax and Tmax that are necessary for seizurecontrol.

A limitation of the present study was that it was carriedout in a controlled, adult population without epilepsy, whichdid not include elderly or pediatric individuals. Bioavailabil-ity may vary in these populations.

In conclusion, all tested doses and formulations ofdiazepam in this study were well tolerated with expectedsafety profiles. Although mild to moderate local TEAEswere reported with the intranasal formulation, they didnot affect the use of the product nor did they impactthe overall diazepam exposure. Based on the PK profileand bioavailability, the 20-mg diazepam-NS formulationwas pharmacokinetically comparable to 20 mg diazepam-RG, suggesting that the intranasal formulation may providecaregivers and patients with a more socially acceptable andmore convenient alternative to rectal gel in the acute treat-ment of seizure clusters.

Funding

This study was funded by Neuronex, Inc., a wholly ownedsubsidiary of Acorda Therapeutics, Inc.

Conflict of interest

H.R.H was an employee of Acorda Therapeutics, Inc., withstock options at the time of the study. A.L.R. and E.J.C. areemployees of Acorda Therapeutics, Inc., with stock options.M.R.S. is a consultant for Acorda Therapeutics, Inc., andelectroCore Medical. G.B. was an employee of Neuronex,Inc., a wholly owned subsidiary of Acorda Therapeutics, Inc.,at the time of the study.

Acknowledgments

Authors thank Bagi Ravishankar, PhD, David Squillacote, MD,and Risa Torkin, PhD, for their assistance in manuscriptdevelopment. Editorial assistance was provided by InfusionCommunications and funded by Acorda Therapeutics, Inc.

References

Bhattacharyya, M., Kalra, V., Gulati, S., 2006. Intranasal midazolamvs rectal diazepam in acute childhood seizures. Pediatr. Neurol.34, 355—359.

Bream, G., Leibowitz, M.T., Wargin, W., Abernathy, K., Ward,D., 2013. Assessment of pharmacokinetic linearity and relativebioavailability of a nasal diazepam formulation compared withdiazepam rectal gel in healthy adult subjects. In: Presented atthe 65th Annual Meeting of the American Academy of Neurology,March 16—23, 2013, San Diego, CA.

Cereghino, J.J., 2007. Identification and treatment of acute repeti-tive seizures in children and adults. Curr. Treat. Options Neurol.9, 249—255.

Chen, J.W., Wasterlain, C.G., 2006. Status epilepticus: pathophysi-ology and management in adults. Lancet Neurol. 5, 246—256.

de Haan, G.J., van der Geest, P., Doelman, G., Bertram, E., Edel-broek, P., 2010. A comparison of midazolam nasal spray and

T

W

1211

diazepam rectal solution for the residential treatment of seizureexacerbations. Epilepsia 51, 478—482.

005. DIASTAT® AcuDialTM (diazepam rectal gel) [prescribing infor-mation]. Valeant Pharmaceuticals North America, LLC, Durham,NC.

reifuss, F.E., Rosman, N.P., Cloyd, J.C., Pellock, J.M., Kuzniecky,R.I., Lo, W.D., Matsuo, F., Sharp, G.B., Conry, J.A., Bergen,D.C., Bell, W.E., 1998. A comparison of rectal diazepam geland placebo for acute repetitive seizures. N. Engl. J. Med. 338,1869—1875.

riedman, D., French, J.A., 2012. Clinical trials for therapeuticassessment of antiepileptic drugs in the 21st century: obstaclesand solutions. Lancet Neurol. 11, 827—834.

014. Guidance for industry: bioavailability and bioequivalencestudies for orally administered drug products-general consider-ations, http://www.fda.gov/cder/guidance/index.htm

aut, S.R., 2006. Seizure clustering. Epilepsy Behav. 8, 50—55.olsti, M., Dudley, N., Schunk, J., Adelgais, K., Greenberg, R.,

Olsen, C., Healy, A., Firth, S., Filloux, F., 2010. Intranasal mida-zolam vs rectal diazepam for the home treatment of acuteseizures in pediatric patients with epilepsy. Arch. Pediatr. Ado-lesc. Med. 164, 747—753.

olsti, M., Sill, B.L., Firth, S.D., Filloux, F.M., Joyce, S.M., Fur-nival, R.A., 2007. Prehospital intranasal midazolam for thetreatment of pediatric seizures. Pediatr. Emerg. Care 23, 148—153.

vaturi, V., Kriel, R., Brundage, R., Loewen, G., Mansbach, H.,Cloyd, J., 2013. Bioavailability of intranasal vs. rectal diazepam.Epilepsy Res. 103, 254—261.

riel, R.L., Cloyd, J.C., Hadsall, R.S., Carlson, A.M., Floren, K.L.,Jones-Saete, C.M., 1991. Home use of rectal diazepam for clus-ter and prolonged seizures: efficacy, adverse reactions, qualityof life, and cost analysis. Pediatr. Neurol. 7, 13—17.

riel, R.L., Cloyd, J.C., Pellock, J.M., Mitchell, W.G., Cereghino,J.J., Rosman, N.P., 1999. Rectal diazepam gel for treatmentof acute repetitive seizures. The North American Diastat StudyGroup. Pediatr. Neurol. 20, 282—288.

yrkou, M., Harbord, M., Kyrkou, N., Kay, D., Coulthard, K., 2006.Community use of intranasal midazolam for managing prolongedseizures. J. Intellect. Dev. Disabil. 31, 131—138.

amson, M.J., Sitki-Green, D., Wannarka, G.L., Mesa, M., Andrews,P., Pellock, J., 2011. Pharmacokinetics of diazepam admin-istered intramuscularly by autoinjector versus rectal gel inhealthy subjects: a phase I, randomized, open-label, single-dose, crossover, single-centre study. Clin. Drug Investig. 31,585—597.

illigan, N., Dhillon, S., Richens, A., Oxley, J., 1981. Rectaldiazepam in the treatment of absence status: a pharma-codynamic study. J. Neurol. Neurosurg. Psychiatry 44, 914—917.

illigan, N., Dhillon, S., Oxley, J., Richens, A., 1982. Absorption ofdiazepam from the rectum and its effect on interictal spikes inthe EEG. Epilepsia 23, 323—331.

illigan, N., Dhillon, S., Griffiths, A., Oxley, J., Richens, A., 1984.A clinical trial of single dose rectal and oral administrationof diazepam for the prevention of serial seizures in adultepileptic patients. J. Neurol. Neurosurg. Psychiatry 47, 235—240.

ellock, J.M., 2007. Overview: definitions and classifications ofseizure emergencies. Child Neurol. 22, 9S—13S.

illanpää, M., Schmidt, D., 2008. Seizure clustering during drugtreatment affects seizure outcome and mortality of childhood-onset epilepsy. Brain 131 (Pt 4), 938—944.

atum, W.O., 2002. Adult patient perceptions of emergency rec-

tal medications for refractory seizures. Epilepsy Behav. 3, 535—538.

ilson, M.T., Macleod, S., O’Regan, M.E., 2004. Nasal/buccal mida-zolam use in the community. Arch. Dis. Child. 89, 50—51.