Clin Drug Invest 2009; 29 (2): 87-100

ORIGINAL RESEARCH ARTICLE 1173-2563/09/0002-0087/$49.95/0

© 2009 Adis Data Information BV. All rights reserved.

Pharmacokinetic Profile of
Armodafinil in Healthy Subjects
Pooled Analysis of Data from Three Randomized Studies
Mona Darwish,1 Mary Kirby,1 Edward T. Hellriegel,2 Ronghua Yang3 and
Philmore Robertson Jr2
1 Clinical Pharmacology Department, Cephalon, Inc., Frazer, Pennsylvania, USA
2 Drug Safety and Disposition Department, Cephalon, Inc., West Chester, Pennsylvania, USA
3 Biometrics Department, Cephalon, Inc., Frazer, Pennsylvania, USA

Abstract Background and objectives: Armodafinil (R-modafinil) is the R- and longer-

lasting isomer of the racemic compound modafinil, a wakefulness-promoting
medication. Armodafinil is eliminated approximately three times more slowly
than the S-isomer of racemic modafinil. Published studies have demonstrated the
efficacy of armodafinil for treating excessive sleepiness associated with obstruc-
tive sleep apnoea, shift work disorder and narcolepsy. The objectives of this study
were to describe the pharmacokinetic profile, tolerability and safety of
armodafinil in healthy subjects.
Methods: Pooled pharmacokinetic data from three separate randomized studies in
119 healthy subjects who received single or multiple (once daily for up to 14 days)
oral doses of armodafinil ranging between 50 and 400 mg were analysed. The
impact of food on the single-dose pharmacokinetic profile of armodafinil was also
assessed in subjects following an overnight fast and after the consumption of a
standard fatty meal.
Results: Armodafinil was readily absorbed and exhibited linear pharmacokinetics
over the 50–400 mg dose range. Peak plasma concentrations were reached around
2 hours after administration in the fasted state. Food had no effect on the overall
bioavailability of armodafinil; however, the peak concentration was delayed by
approximately 2–4 hours. In the multiple-dose study, dose proportionality was
confirmed by linear regression analyses of the log-transformed area under the
plasma concentration versus time curve (AUC) and maximum plasma concentra-
tion (Cmax) values as a function of dose. After reaching the peak, plasma
concentrations of armodafinil declined in a monophasic manner, with a mean
elimination half-life of approximately 15 hours. Steady state appeared to be
reached within 7 days. At steady state, the systemic exposure to armodafinil was
1.8 times that observed after single-dose administration. Armodafinil was general-
88 Darwish et al.

ly well tolerated, the most frequent adverse events being headache, dizziness and
nausea.
Conclusions: In the present analysis, armodafinil exhibited linear pharmaco-
kinetics over the dose range of 50–400 mg. While food affected the rate but not the
extent of absorption, peak plasma concentrations were reached in approximately
2 hours when the drug was taken on an empty stomach. With once-daily dosing,
steady state appeared to be reached within 7 days. After reaching peak plasma
levels, concentrations of armodafinil declined monophasically, with a mean
elimination half-life of around 15 hours. Armodafinil was generally well tolerat-
ed.

Background phasic) clearance of the R-isomer, it was hypothe-

Armodafinil (R-modafinil) is the longer-lived of
the two isomers composing the racemic compound
modafinil, a wakefulness-promoting medication.
Armodafinil has been shown to improve wakeful-
ness significantly in patients with excessive sleepi-
ness associated with obstructive sleep apnoea, shift
work disorder and narcolepsy.[1-4] The longer-lasting
R-isomer was developed in an effort to improve
upon the pharmacokinetic profile of modafinil,
which has effects that are not always maintained
throughout the day with once-daily administration,
making it necessary either to increase the dose and/
or to split the dose between morning and midday
(undermining the convenience of once-daily admin-
istration) in order to obtain effective control of ex-
cessive sleepiness later in the period patients are
awake.[5-7]
Previous studies have shown that the R- and S-
isomers of modafinil have an approximately equipo-
tent pharmacological activity,[8] but different phar-
macokinetic profiles. The elimination half-life (t1/2β)
of R-modafinil (the levorotatory isomer) is approxi-
mately 15 hours in humans, whereas that of S-
modafinil (the dextrorotatory isomer) is approxi-
mately 4 hours. This stereospecific difference in the
clearance of its two isomers results in a biphasic
decline in modafinil plasma concentrations from
peak levels.[9,10] As a result of the slower (i.e. mono-
sized that armodafinil might maintain effective plas-
ma concentrations for longer periods than modafinil
with once-daily oral administration. In a comparison
of the pharmacokinetic and pharmacodynamic ef-
fects of armodafinil and modafinil in healthy sub-
jects undergoing acute sleep deprivation, doses of
armodafinil 200 mg achieved comparable peak plas-
ma concentrations but higher concentrations at 6–14
hours after administration than modafinil 200 mg, as
well as increased wakefulness and attention per-
formance for longer periods post-dosing.[11]
The present analysis was undertaken to further
characterize the pharmacokinetic profile of ar-
modafinil and to examine its tolerability and safety
in healthy volunteers. The pharmacokinetic profile
described in this analysis is derived from data com-
piled from three randomized studies in healthy sub-
jects who received single (with or without food) or
multiple oral doses of armodafinil ranging from
50 mg to 400 mg.
Subjects and Methods
The three pharmacokinetic studies were random-
ized investigations conducted either in the UK (stud-
ies 1 and 2) or the US (study 3). All three were
approved by the independent ethics committees/
institutional review boards at the respective study
sites, and were conducted in accordance with Inter-

© 2009 Adis Data Information BV. All rights reserved. Clin Drug Invest 2009; 29 (2)
Pharmacokinetic Profile of Armodafinil in Healthy Subjects
national Conference on Harmonisation (ICH) good
clinical practice guidelines,[12] and applicable na-
tional and local laws. Written consent was obtained
from each subject before any procedures or assess-
ments were performed and after the aims, methods
and potential hazards had been explained.
The demographic characteristics of the subjects
in the three studies were similar, as were the entry
criteria for enrolment and the methodologies used to
measure the concentrations of armodafinil and
metabolites in plasma and to determine the drug’s
pharmacokinetic parameters. For these reasons,
pooling of data from these studies was considered
acceptable.
Study Design and Dosing
Studies 1 and 2 were randomized, double-blind,
placebo-controlled, parallel-group investigations of
the pharmacokinetics and tolerability of armodafinil
in healthy men. In both studies, randomization was
performed using a randomization code, while blind-
ing was achieved by matching placebo to 50 mg
capsules of the study drug. Each investigator was
accountable for the study drug and placebo through-
out the study.
Subjects enrolled in study 1 were randomized to
receive a single dose of armodafinil 50, 100, 200,
300 or 400 mg or matching placebo in the morning
after an overnight fast. Those who received the
100 mg dose returned to the clinic after 1 week to
receive a second 100 mg dose after consuming a
standard fatty meal.
Subjects enrolled in study 2 were randomized to
receive armodafinil 50, 100, 200, 300 or 400 mg or
matching placebo once daily for 14 consecutive
days. Subjects received armodafinil each morning
after an overnight fast either as inpatients (on days 1,
7 and 14) or outpatients. Blinded assessment of the
safety data for the 200, 300 and 400 mg dose groups
indicated that 400 mg appeared to be an intolerable
dose and this dose was discontinued after 7 days of
© 2009 Adis Data Information BV. All rights reserved.
89
administration. Because 200 and 300 mg doses ap-
peared to be well tolerated, an intermediate dose of
250 mg was added to further evaluate the tolerability
profile of armodafinil. Subjects in this dose group
followed the same experimental procedures as those
used for the initial dose groups.
Study 3 was a randomized, open-label, crossover
investigation to compare the bioavailability of the
highest dosage strength of uncoated tablets planned
for commercialization versus that of an equivalent
dose of multiple 50 mg coated tablets used during
early clinical testing. Healthy men or women re-
ceived a single dose of armodafinil as 5 × 50 mg
coated tablets or as 1 × 250 mg uncoated tablet.
Subjects were randomized to receive either dose in
the morning after an overnight fast. Following a
7-day washout period, the alternative dose was ad-
ministered, again after an overnight fast.
In all three studies, subjects were not permitted to
take any other prescription or over-the-counter
medications (with the exception of paracetamol
[acetaminophen]) or alcohol within 48 hours before
each administration of the study medication or
throughout each pharmacokinetic sampling period.
Subjects
Studies 1 and 2 were to be conducted in men aged
21–40 years who were in good health and had a
body mass index (BMI) ≤30 kg/m2. Exclusion crite-
ria were: (i) clinically significant, uncontrolled
medical conditions; (ii) abnormal laboratory values,
vital signs, ECG or physical examination findings;
(iii) a history of smoking or alcohol or drug abuse,
or excessive consumption of caffeine-containing
beverages; (iv) positive test results for hepatitis B
surface antigen or hepatitis C antibodies; or (v) a
history of HIV-positive status.
Study 3 was to be conducted in men or women
aged 18–45 years who were in good health and had a
BMI ≤30 kg/m2. The exclusion criteria were similar
to those for studies 1 and 2; women enrolled in the
Clin Drug Invest 2009; 29 (2)
90
study were required to be surgically sterile, 2 years
postmenopausal, or using a medically accepted
method of birth control if they were of child-bearing
potential. Steroidal contraceptives had to be used in
conjunction with a barrier method.
Pharmacokinetic Assessments
In all three studies, plasma samples were collect-
ed before armodafinil administration on day 1 and at
prespecified times after administration. In study 1,
plasma samples were collected at 0.5, 1, 1.5, 2, 3, 4,
6, 8, 10, 13, 16, 24, 36, 48, 72 and 96 hours after
drug administration. In study 2 (multiple-dose
study), plasma samples were collected on days 1, 7
and 14 at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, 13, 16 and 24
hours after drug administration, with additional
samples at 48, 72 and 96 hours after the last dose on
day 14. In addition, pre-dose samples for estimation
of trough concentrations were collected on days 5, 6,
7, 12, 13 and 14. In study 3, plasma samples were
collected at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, 13, 16, 24,
36, 48, 60 and 72 hours after administration of the
first dose, and at the same times following the
alternative regimen administered after the 7-day
washout period.
Plasma samples from studies 1 and 2 were
analysed for concentrations of armodafinil and its
two major circulating metabolites, R-modafinil acid
and modafinil sulfone (figure 1), by the Department
of Drug Safety and Disposition of Cephalon, Inc.
(West Chester, PA, USA), using a validated high-
performance liquid chromatography (HPLC)
method with UV detection. The assay utilized liq-
uid-liquid extraction of armodafinil and its two cir-
culating metabolites from 0.20 mL of plasma to
which had previously been added the internal
standard, (phenylthio)acetic acid. After mixing and
centrifugation, the samples were placed on dry ice
until the aqueous layer had frozen. The organic layer
was then poured into a clean glass centrifuge tube,
and the solvent was evaporated under a stream of
© 2009 Adis Data Information BV. All rights reserved.
Darwish et al.
O O
S
N
Armodafinil
O O
O S
O N
S O
O
R-Modafinil acid Modafinil sulfone
Fig. 1. Structural configuration of armodafinil and its two major
circulating metabolites, R-modafinil acid and modafinil sulfone.
N2. The dried residue was reconstituted in 0.20 mL
of acetonitrile : 0.02 mol/L KH2PO4 buffer, pH 2.5,
30 : 70 (v/v), and 30 μL was injected onto a Hyper-
sil® BDS phenyl column (2.1 × 150 mm; 5 μm
particle size) maintained at 35°C. Chromatography
was isocratic, with a mobile phase consisting of a
32 : 68 ratio of methanol : 0.02 mol/L KH2PO4
buffer, pH 2.5 (v/v). Elution during the 24-minute
run was monitored at 225 nm; armodafinil, R-moda-
finil acid, modafinil sulfone and the internal
standard eluted at approximately 12.5, 14.5, 15.5
and 6.8 minutes, respectively. Chiral chromatogra-
phy was not required because there is no intercon-
version of armodafinil or R-modafinil acid with its
corresponding S-isomer. Within-run and between-
run coefficients of variation over the assay range
(0.20–50.0 μg/mL for each analyte) were ≤4.9%
and ≤11.1%, respectively. Within-run accuracy and
between-run accuracy were 87.9–107% and
92.9–101%, respectively.
Plasma samples from study 3 were analysed for
concentrations of armodafinil, R-modafinil acid and
modafinil sulfone by PPD Development (Middle-
ton, WI, USA) using a similar, validated HPLC-UV
Clin Drug Invest 2009; 29 (2)
Pharmacokinetic Profile of Armodafinil in Healthy Subjects
method. Sample processing was conducted the same
way as outlined above, except that the internal
standard was CEP-20253, a structural analogue of
modafinil. Chromatography was performed on a
Zorbax® Eclipse XDB-C18 column (2.1 × 30 mm;
3.5 μm particle size) with monitoring of analyte
elution at 235 nm. Armodafinil, R-modafinil acid,
modafinil sulfone and the internal standard eluted at
approximately 3.6, 5.1, 6.5 and 10.5 minutes, re-
spectively. Assay precision and accuracy over the
assay range (0.20–50.0 μg/mL for each analyte)
were consistent with those reported above.
Noncompartmental pharmacokinetic analysis
was performed for all three studies. Standard phar-
macokinetic parameters of armodafinil, R-modafinil
acid and modafinil sulfone after single and multiple
doses were determined, as applicable, using Win-
Nonlin® (Enterprise Versions 4.0.1 or 4.1; Pharsight
Corporation, Mountain View, CA, USA), including:
(i) the maximum plasma concentration (Cmax);
(ii) time to Cmax (tmax); (iii) minimum (trough)
plasma concentration in a dosing interval (Cmin);
(iv) area under the plasma concentration versus time
curve (AUC) from time zero to infinity (AUC∞) for
a single dose and AUC over one dosing interval (τ)
for the multiple-dose regimen (AUCτ); (v) apparent
Table I. Demographic characteristics of the subjects who received at least one dose of armodafinil (range 50–400 mg) or placebo in the
three studies
Characteristic Armodafinil
(n = 97)
Age (y) [mean ± SD] 28.3 ± 6.6
Sex [n (%)]
male 85 (88)
female 12 (12)
Race [n (%)]
White 72 (74)
Black 7 (7)
Asian 1 (1)
other 17 (18)
Weight (kg) [mean ± SD] 77 ± 10.8
Height (cm) [mean ± SD] 176.6 ± 8.7
BMI (kg/m2) [mean ± SD] 24.7 ± 2.9
BMI = body mass index.
© 2009 Adis Data Information BV. All rights reserved.
91
total oral clearance (CL/F); (vi) t1/2β; and (vii) appar-
ent volume of distribution (Vz/F). In study 2, the
steady state and observed accumulation ratios (Rss
and Robs, respectively) were determined as the ratio
of AUCτ on days 7 or 14 to AUC∞ on day 1 (Rss) or
to AUC from time zero to 24 hours (AUC24) on day
1 (Robs).
Where possible, standard pharmacokinetic para-
meters of R-modafinil acid and modafinil sulfone
were also determined in all three studies. Dose-
normalization of pharmacokinetic parameters for
armodafinil and its metabolites to a 50 mg dose was
performed through a proportional adjustment of the
values for each subject by multiplying by the appro-
priate dose-normalization factor (i.e. 50 mg/dose
from which the value was derived) and then deter-
mining the mean value.
Tolerability/Safety Assessments
The tolerability of armodafinil was assessed in
each study by an adverse event inquiry each time a
blood sample was collected. Safety was evaluated
by monitoring clinical laboratory test results, vital
sign measurements, ECG and physical examination
findings. The latter assessments were performed at
baseline and at regular intervals after armodafinil
Placebo
(n = 22)
27.4 ± 5.9
22 (100)
0
21 (96)
1 (4)
0
0
76 ± 7.8
176.8 ± 6.9
24.3 ± 2.2
Clin Drug Invest 2009; 29 (2)
 92
Table II. Mean (± SD) maximum plasma concentrations (Cmax) and areas under the plasma concentration vs time curve (AUC) of armodafinil following single and multiple doses
administered in the fasting state in the three studies

Parameter/ Study 1a (single Study 2b (multiple ascending doses) Study 3c (single-dose/ bioequivalence)
armodafinil ascending doses) [n = 34] [n = 25]
dose (mg) [n = 30] day 1 day 7 day 14 5 × 50 mg 1 × 250 mg
Cmax (μg/mL)

50 1.1 ± 0.4 1.3 ± 0.2 1.8 ± 0.2 1.8 ± 0.1

100 2.4 ± 0.4 2.6 ± 0.4 4.0 ± 0.7 4.0 ± 0.9

200 4.1 ± 1.1 4.5 ± 1.5 7.4 ± 2.2 7.4 ± 1.8

250 DNA 5.9 ± 0.7 9.2 ± 0.7 10.5 ± 2.4 8.5 ± 1.3 8.5 ± 1.7d

© 2009 Adis Data Information BV. All rights reserved.

300 7.0 ± 1.4 6.5 ± 1.1 10.9 ± 1.3 10.0 ± 1.0

400 6.9 ± 0.9 9.7 ± 1.8 13.4 ± 5.3 DNA

AUC (μg • h/mL)e

50 15.9 ± 3.7 21.3 ± 7.4 25.4 ± 4.1 23.4 ± 3.4

100 40.6 ± 7.4 41.8 ± 6.2 54.2 ± 8.2 56.2 ± 8.9

200 75.9 ± 17.9 91.9 ± 33.0 111.8 ± 39.4 105.9 ± 25.0

250 DNA 129.2 ± 15.0 148.3 ± 9.6 136.1 ± 8.2 148.6 ± 24.3 144.7 ± 26.3f

300 146.0 ± 40.3 139.6 ± 9.5 165.4 ± 13.8 150.4 ± 12.7

400 142.9 ± 16.3 200.1 ± 52.8 189.5 ± 77.8 DNA

a Six subjects in each dose group.

b Six subjects in each dose group with the exception of the 400 mg dose group (n = 4). The 400 mg dose group was discontinued after day 7 assessments.

c Crossover study.

d Geometric mean ratio of Cmax (expressed as a percentage) for 1 × 250 mg tablet (test tablet) vs 5 × 50 mg tablets (reference formulation) = 98.5% (90% CI 93.4, 103.9).

e AUC∞ for single doses (day 1); AUCτ for multiple doses (days 7 and 14).

f Geometric mean ratio of AUC (expressed as a percentage) for 1 × 250 mg tablet (test tablet) vs 5 × 50 mg tablets (reference formulation) = 98.5% (90% CI 93.4, 103.9).

AUC∞ = AUC from time zero to infinity; AUCτ = AUC over one dosing interval (τ); DNA = dose not administered.

Clin Drug Invest 2009; 29 (2)

Darwish et al.
Pharmacokinetic Profile of Armodafinil in Healthy Subjects 93

a
16 50 mg (n = 6)
sion 7 of the Medical Dictionary for Regulatory
100 mg (n = 6) Activities (MedDRA)[14] in study 3.
Mean armodafinil plasma concentration (μg/mL)

200 mg (n = 6)
14
250 mg (n = 6)
300 mg (n = 6) Statistical Analysis
12 400 mg (n = 4)

10 Descriptive statistics were used to summarize the

pharmacokinetic, tolerability and safety parameters.
8 Pooled pharmacokinetic parameters were used to
characterize the single-dose pharmacokinetics of
6
armodafinil and its metabolites (R-modafinil acid
4 and modafinil sulfone) across all three studies; mul-
tiple-dose parameters were obtained only in study 2.
2
For study 3, in which each subject received
0 armodafinil twice, the observations for each subject
0 4 8 12 16 20 24 were averaged such that each subject within the
Time after dose administration (h)
study contributed only one observation to the pooled
b mean values following the administration of
16
armodafinil. The summaries of combined pharma-
Mean armodafinil plasma concentration (μg/mL)

14 cokinetic parameters included all subjects with data

that were adequate to fully characterize the pharma-
12 cokinetic profile using a noncompartmental ap-
10
proach.
Confirmation of steady state was evaluated via
8 analysis of plasma concentrations of armodafinil
normalized to a 50 mg dose on days 1, 7 and 14 of
6
administration in subjects participating in the multi-
4 ple-dose study. Attainment of steady state was fur-
ther assessed for the 50, 100, 200, 250 and 300 mg
2
doses by calculating the difference of log-trans-
0 formed AUCτ and Cmax values between days 7 and
0 4 8 12 16 20 24 14 at each dose level via ANOVA, determining the
Time after dose administration (h)
mean difference of the log value and the 90% CI
Fig. 2. Mean (± standard error of the mean) armodafinil plasma
concentrations on (a) day 1 and (b) day 7 in the multiple-dose study values at each dose, and then converting these val-
(n = 34) following doses of 50–400 mg in healthy men. ues to geometric mean ratios and their 90% CIs.
90% CIs within the range 80–125% were taken as
administration (i.e. at each pharmacokinetic sample indicating equivalence of the day 7 and day 14
collection in the case of vital signs and ECG, and at values at each dose and the attainment of steady
the conclusion of the study in the case of clinical state.
laboratory tests and physical examinations). Ad- Dose-proportionality was assessed in study 2 by
verse events were classified according to the Coding visual examination of the magnitude of increase in
Symbols for a Thesaurus of Adverse Reaction the mean values of AUC (AUC∞ on day 1; AUCτ on
Terms (COSTART)[13] in studies 1 and 2, and ver- days 7 and 14) and Cmax relative to the magnitude of

© 2009 Adis Data Information BV. All rights reserved. Clin Drug Invest 2009; 29 (2)
94 Darwish et al.

a
20 350 Individual
Mean
300

Armodafinil AUC∞ (μg • h/mL)

16
Armodafinil Cmax (μg/mL)

250

12
200

8 150

100
4
50

0 0
0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400
Armodafinil dose (mg) Armodafinil dose (mg)

b
20 350

300
16
Armodafinil AUCτ (μg • h/mL)
Armodafinil Cmax (μg/mL)

250

12
200

150
8

100
4
50

0 0
0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400
Armodafinil dose (mg) Armodafinil dose (mg)
Fig. 3. Individual and mean area under the concentration vs time curve (AUC) from time zero to infinity (AUC∞), AUC over one dosing
interval (AUCτ), and maximum plasma concentration (Cmax) values of armodafinil following doses of 50, 100, 200, 250, 300 and 400 mg on
(a) day 1 and (b) day 7 in the multiple-dose study in healthy men. The lines represent the best fit of the mean data and were obtained by
linear regression analysis of AUC and Cmax data over the dose range.

increase in armodafinil dose. In addition, the expo- Results

sure-dose relationship was further evaluated using
Of 256 subjects screened for participation in the
regression analyses of the log-transformed AUC and three studies, 119 met the inclusion criteria and were
Cmax values as the dependent variable and the log randomized to receive armodafinil (either as a single
value of dose as the independent variable on days 1, dose or as multiple doses over 14 days; n = 97) or
placebo (n = 22); 40 subjects were randomized in
7 and 14. Dose-proportionality was indicated when
study 1, 49 in study 2, and 30 in study 3. All 119
the 90% CI of the slope fell within the reference
subjects received at least one dose of study med-
range of 0.8927–1.1073 for days 1 and 7 and ication and composed the safety analysis set. The
0.8755–1.1245 for day 14.[15] demographic features of the safety analysis set are

© 2009 Adis Data Information BV. All rights reserved. Clin Drug Invest 2009; 29 (2)
Pharmacokinetic Profile of Armodafinil in Healthy Subjects
shown in table I. The mean ages, bodyweight, height
and BMI of subjects who received the various doses
of armodafinil were similar to those of subjects who
received placebo. The pharmacokinetic analysis set,
which included all subjects in the safety analysis set
who had at least one complete pharmacokinetic as-
sessment, comprised 115 subjects (93 armodafinil,
22 placebo).
Of the 119 randomized subjects, 103 (85
armodafinil, 18 placebo) were considered ‘study
completers’, having concluded all phases of their
respective studies. 105 subjects (87 armodafinil, 18
placebo) completed all pharmacokinetic assess-
ments and were considered ‘pharmacokinetic com-
pleters’; this number included two subjects who
subsequently missed the final safety assessment.
The 14 subjects who did not complete the pharma-
cokinetic assessments included three who were
withdrawn because of adverse events, six in the
400 mg dose group in whom dosing was halted after
day 7, one subject who withdrew consent, one who
was a protocol violator, and three who were with-
drawn for noncompliance or other reasons.
Table III. Slopes and 90% CIs for dose-proportionality using log-
transformed values of areas under the plasma concentration vs
time curve (AUC) and maximum plasma concentrations (Cmax) as a
function of dose following single and multiple doses in study 2
Parameter Slope 90% CI
Day 1a
AUC∞ (μg • h/mL) 1.10 1.0085, 1.1933
Cmax (μg/mL) 0.93 0.8565, 1.0033
Day 7a
AUCτ (μg • h/mL) 1.01 0.9101, 1.1022
Cmax (μg/mL) 0.95 0.8623, 1.0379
Day 14b
AUCτ (μg • h/mL) 1.04 0.9684, 1.1142
Cmax (μg/mL) 1.00 0.9135, 1.0813
a Reference 90% CI range for slope is 0.8927–1.1073 as dose
range was 50–400 mg on days 1 and 7.
b Reference 90% CI range for slope is 0.8755–1.1245 as dose
range was 50–300 mg on day 14.
AUC∞ = AUC from time zero to infinity; AUCτ = AUC over one
dosing interval (τ).
© 2009 Adis Data Information BV. All rights reserved.
95
Pharmacokinetics of Armodafinil
Dose-Proportionality
Dose-proportionality of armodafinil was as-
sessed across the range of 50–400 mg on days 1 and
7 and 50–300 mg on day 14 (table II). The
pharmacokinetics, after both single and multiple
doses, appeared to be linear, based on the approxi-
mately dose-proportional increases in AUC and
Cmax over the dose ranges studied (figure 2 and
figure 3). The slopes, calculated by linear regression
analyses of log-transformed AUC and Cmax values
as a function of log-transformed dose value, on days
1, 7 and 14 fell between 0.9 and 1.1 (table III).
Although this pharmacokinetic study was not pow-
ered to support extensive statistical analysis, the
90% CIs of the slopes for AUCτ on day 7 and for
Cmax and AUCτ on day 14 were found to fall within
the corresponding reference ranges. The 90% CIs of
the slopes for these parameters on day 1 and for
Cmax on day 7 were only marginally outside the
reference range.
Mean Pharmacokinetic Parameters Following
Single and Multiple Doses
Mean pharmacokinetic parameters of ar-
modafinil following single and multiple doses de-
rived from all three studies and normalized for a
50 mg dose are shown in table IV. These data
confirm that armodafinil is readily absorbed after
oral administration, and that Cmax is achieved ap-
proximately 2 hours after single- or multiple-dose
administration in fasting subjects. After reaching
peak, plasma concentrations declined in a relatively
slow, apparently monophasic manner, with mean
t1/2β and CL/F values of approximately 14 hours and
39 mL/min, respectively, after single doses, and
17 hours and 33 mL/min, respectively, after 14 days
of administration. The mean Vz/F was approximate-
ly 42 L after single doses and 47 L after 14 days of
administration. These values approximate the esti-
mated total body water in humans, indicating that
Clin Drug Invest 2009; 29 (2)
96 Darwish et al.

Table IV. Pharmacokinetic parameters for armodafinil and its metabolites R-modafinil acid and modafinil sulfone following single and
multiple doses, normalized to a 50 mg dose of armodafinil (combined data)a
Pharmacokinetic parameter Single-dose values Multiple-dose values
day 7 day 14
Armodafinil
n 93 34 30
Cmax (μg/mL) 1.3 ± 0.4 1.8 ± 0.4 1.9 ± 0.4
tmax (h)b 1.5 (0.5–6.0) 2 (0.5–6.0) 1.8 (0.0–4.0)
AUC∞ (μg • h/mL) 24.1 ± 6.9 NA NA
AUCτ (μg • h/mL) NA 27.1 ± 5.8 26.1 ± 4.0
CL/F (mL/min) 38.6 ± 9.9 32.4 ± 8.7 32.7 ± 5.2
t1/2β (h) 13.8 ± 3.3c 15.3 ± 3.0d 16.9 ± 3.1
Vz/F (L) 42.4 ± 12.5 54.5 ± 31.4d 47.4 ± 8.7
Rss NA 1.2 ± 0.2 1.2 ± 0.2
Robs NA 1.8 ± 0.3 1.7 ± 0.2

R-Modafinil acid
n 57 22 18
Cmax (μg/mL) 0.1 ± 0.0 0.1 ± 0.0 0.1 ± 0.0
tmax (h)b 3.0 (1.0–6.0) 2.0 (0.0–24.0) 2.0 (0.0–4.0)
AUC∞ (μg • h/mL) 2.7 ± 0.8 NA NA
AUCτ (μg • h/mL) NA 2.1 ± 0.7 1.9 ± 0.7
t1/2β (h) 15.3 ± 4.3 17.4 ± 5.5d 20.7 ± 9.1e

Modafinil sulfone
n 39 ND 24
Cmax (μg/mL) 0.1 ± 0.0 ND 0.7 ± 0.5
tmax (h)b 24.0 (13.0–42.0) ND 3.0 (0.0–16.0)
AUC∞ (μg • h/mL) 8.0 ± 2.3 ND NA
AUCτ (μg • h/mL) NA ND 14.7 ± 10.2
t1/2β (h) 37.6 ± 12.8 ND 39.2 ± 17.3
a Mean values (± SD), except where otherwise stated.
b Median values (range).
c Data from 87 subjects.
d Data from four subjects.
e Data from 15 subjects.
AUC = area under the plasma concentration vs time curve; AUC∞ = AUC from time zero to infinity; AUCτ = AUC over one dosing interval (τ);
Cmax = maximum plasma concentration; CL/F = apparent total oral clearance; NA = not applicable; ND = not done; Robs = observed
accumulation ratio (ratio of AUCτ on day 7 or 14 to AUC from time zero to 24 hours on day 1); Rss = steady-state accumulation ratio (ratio of
AUCτ on day 7 or 14 to AUC∞ on day 1); t1/2β = terminal elimination half-life; tmax = time to Cmax; Vz/F = apparent volume of distribution.

armodafinil is widely distributed outside the vascu-
lature, but does not appear to have high affinity for
tissues.
Plasma concentrations of armodafinil normalized
to a 50 mg dose on days 7 and 14 in the multiple-
dose study were similar, indicating that steady state
had been attained within 7 days of once-daily ad-
ministration (figure 4). Further confirmation of
stable status was provided by the 90% CIs for geo-
metric mean AUCτ ratios on day 7 versus day 14 for
the armodafinil 50, 100, 200, 250 and 300 mg doses.
For all five doses, the 90% CI bounds were well
within the range 80–125% (data not shown). When
this analysis was performed for Cmax, the 90% CI
values were also well within the range 80–125% for
all doses except the 250 mg dose (90% CI 74.9,
106.4 at this dose level).

© 2009 Adis Data Information BV. All rights reserved. Clin Drug Invest 2009; 29 (2)
Pharmacokinetic Profile of Armodafinil in Healthy Subjects 97
Mean armodafinil plasma concentration (μg/mL)

2.5 Day 1 (n = 34)

Day 7 (n = 34) Plasma concentrations of armodafinil after multi-
Day 14 (n = 30) ple-dose administration were increased compared
2.0
with those measured after single-dose administra-
tion (table IV), and the mean Robs was 1.8 on day 7
1.5
and 1.7 on day 14. The mean Rss was approximately
1.2 at both times, indicating that AUC values were
1.0 only slightly larger on days 7 and 14 than would be
predicted on the basis of single-dose data on day 1.
0.5
Influence of Food on Absorption
0.0 In the six subjects enrolled in study 1 who re-
0 2 4 6 8 10 12 14 16 18 20 22 24
ceived single doses of armodafinil 100 mg both in
Time after dose administration (h)
Fig. 4. Mean (± standard error of the mean) plasma concentrations
the fasted state and after consumption of a fatty
of armodafinil normalized to a 50 mg dose over a 24-hour period on meal, absorption of the drug was slowed by the
days 1, 7 and 14 of administration in the multiple-dose study in
presence of food. Median tmax was approximately
healthy men. The 400 mg dose was discontinued after day 7.
4 hours longer when armodafinil was administered
after a fatty meal compared with that following
Mean armodafinil plasma concentration (μg/mL)

administration in the fasted state (6.0 vs 2.3 hours,

2.5 Fed respectively) [figure 5]. However, food had no ef-
Fasted
fect on the overall bioavailability of armodafinil
2
compared with fasting (Cmax 2.2 ± 0.1 vs 2.4 ± 0.4
1.5
μg/mL, respectively; AUC∞ 43.8 ± 8.2 vs 40.6 ± 7.4
μg • h/mL, respectively) [table V].
1 Although the effect of food following multiple-
dose administration was not studied, superimposi-
0.5
tion of the single-dose data indicated that a delay of
0
approximately 2 hours in median tmax could be
0 6 12 18 24 30 36 anticipated at steady state in the fed state compared
Time after dose administration (h)
with the fasted state (4.1 vs 1.9 hours, respectively).
Fig. 5. Mean (± standard error of the mean) plasma concentrations
of armodafinil following doses of 100 mg in the fed and fasted
states in the single-dose study in healthy men (n = 6). Pharmacokinetics of R-Modafinil Acid and
Modafinil Sulfone
The pharmacokinetic parameters determined for
the two principal circulating metabolites of ar-

Table V. Pharmacokinetic parameters of armodafinil following single 100 mg doses administered in the fasting state and after consumption
of a standard fatty meal
Subjects n Cmax (μg/mL)a tmax (h)b AUC∞ (μg • h/mL)a
Fasted 6 2.4 ± 0.4 2.3 (0.5–4.0) 40.6 ± 7.4
Fed 6 2.2 ± 0.1 6.0 (3.0–6.0) 43.8 ± 8.2
a Mean values (± SD).
b Median values (range).
AUC∞ = area under the plasma concentration vs time curve from time zero to infinity; Cmax = maximum plasma concentration; tmax = time to
Cmax.

© 2009 Adis Data Information BV. All rights reserved. Clin Drug Invest 2009; 29 (2)
98 Darwish et al.

Table VI. Adverse events [n (%)] occurring in ≥5% of subjects

Tolerability and Safety
receiving armodafinil 50–400 mg or placebo (combined data)
Adverse event Armodafinil Placebo
(preferred term) (n = 97) (n = 22) Single doses of armodafinil were well tolerated at
Headache 41 (42) 2 (9) doses up to 400 mg. In subjects given multiple
Dizziness 21 (22) 1 (5) doses, doses up to 300 mg/day were well tolerated.
Nausea 18 (19) 1 (5)
The most frequent adverse events observed with
Insomnia 13 (13) 0
Palpitation 12 (12) 0
armodafinil were headache, dizziness and nausea,
Anxiety 10 (10) 0 which occurred in 42%, 22% and 19% of subjects,
Fatigue 9 (9) 0 respectively (table VI); these adverse events ap-
Heart rate increased 7 (7) 0 peared to be dose related, being more common with
Rhinitis 6 (6) 1 (5)
doses ≥250 mg. Most adverse events occurring with
Anorexia 6 (6) 0
Vasodilatation 6 (6) 0
armodafinil were mild to moderate in nature and did
Asthenia 5 (5) 2 (9) not result in study withdrawal. The exception was
Pharyngitis 5 (5) 2 (9) the 400 mg group (multiple-dose), from which two
Energy increased 5 (5) 0 subjects withdrew prior to the day 7 dose. Because
Rash 2 (2) 1 (5)
of the high number of adverse events in the
Cough increased 1 (1) 1 (5)
Pain 1 (1) 1 (5)
armodafinil-treated subjects in this group, dosing of
ALT increased 1 (1) 1 (5) the other six subjects scheduled to receive 400 mg,
ALT = alanine aminotransferase. including those receiving matched placebo, was
halted after day 7.
modafinil are summarized in table IV. Systemic No clinically meaningful changes in mean labor-
exposure to R-modafinil acid, as assessed by dose- atory test values, physical examination or ECG find-
normalized, mean AUC values, was approximately ings were detected, and there were no overall trends
11% of that of the parent drug after single-dose suggesting a relationship to changes in mean vital
administration, and approximately 7% after multi- sign parameters.
ple-dose administration. The median tmax and mean
apparent t1/2β values for this metabolite were similar Discussion
to those of the parent drug. R-Modafinil acid
achieved steady state within 7 days after initiation of Oral administration of modafinil results in signif-
icant differences in circulating concentrations of the
dosing with armodafinil.
two isomers of the drug, R and S, because of stereo-
For modafinil sulfone, the dose-normalized mean specific differences in their elimination. For this
AUC values (table IV) suggested that systemic ex- reason, the plasma concentrations of R-modafinil
posure to this metabolite was approximately 33% of (armodafinil) may be up to three times higher than
that of the parent drug after single-dose administra- those of S-modafinil.[9] Preclinical studies have
tion, and approximately 56% after multiple-dose shown that the isomers have approximately equipo-
administration. The mean t1/2β values for modafinil tent pharmacological effects.[8] This suggests that
sulfone were considerably higher than those of the the majority of the effects observed with modafinil
parent drug, i.e. approximately 39 hours on day 14 are attributable to the longer-lasting R-isomer.
versus 17 hours for armodafinil. Modafinil sulfone When given once daily in a previous study to
was not at steady state on day 7 but appeared to be at healthy subjects who underwent acute sleep depri-
steady state by day 14. vation, single armodafinil 200 mg doses produced

© 2009 Adis Data Information BV. All rights reserved. Clin Drug Invest 2009; 29 (2)
Pharmacokinetic Profile of Armodafinil in Healthy Subjects
comparable Cmax values but higher concentrations
later in the day (i.e. at 6–14 hours after administra-
tion) compared with modafinil 200 mg.[11] In addi-
tion, armodafinil 200 mg improved wakefulness and
sustained attention for a longer period than modafi-
nil 200 mg, as demonstrated objectively via the
Maintenance of Wakefulness Test and Psychomotor
Vigilance Task.[11] These findings suggest that be-
cause of the presence of the short-lived S-isomer in
modafinil, the Cmax for modafinil would need to be
substantially higher than that of armodafinil to
maintain similar late-day exposure.
The present analysis involved pooling data from
three separate, randomized investigations of ar-
modafinil in healthy volunteers in order to assess the
single- and multiple-dose pharmacokinetics and
safety/tolerability of armodafinil and its two circu-
lating metabolites, R-modafinil acid and modafinil
sulfone. Armodafinil displayed linear pharmaco-
kinetics over the 50–400 mg dose range, with Cmax
occurring approximately 2 hours after administra-
tion in fasting subjects. Dose-proportionality at
steady state was confirmed by linear regression
analyses of the log-transformed AUC and Cmax val-
ues as a function of dose in the multiple-dose study.
Compared with the biphasic decline from peak
shown by modafinil,[9,10] plasma concentrations of
armodafinil declined from peak in a monophasic
manner, with a mean t1/2β of approximately 15 hours.
When assessed in subjects who had eaten a fatty
meal, tmax occurred later, but systemic exposure
(Cmax and AUC) appeared to be only minimally
affected, suggesting that the presence of food may
slow the rate of absorption of armodafinil without
materially affecting the extent of absorption.
The results of the current analysis are consistent
with those of an earlier study by Dinges et al.,[11] in
which single oral doses of armodafinil in the range
100–300 mg were administered to 71 healthy male
volunteers aged 18–40 years. Armodafinil displayed
linear pharmacokinetics over this dose range, as it
© 2009 Adis Data Information BV. All rights reserved.
99
did in the present analysis, and Cmax values were
comparable to those noted in our investigation. Be-
cause subjects in the study by Dinges et al.[11] re-
ceived armodafinil doses immediately followed by a
standardized meal, tmax occurred later than those
reported for subjects who had fasted (5–6.5 hours vs
1.5 hours), but were comparable to those found in
subjects who received an armodafinil dose of
100 mg after a fatty meal in this analysis.
The results of the present analysis are also consis-
tent with those previously published for armodafinil
when plasma samples were analysed using chiral
bioanalytical methods after administration of moda-
finil. In healthy male subjects who received modafi-
nil 400 mg once daily for 7 days,[9] pharmacokinetic
parameters for armodafinil were similar to those in
subjects who received armodafinil 200 mg once
daily in the current multiple-dose study. On day 7,
R-modafinil Cmax, CL/F and AUCτ values were
6.8 μg/mL, 34.0 mL/min and 104.0 μg • h/mL, re-
spectively, after administration of modafinil 400 mg
in the study by Wong et al.[9] compared with 7.4 μg/
mL, 32.4 mL/min and 111.8 μg • h/mL after
armodafinil 200 mg in the current multiple-dose
study. The similarity of the results for armodafinil
after administration with those obtained after ad-
ministration of modafinil suggests that there is little,
if any, pharmacokinetic interaction between the iso-
mers in humans.
Analysis of the pooled safety data from the three
randomized studies showed that armodafinil was
generally well tolerated after single and multiple
doses, particularly at doses <250 mg/day, in healthy
subjects. The most frequent adverse events noted
with armodafinil were headache, dizziness and nau-
sea, which appeared to be dose-dependent.
Conclusions
In the present analysis, armodafinil displayed
linear pharmacokinetics over the dose range of
50–400 mg. Cmax was reached in approximately
Clin Drug Invest 2009; 29 (2)
100
2 hours when the drug was taken on an empty
stomach; food affected the rate but not the extent of
absorption. Steady state appeared to be reached
within 7 days after commencing once-daily admin-
istration. After reaching Cmax, concentrations of
armodafinil exhibited an apparent monophasic de-
cline, with a mean t1/2β of around 15 hours.
Armodafinil was generally well tolerated in these
healthy volunteers; adverse events were dependent
upon dose.
Acknowledgements
Cephalon, Inc., Frazer, PA, USA, sponsored this pooled
analysis and, with guidance from the authors, was responsible
for the design of the study and data analysis. The authors were
employees of Cephalon, Inc. at the time of the analyses and
manuscript development. The authors wish to acknowledge
the study investigators who enrolled subjects and collected
data under the direction of the sponsor: Stephen Freestone
(Inveresk Research [now known as Charles River Laborato-
ries], Tranent, Scotland, UK; studies 1 and 2), and Dennis
Swearingen (MDS Pharma Services, Phoenix, AZ, USA;
study 3). John G. Jiang, a former employee of Cephalon, Inc.,
contributed to the analysis of the data. Virginia Schobel, an
employee of Cephalon, Inc., and Thomson Reuters (Hor-
sham, PA, USA) provided editorial assistance to the authors.
All authors had full access to the data and contributed to data
interpretation and preparation of this report. The correspond-
ing author had final responsibility for the decision to submit
this report for publication.
References
1. Roth T, White D, Schmidt-Nowara W, et al. Effects of
armodafinil in the treatment of residual excessive sleepiness
associated with obstructive sleep apnea/hypopnea syndrome: a
12-week, multicenter, double-blind, randomized, placebo-con-
trolled study in nCPAP-adherent adults. Clin Ther 2006; 28
(5): 689-706
2. Hirshkowitz M, Black JE, Wesnes K, et al. Adjunct armodafinil
improves wakefulness and memory in obstructive sleep apnea/
hypopnea syndrome. Respir Med 2007; 101 (3): 616-27
3. Drake C, Walsh J, Roth T. Armodafinil improves sleep latency
in patients with shift work sleep disorder [abstract no. 0189].
Sleep 2006; 29 Suppl.: A64
4. Harsh JR, Hayduk R, Rosenberg R, et al. The efficacy and safety
of armodafinil as treatment for adults with excessive sleepiness
© 2009 Adis Data Information BV. All rights reserved.
Darwish et al.
associated with narcolepsy. Curr Med Res Opin 2006; 22 (4):
761-74
5. Schwartz JR, Nelson MT, Schwartz ER, et al. Effects of modafi-
nil on wakefulness and executive function in patients with
narcolepsy experiencing late-day sleepiness. Clin Neuro-
pharmacol 2004; 27 (2): 74-9
6. Schwartz JR, Feldman NT, Bogan RK, et al. Dosing regimen
effects of modafinil for improving daytime wakefulness in
patients with narcolepsy. Clin Neuropharmacol 2003; 26 (5):
252-7
7. Schwartz JR, Feldman NT, Bogan RK. Dose effects of modafi-
nil in sustaining wakefulness in narcolepsy patients with
residual evening sleepiness. J Neuropsychiatry Clin Neurosci
2005; 17 (3): 405-12
8. Data on file, Cephalon, Inc., West Chester (PA), USA, 1998
9. Wong YN, Simcoe D, Hartman LN, et al. A double-blind,
placebo-controlled, ascending-dose evaluation of the
pharmacokinetics and tolerability of modafinil tablets in
healthy male volunteers. J Clin Pharmacol 1999; 39 (1): 30-40
10. Wong YN, King SP, Simcoe D, et al. Open-label, single-dose
pharmacokinetic study of modafinil tablets: influence of age
and gender in normal subjects. J Clin Pharmacol 1999; 39 (3):
281-8
11. Dinges DF, Arora S, Darwish M, et al. Pharmacodynamic
effects on alertness of single doses of armodafinil in healthy
subjects during a nocturnal period of acute sleep loss. Curr
Med Res Opin 2006; 22 (1): 159-67
12. Expert Working Group (Efficacy) of the International Confer-
ence on Harmonisation of Technical Requirements for Regis-
tration of Pharmaceuticals for Human Use (ICH). Good Clin-
ical Practice: Consolidated Guidance E6(R1). ICH, 1996 [on-
line]. Available from URL: http://www.fda.gov/cder/
guidance/959fnl.pdf [Accessed 2008 Nov 25]
13. United States Food and Drug Administration, Center for Drug
Evaluation and Research. Coding symbols for thesaurus of
adverse reaction terms. 5th ed. Rockville (MD): Department of
Health and Human Services, Center for Drug Evaluation and
Research, Office of Epidemiology Biostatistics, 1995
14. International Federation of Pharmaceutical Manufacturers As-
sociations. Medical dictionary for regulatory activities (Med-
DRA): version 7.0. Chantilly (VA): MedDRA MSSO, 2004
Mar
15. Smith BP, Vandenhende FR, DeSante KA, et al. Confidence
interval criteria for assessment of dose proportionality. Pharm
Res 2000; 17 (10): 1278-83
Correspondence: Dr Mona Darwish, Clinical Pharmacology
Department, Cephalon Inc., 41 Moores Road, Frazer, PA
19355, USA.
E-mail: mdarwish@cephalon.com
Clin Drug Invest 2009; 29 (2)