This randomized, double-blind, placebo-controlled, parallel-group study was conducted between September 18, 2008, and February 9, 2009. Sites were selected based on the requirement that participants undergo a 6-hour time zone change. Specific sites were chosen based on their expertise and on logistic considerations, such as being close to an appropriate airport that accommodated private chartered jets and having the capacity required to conduct the study. The protocol was approved by the central institutional review board in the United States and regional independent ethics committees in France, and complied with the International Conference on Harmonisation's Good Clinical Practice Consolidated Guidance11
and applicable laws and regulations. Study participants provided written informed consent at the origination study centers and reconfirmed their consent at the destination study centers.
Participants were screened at 2 visits during a 1- to 8-week period and, if eligible, flew overnight (day 0) on a nonstop, private chartered flight simulating coach-class accommodations from their origination study center in the United States (New York, NY; Atlanta, GA; Columbia, SC; or Crestview Hills, KY) to their destination study center in France (Rouffach or Toulouse). The flight duration was 8.0 to 10.5 hours, and the difference between time zones was 6 hours. Before participants boarded the plane, a breathalyzer test, urine drug screen, and pregnancy test were administered. Passengers were randomly assigned seats, which could not be reclined more than 37°, and were not allowed to consume alcohol or caffeine during the trip. Cabin lights were turned off after dinner and turned back on before breakfast, as is customary on commercial flights. Study participants were not given instructions regarding sleep and may or may not have slept during the flight. They were served a light breakfast approximately 1 hour before landing. After arriving early the following morning, participants were transported by chartered vehicle to the destination study center (arriving at approximately 7:00 am) where they stayed for the entire 3-day study period (days 1-3; ).
During the 3-day study period in France, participants remained indoors (to limit confounding variables, such as access to caffeine, and because of logistic constraints) and slept and ate in accordance with the local time zone. They were exposed to natural light through windows, and the overall degree of light exposure (artificial or natural) was not specified by the protocol, except during nocturnal polysomnography (NPSG) and the Multiple Sleep Latency Test (MSLT). Snacks, decaffeinated drinks, and juices were available throughout the day. Study participants slept in single rooms with lights turned out at 10:00 pm±30 minutes; their wake-up time was 8 hours later.
Participants flew back to their origination study center in the United States on the morning of day 4. A final evaluation was performed after arrival at the origination study center in the United States. Participants were also contacted via telephone approximately 48 hours and 7 days after discharge from the origination study center for follow-up of adverse events.
Men and women (18-65 years old) who previously experienced symptoms consistent with the diagnostic criteria for jet lag disorder (as defined by The International Classification of Sleep Disorders
, Second Edition1
) were enrolled at the 4 origination study centers in the United States. Participants were required to have experienced jet lag symptoms after jet travel (with a time zone change of ≤6 hours) at least once in the past 5 years. All participants were required to report sleeping 6.5 to 9 hours per night, on average, during the month preceding screening, at screening visits, and on the day of departure from the origination city (day 0). In addition, participants had to have a self-reported bedtime between 9:00 pm
and 12:00 am
during the month preceding screening and during the 2 nights before day 0. Major exclusion criteria included traveling across time zones with a 4-hour or greater difference within 2 weeks before the end of the screening period or traveling outside the origination time zone within 1 week before the end of the screening period. Participants were also excluded if they had a history (ie, in the preceding 12 months) or diagnosis of narcolepsy, obstructive sleep apnea, or SWD; any history of hypersomnia, insomnia, or sleep disorder; a history of deep venous thrombosis; or any current diagnosis of a clinically relevant medical disorder. Participants with a history of any psychiatric disorder that would affect study participation or with an Epworth Sleepiness Scale score of 10 or higher, MSLT score lower than 8 minutes, or State and Trait Anxiety Inventory (STAI) score greater than 50 on either of the 2 scales (trait or state anxiety) were excluded. Participants were excluded for taking medications that might influence sleep (eg, hypnotics, melatonin, and stimulants) within 7 days before completing screening and for caffeine intake greater than 300 mg/d within 2 weeks before completing screening. Participants with any history of substance abuse or dependence (except nicotine dependence >5 years ago), any use of nicotine within 3 months, or current alcohol consumption greater than 14 units/week were excluded. Alcohol consumption was prohibited during the study period.
Eligible participants were randomized in a 1:1:1 ratio to receive either 50 mg or 150 mg of armodafinil or placebo once daily. The 150-mg dose was the dose evaluated in relation to the primary efficacy assessment; the 50-mg dose was evaluated as a minimally effective dose in secondary assessments. Armodafinil was provided in tablets of 50 mg/d that were identical in appearance to placebo tablets.
At approximately 8:00 am on each of the 3 laboratory study days, participants were given 3 tablets of study drug (of which 0, 1, or 3 were armodafinil at 50 mg/d), 1 from each of 3 bottles prelabeled with their randomization number. The sponsor generated the randomization code, as well as packaged and labeled the study medication. Randomization was performed in blocks of 3 and stratified according to destination site. The participants, the investigators and their study site personnel, and the study sponsor's personnel were blinded to identification of treatment.
Wakefulness was assessed using 2 prespecified coprimary outcomes, the mean score on the MSLT12
and participants' rating of their overall condition (in relation to their jet lag symptoms) using the Patient Global Impression of Severity (PGI-S),13
both of which were averaged across days 1 and 2 when, on the basis of the number of time zones crossed, participants' sleepiness and jet lag symptoms were expected to be maximal.4,14
Secondary outcomes included scores on the MSLT and PGI-S ratings from individual days 1 to 3, and the Karolinska Sleepiness Scale (KSS) scores averaged across days 1 and 2 and from individual days 1 to 3.
The MSLT is a validated, objective assessment of excessive sleepiness,12
the key characteristic of jet lag as defined by The International Classification of Sleep Disorders
This measure has been used in studies of other circadian sleep disorders, including SWD.10
As specified in the MSLT protocol, participants are instructed to attempt to fall asleep while lying quietly in a darkened room. Four 20-minute (maximum) MSLT sessions were performed. Each MSLT session was performed during a 20-minute period at individual test times: 10:00 am
, 12:00 pm
, 2:00 pm
, and 4:00 pm
at screening visit 2 (to determine eligibility for inclusion in the study) and during the 3 laboratory days. After a participant fell asleep, he or she was immediately awakened and kept awake for the rest of the period.12
If a participant did not fall asleep in the given time, the MSLT was terminated after 20 minutes.12
The PGI-S is a scale that allows individuals to assess their overall condition.13
Participants were asked to assess how they felt overall in relation to their jet lag symptoms. Participants were instructed to rate their overall condition according to 7 categories: normal (shows no sign of illness), borderline ill, mildly (slightly) ill, moderately ill, markedly ill, severely ill, and extremely ill. Study site personnel described the term ill
as follows: “Ill
refers to symptoms of jet lag and any other symptoms. Excessive sleepiness is one of the symptoms of jet lag. Irritability, malaise, gastrointestinal disturbance, and difficulty maintaining one's usual level of performance are other symptoms of jet lag.”
The PGI-S was assessed at screening visit 2 and during the 3 laboratory days, just before the first MSLT session.
The KSS is a validated, participant-rated instrument for measuring excessive sleepiness based on a scale of 1 to 9 (with 1 indicating “very alert” and 9 indicating “very sleepy, great effort to stay awake, fighting sleep”).15
The KSS was administered before each MSLT session at screening visit 2, and before each MSLT session and at bedtime on the 3 laboratory days.
Efficacy data (MSLT, PGI-S, and KSS), collected at screening visit 2, served as the baseline for statistical analyses.
Safety and Tolerability
Mini International Neuropsychiatric Interview,17
and the Columbia Suicide History Scale18
were performed at screening to assess participants' psychiatric status. The state anxiety subscale of the STAI was performed at screening, on each of the 3 laboratory days after the last MSLT session, and at any other time if clinically indicated. Clinical laboratory tests (chemistry, hematology, and urinalysis) and physical examination were performed at screening, and the final evaluation was performed on discharge (day 4). Nocturnal polysomnography was performed at screening and on nights 1 and 2. All safety screening assessments occurred at the first screening visit except NPSG, which was performed at the second (overnight) screening visit. Adverse events and vital signs were monitored on all study days.
Sample size was calculated on the basis of findings from a clinical study of armodafinil for excessive sleepiness associated with SWD.19
A sample size of 133 evaluable participants per treatment group was expected to provide at least 90% power to detect a 1.6-minute difference (assuming a pooled SD of 4.0) in sleep latency on the MSLT and a 1-point mean treatment difference in PGI-S rating (assuming a pooled SD of 2.5). The power to test the joint hypothesis was calculated as at least 81%.
Scores on the MSLT across days 1 and 2 (ie, average of all 8 test sessions across days 1 and 2) were analyzed using analysis of covariance (ANCOVA), with treatment as the main factor and baseline MSLT value as the covariate. A Cochran-Mantel-Haenszel test was used to analyze mean PGI-S ratings on days 1 and 2. The primary comparison was that made between the group who received armodafinil at 150 mg/d and the placebo group for both these outcomes. Type 1 error was controlled at the 0.05 level for the analyses of primary efficacy variables. The secondary outcomes, mean KSS score (average of 4 sessions given before each MSLT) and MSLT scores on days 1, 2, and 3, were analyzed using ANCOVA, with treatment as the main factor and baseline value as covariate. The PGI-S was analyzed using a Cochran-Mantel-Haenszel test on days 1, 2, and 3. Data from NPSG were also analyzed using ANCOVA. Adverse events were analyzed using the Fisher exact test. Statistical analyses completed for all secondary outcome measures were not adjusted for multiplicity; therefore, nominal P values are reported. All other tolerability outcomes are summarized with descriptive statistics.
The safety analysis set included all randomized participants who took at least one dose of study medication, and the efficacy analysis set included all participants in the safety analysis set who had at least one postbaseline, primary efficacy assessment on day 1 or 2.