Study to Evaluate Armodafinil Treatment in Improving Prefrontal Cortical Activation and Working Memory Performance

Excessive Sleepiness Clinical Trial

Official title:

Double-Blind, Placebo-Controlled, Functional Neuroimaging Study of Armodafinil (200 mg/Day) on Prefrontal Cortical Activation in Patients With Residual Excessive Sleepiness Associated With Obstructive Sleep Apnea/Hypopnea

Clinical Trial Details — Status: Completed

Administrative data

• NCT number: NCT00711516
• Other study ID #: C10953/4026/AP/US
• Status: Completed
• Phase: Phase 4
• First received: July 8, 2008
• Last updated: July 12, 2013
• Start date: September 2008
• Est. completion date: October 2009

Study information

• Verified date: July 2013
• Source: Teva Pharmaceutical Industries
• Contact: n/a
• Is FDA regulated: No
• Health authority: United States: Food and Drug Administration
• Study type: Interventional

Clinical Trial Summary

The primary objective of this study is to determine whether treatment with armodafinil will provide improvements in prefrontal cortical activation in patients with OSAHS (Obstructive Sleep Apnea/Hypopnea Syndrome) who have residual sleepiness despite receiving nCPAP therapy.

Recruitment information / eligibility

• Status: Completed
• Enrollment: 40
• Est. completion date: October 2009
• Est. primary completion date: September 2009
• Accepts healthy volunteers: No
• Gender: Both
• Age group: 18 Years to 60 Years

Eligibility

Inclusion Criteria:

• Patient has a current diagnosis of OSAHS and has a complaint of excessive sleepiness despite effective nCPAP therapy.

• Patient has excessive sleepiness as evidenced by a mean sleep latency of less than 8 minutes, as determined by the MSLT.

• Patient has an ESS score of 10 or more at the initial screening visit.

• Patient has a habitual sleep time beginning no earlier than 2100 and ending no later than 0700.

• Patient is right-handed. Patients who are ambidextrous may be eligible following consultation with the medical monitor.

• Women of childbearing potential must use a medically accepted method of contraception and must agree to continue use of this method for the duration of the study and for 30 days after participation in the study.

• Patient exhibits reasonable accuracy (=80%) on the 2-back working memory task during the training session at the second screening visit.

Exclusion Criteria:

The Patient:

• The patient is a current smoker or has a prior history of smoking (defined as =1 pack-year) within 2 years prior to the screening visit.

• consumes caffeine including coffee, tea and/or other caffeine-containing beverages or food averaging more than 400 mg of caffeine per day (approximately equivalent to 4 or more cups of coffee).

• has NART-predicted verbal IQ and QIDS-SR16 scores within protocol-specific exclusionary ranges.

• has a clinically significant, uncontrolled medical or psychiatric conditions (treated or untreated).

• has a confirmed or probable diagnosis of a current sleep disorder other than OSAHS.

• has used any excluded prescription drugs or procedures for prohibited and allowed drugs within the excluded timeframe.

• has a history of alcohol, narcotic, or any other drug abuse.

• has a positive UDS, without medical explanation, at the screening visit.

• has a clinically significant deviation from normal in the physical examination.

• is a pregnant or lactating woman. Any woman becoming pregnant during the study will be withdrawn from the study.

• has a past or present seizure disorder, head trauma that is clinically significant, or past neurosurgery.

• has used an investigational drug within 1 month before the screening visit.

• has any disorder that may interfere with drug absorption, distribution, metabolism, or excretion (including gastrointestinal surgery).

• has a known hypersensitivity to armodafinil or modafinil, or any other component of the study drug tablets.

• has a history of any clinically significant cutaneous drug reaction, or a history of clinically significant hypersensitivity reaction, including multiple allergies or drug reactions.

• has known human immunodeficiency virus (HIV).

• has clinical laboratory test value(s) outside the range(s) specified in the Protocol, or presents a clinically significant laboratory abnormality without prior written approval by the medical monitor.

• has worked the night shift within 28 days of the baseline visit, or will work the night shift during the double-blind segment of the study.

• anticipates any travel across more than 3 time zones at any time during the study.

• needs to use any of the excluded medications identified in this protocol.

• is unable to complete neuroimaging studies, performance tasks, self-rating scales, and all other study assessments.

• has a contraindication to fMRI scanning, (such as an implanted pacemaker/defibrillator, aneurysm clips, drug infusion device or metallic foreign body).

• is suspected to be unable to tolerate fMRI scanning (eg, claustrophobic) and/or the testing paradigm.

• has physical or other characteristics that suggest imaging data will be unobtainable or degraded.

Study Design

Allocation: Randomized, Intervention Model: Parallel Assignment, Masking: Double Blind (Subject, Investigator), Primary Purpose: Treatment

Related Conditions & MeSH terms

• Disorders of Excessive Somnolence
• Excessive Sleepiness

Intervention

Drug:
• Armodafinil
Armodafinil once-daily (50 mg/day (1 tablet) on Day 1; increased to 100 mg/day (2 tablets) starting on Day 2; increased to 150 mg/day (3 tablets) starting on Day 5; increased to 200 mg/day (4 tablets) starting on Day 8). Then continue 200 mg/day dosage through Day 14.
• Placebo
Matching Placebo dosed once-daily (50 mg/day (1 tablet) on Day 1; increased to 100 mg/day (2 tablets) starting on Day 2; increased to 150 mg/day (3 tablets) starting on Day 5; increased to 200 mg/day (4 tablets) starting on Day 8). Then continue 200 mg/day dosage through Day 14.

Locations

Country	Name	City	State
United States	Beth Israel Deaconess Medical	Boston	Massachusetts
United States	Peninsula Sleep Center	Burlingame	California
United States	Duke University Medical Center	Durham	North Carolina
United States	Neurocare, Inc.	Newton	Massachusetts
United States	Wake Research Associates	Raleigh	North Carolina
United States	Pacific Research	San Diego	California
United States	VA San Diego Healthcare System	San Diego	California
United States	Washington University	St. Louis	Missouri
United States	Stanford University	Stanford	California

Sponsors (1)

Lead Sponsor	Collaborator
Cephalon

Country where clinical trial is conducted

United States, 

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Change From Baseline to Endpoint in Number of Contiguous Activated Voxels Meeting Predefined Threshold in Dorsolateral Prefrontal Cortex (DLPFC) on Functional Magnetic Resonance Imaging (fMRI) as a Measure of Prefrontal Cortical Activation	The primary outcome was the change from baseline in number of contiguous activated voxels in the dorsolateral prefrontal cortex (DLPFC) on functional magnetic resonance imaging (fMRI) at Week 2(or last observation after baseline). Each voxel is compared to the reference wave form. If it differs from that value p<0.05, the voxel is considered active. fMRI is a brain imaging technique that identifies neuronal activation related to specific tasks or sensory stimulation. Increased neuronal activity increases blood flow and oxygen content to the activated part of the brain, altering fMRI signal.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Change From Baseline to Endpoint in Mean Response Latency in the 2-Back Working Memory Test at Endpoint - Mean Performance Speed	The 2-Back is a verbal working memory test in which random letters are presented visually every 4 sec, with each stimulus lasting 500 msec. Subjects are asked to make a yes/no response following each letter indicating whether it was the same or different from the letter presented two earlier. The load on working memory was the ordering, retention, updating, and manipulation of 2 letters and consideration of the relationship to a 3rd newly presented letter, which could have been a target or a nontarget. The change from baseline in response latency at endpoint is presented here.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Change From Baseline to Endpoint in the Number of Contiguous Activated Voxels Meeting the Predefined Threshold in the Anterior Cingulate Cortex (ACC)	The outcome was the change from baseline in number of contiguous activated voxels in the anterior cingulate cortex (ACC) on functional magnetic resonance imaging (fMRI) at Week 2(or last observation after baseline). Each voxel is compared to the reference wave form. If it differs from that value p<0.05, the voxel is considered active. fMRI is a brain imaging technique that identifies neuronal activation related to specific tasks or sensory stimulation. Increased neuronal activity increases blood flow and oxygen content to the activated part of the brain, altering fMRI signal.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Change From Baseline to Endpoint in the Number of Contiguous Voxels Meeting the Predefined Threshold in the Posterior Parietal Cortex (PPC)	The outcome was the change from baseline in number of contiguous activated voxels in the posterior parietal cortex (PPC) on functional magnetic resonance imaging (fMRI) at Week 2(or last observation after baseline). Each voxel is compared to the reference wave form. If it differs from that value with p<0.05, the voxel is considered active. fMRI is a brain imaging technique that identifies neuronal activation related to specific tasks or sensory stimulation. Increased neuronal activity increases blood flow and oxygen content to the activated part of the brain, altering fMRI signal.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Change From Baseline to Endpoint in the Number of Contiguous Activated Voxels Meeting the Predefined Threshold in the Thalamus	The outcome was the change from baseline in number of contiguous activated voxels in the thalamus on functional magnetic resonance imaging (fMRI) at Week 2(or last observation after baseline). Each voxel is compared to the reference wave form. If it differs from that value significantly (p<0.05), the voxel is considered active. fMRI is a brain imaging technique that identifies neuronal activation related to specific tasks or sensory stimulation. Increased neuronal activity increases blood flow and oxygen content to the activated part of the brain, altering fMRI signal.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Pattern Recognition Memory (PRM) Percent Correct (Immediate) From the CANTAB Battery-Change From Baseline to Endpoint	The PRM test from the Cambridge Neuropsychological Test Automated Battery (CANTAB) assesses episodic memory by a patient's ability to encode and recognize visual information. Patterns appear sequentially on the screen, and patients are instructed to remember them. Immediately afterwards a recognition test is performed, in which each pattern shown earlier is presented with another pattern of similar form and color. Patient has to touch the pattern seen earlier. Change from baseline to endpoint in % correct responses with immediate recall is presented. Subjects complete 24 trials per assessment.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Pattern Recognition Memory (PRM) Percent Correct (Delayed) From the CANTAB Battery-Change From Baseline to Endpoint	The PRM test from the Cambridge Neuropsychological Test Automated Battery (CANTAB) assesses episodic memory as measured by a patient's ability to encode and recognize visual information. Patterns appear sequentially on the screen, and patients are instructed to remember them. Twenty minutes following the immediate recognition test, another "delayed" recognition test is performed, featuring the same stimuli as in the first phase. The change from baseline to endpoint in percent correct responses of this delayed test are presented here. Subjects complete 24 trials per assessment.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Reaction Time Index (RTI) Median Correct Latency, Five Choice Test From the CANTAB Battery-Change From Baseline to Endpoint	The RTI is a measure of simple and choice reaction time, movement time and spatio-temporal vigilance during simple and 5 choice reaction time trials. This task also permits measurement of anticipatory/premature responding and perseverative responding. The patient responded to a yellow spot appearing on the screen by letting go of the press pad and touching the location in which the spot appeared. The yellow spot appeared in any 1 of 5 locations in the 5 choice reaction time phase. The change from baseline to endpoint in median correct latency is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Reaction Time Index (RTI) Median Correct Latency, One Choice Test From the CANTAB Battery-Change From Baseline to Endpoint	The RTI is a measure of simple and choice reaction time, movement time and spatio-temporal vigilance during simple and 5 choice reaction time trials. This task also permits measurement of anticipatory/premature responding and perseverative responding. The patient responded to a yellow spot appearing on the screen by letting go of the press pad and touching the location in which the spot appeared. The yellow spot appeared in a single location during the simple reaction time phase. The change from baseline to endpoint in median correct latency is presented here.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	One Touch Stockings of Cambridge (OTS) Mean Correct Latency, (Easy) From the CANTAB Battery-Change From Baseline to Endpoint	OTS is a spatial planning test based on Tower of London test and the CANTAB Stockings of Cambridge test, and measures frontal lobe function. Subject is shown 2 displays containing 3 colored balls and a row of boxes containing numbers. The patient was shown one demonstration problem and then had to solve 3 additional problems (easy). The problems increased in complexity, from one to six moves. With additional problems subject had to work out how many moves the solutions required in their heads (hard). Change from baseline to endpoint in Mean correct latency for the easy problems is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	One Touch Stockings of Cambridge (OTS) Mean Correct Latency, (Hard) From the CANTAB Battery-Change From Baseline to Endpoint	OTS is a spatial planning test based on Tower of London test and the CANTAB Stockings of Cambridge test, and measures frontal lobe function. Subject is shown 2 displays containing 3 colored balls and a row of boxes containing numbers. The patient was shown one demonstration problem and then had to solve 3 additional problems (easy). The problems increased in complexity, from one to six moves. With additional problems subject had to work out how many moves the solutions required in their heads (hard). Change from baseline to endpoint in Mean correct latency for the hard problems is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	One Touch Stockings of Cambridge (OTS) Mean Choices to Correct, (Easy) From the CANTAB Battery-Change From Baseline to Endpoint	OTS is a spatial planning test based on the Tower of London and the CANTAB Stockings of Cambridge test, and measures frontal lobe function. Patient shown 2 displays containing 3 colored balls and a row of boxes containing numbers. The patient is shown one demonstration problem and then solves 3 additional problems (easy). Problems increased in complexity, from one to six moves. With additional problems (hard) the patient has to work out how many moves the solutions required in their heads. Mean change from Baseline to endpoint in number of choices to correct for easy problems is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	One Touch Stockings of Cambridge (OTS) Mean Choices to Correct, (Hard) From the CANTAB Battery-Change From Baseline to Endpoint	OTS is a spatial planning test based on the Tower of London and the CANTAB Stockings of Cambridge test, and measures frontal lobe function. Patient shown 2 displays containing 3 colored balls and a row of boxes containing numbers. The patient was shown one demonstration problem and then solves 3 additional problems (easy). Problems increased in complexity, from one to six moves. With additional problems (hard) the patient had to work out how many moves the solutions required in their heads. Mean change from baseline to endpoint in number of choices to correct for hard problems is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Epworth Sleepiness Scale Change From Baseline to Endpoint	The patient's evaluation of excessive daytime sleepiness was measured by the patient reported measure, ESS (Johns1991). The ESS score was based on responses to questions referring to 8 everyday situations (eg, sitting and reading, talking to someone, being stopped in traffic) and reflected a patient's propensity to fall asleep in those situations. The ESS score was derived from the sum of the values from questions corresponding to the 8 situations. Scores for the ESS ranged from 0 to 24, with a higher score indicating a greater daytime sleepiness. Change from baseline to endpoint is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Clinical Global Impression of Change (CGI-C)- Number of Responders at Endpoint	Severity of sleepiness, was assessed by the Clinical Global Impression of Severity (CGI-S) at Baseline. The clinician assessed the change from baseline in the patient's condition, as related to excessive sleepiness, in response to treatment using the CGI-C, which consisted of the following 7 categories and scoring assignments: very much improved, much improved, minimally improved, no change, minimally worse, much worse, and very much worse. Responders had to be at least minimally improved from Baseline to qualify as a responder at Endpoint.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Total Score From the Medical Outcomes Study 6-Item Cognitive Function Scale (MOS-CF6)-Change From Baseline to Endpoint	The MOS-CF6 is an instrument to assess patient self-reported cognitive function. Items were selected to cover 6 relevant aspects of cognitive functioning as follows: confusion, concentration/thinking, attention, memory, reasoning, problem solving, and processing speed. The CF 6 item responses include 6 choices, ranging from "none of the time" to "all of the time." The CF-6 was scored by summing responses across the 6 items and converting the total to a 0 to 100 point scale, with higher scores indicating better cognitive functioning. Change in MOS-CF6 from baseline to endpoint is reported.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Blood Oxygenation Level Dependent (BOLD) Signal Intensity - Percent Change From Baseline to Endpoint in the Dorsolateral Prefrontal Cortex (DLPFC)	Functional magnetic resonance imaging (fMRI) is a brain imaging technique that identifies neuronal activation in regions related to specific tasks or sensory stimulation such as language, vision, hearing, and short-term memory. When neuronal activity increases, blood flow increases to that part of the brain with an increase in the oxygen content of the blood. Increase in oxygen content causes the fMRI signal in that part of the brain to change, and is the basis of the BOLD effect. The percent change in BOLD signal from Baseline to 2 weeks or last observation after baseline is presented here.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Blood Oxygenation Level Dependent (BOLD) Signal Intensity - Percent Change From Baseline to Endpoint in the Anterior Cingulate Cortex (ACC)	Functional magnetic resonance imaging (fMRI) is a brain imaging technique that identifies neuronal activation in regions related to specific tasks or sensory stimulation such as language, vision, hearing, and short-term memory. When neuronal activity increases, blood flow increases to that part of the brain with an increase in the oxygen content of the blood. Increase in oxygen content causes the fMRI signal in that part of the brain to change, and is the basis of the BOLD effect. The percent change in BOLD signal from Baseline to 2 weeks or last observation after baseline is presented here.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Blood Oxygenation Level Dependent (BOLD) Signal Intensity -Change From Baseline to Endpoint in the Posterior Parietal Cortex (PPC)	Functional magnetic resonance imaging (fMRI) is a brain imaging technique that identifies neuronal activation in regions related to specific tasks or sensory stimulation such as language, vision, hearing, and short-term memory. When neuronal activity increases, blood flow increases to that part of the brain with an increase in the oxygen content of the blood. Increase in oxygen content causes the fMRI signal in that part of the brain to change, and is the basis of the BOLD effect. The percent change in BOLD signal from Baseline to 2 weeks or last observation after baseline is presented here.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Blood Oxygenation Level Dependent (BOLD) Signal Intensity - Percent Change From Baseline to Endpoint in the Thalamus	Functional magnetic resonance imaging (fMRI) is a brain imaging technique that identifies neuronal activation in regions related to specific tasks or sensory stimulation such as language, vision, hearing, and short-term memory. When neuronal activity increases, blood flow increases to that part of the brain with an increase in the oxygen content of the blood. Increase in oxygen content causes the fMRI signal in that part of the brain to change, and is the basis of the BOLD effect. The percent change in BOLD signal from Baseline to 2 weeks or last observation after baseline is presented here.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Activation-Performance Relationship Between the Functional Magnetic Resonance Imaging (fMRI) in Dorsolateral Prefrontal Cortex (DLPFC) and 2-Back Working Memory Test - Number of Voxels Activated at Endpoint	With this outcome measure the correlation between the number of voxels activated on fMRI (voxels that differ significantly from reference wave form) in DLPFC versus performance on the 2-back working memory test was evaluated for both Armodafinil and Placebo. Correlation coefficients and P-values are presented for each treatment group.	Endpoint (Week 2 or last observation after baseline)	No
Secondary	Activation-Performance Relationship Between the Functional Magnetic Resonance Imaging (fMRI) in Anterior Cingulate Cortex (ACC) and 2-Back Working Memory Test - Number of Voxels Activated at Endpoint	With this outcome measure the correlation between the number of voxels activated on fMRI (voxels that differ significantly from reference wave form) in ACC versus performance on the 2-back working memory test was evaluated for both Armodafinil and Placebo. Correlation coefficients and P-values are presented for each treatment group.	Week 2 or Last Observation after Baseline	No
Secondary	Activation-Performance Relationship Between Functional Magnetic Resonance Imaging (fMRI) in Posterior Parietal Cortex (PPC) and the 2-Back Working Memory Test -Number of Voxels Activated at Endpoint	With this outcome measure the correlation between the number of voxels activated on fMRI (voxels that differ significantly from reference wave form) in PPC versus performance on the 2-back working memory test was evaluated for both Armodafinil and Placebo. Correlation coefficients and P-values are presented for each treatment group.	Week 2 or Last Observation after Baseline	No
Secondary	Activation-Performance Relationship Between Functional Magnetic Resonance Imaging (fMRI) in the Thalamus and 2-Back Working Memory Test -Number of Voxels Activated at Endpoint	With this outcome measure the correlation between the number of voxels activated on fMRI (voxels that differ significantly from reference wave form) in the thalamus versus performance on the 2-back working memory test was evaluated for both Armodafinil and Placebo. Correlation coefficients and P-values are presented for each treatment group.	Week 2 or Last Observation after Baseline	No
Secondary	Activation-Performance Relationship Between the Functional Magnetic Resonance Imaging (fMRI) in Dorsolateral Prefrontal Cortex (DLPFC) and 2-Back Working Memory Test - Blood Oxygen Level Dependent (BOLD) Signal Intensity at Endpoint	With this outcome measure the correlation between the BOLD signal intensity on fMRI over DLPFC versus performance on the 2-back working memory test was evaluated for both Armodafinil and Placebo. Correlation coefficients and P-values are presented for each treatment group.	Week 2 or Last Observation after Baseline	No
Secondary	Activation-Performance Relationship Between Functional Magnetic Resonance Imaging (fMRI) in Anterior Cingulate Cortex (ACC) and 2-Back Working Memory Test -Blood Oxygen Level Dependent (BOLD) Signal Intensity at Endpoint	With this outcome measure the correlation between the BOLD signal intensity on fMRI in the ACC versus performance on the 2-back working memory test was evaluated for both Armodafinil and Placebo. Correlation coefficients and P-values are presented for each treatment group.	Week 2 or Last Observation after Baseline	No
Secondary	Activation-Performance Relationship on Functional Magnetic Resonance Imaging (fMRI) in Posterior Parietal Cortex (PPC) and 2-Back Working Memory Test - Blood Oxygen Level Dependent (BOLD) Signal Intensity at Endpoint	With this outcome measure the correlation between the BOLD signal intensity on fMRI in PPC versus performance on the 2-back working memory test was evaluated for both Armodafinil and Placebo. Correlation coefficients and P-values are presented for each treatment group.	Week 2 or Last Observation after Baseline	No
Secondary	Activation-Performance Relationship on Functional Magnetic Resonance Imaging (fMRI) in the Thalamus and 2-Back Working Memory Test - Blood Oxygen Level Dependent (BOLD) Signal Intensity at Endpoint	With this outcome measure the correlation between BOLD signal intensity on fMRI in the thalamus versus performance on the 2-back working memory test was evaluated for both Armodafinil and Placebo. Correlation coefficients and P-values are presented for each treatment group.	Week 2 or Last Observation after Baseline	No
Secondary	Number of Contiguous Activated Voxels Meeting Predefined Threshold in the DLPFC on fMRI on the 2 Back Working Memory Test - Change From Baseline-Subgroup-Responders in 2 Back Working Memory Test	This is a subgroup analysis of responders on the 2-back working memory test for the number of voxels meeting the predefined threshold in DLPFC. A responder in the 2-back working memory test was defined as a patient showing a response latency of less than 713 ms at endpoint. This is based on baseline data from the matched control population in a functional imaging study in patients with obstructive sleep apnea. The change from Baseline to Endpoint in the number of activated voxels (that differ significantly from reference wave form) for each treatment group among the responders is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Number of Contiguous Activated Voxels Meeting Predefined Threshold in the ACC on fMRI by 2-Back Working Memory Test -Change From Baseline; Subgroup-Responders in 2 Back Memory Test	This is a subgroup analysis of responders on the 2-back working memory test for the number of activated voxels (that differ significantly from reference wave form) in Anterior Cingulate Cortex (ACC). A responder in the 2-back working memory test was defined as a patient showing a response latency of less than 713 ms at endpoint. This is based on baseline data from the matched control population in a functional imaging study in patients with obstructive sleep apnea. The change from Baseline to Endpoint in the number of activated voxels for each treatment group among the responders is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Number of Contiguous Activated Voxels Meeting Predefined Threshold in the PPC on fMRI by 2-Back Working Memory Test-Change From Baseline; Subgroup-Responders in 2 Back Memory Test	This is a subgroup analysis of responders on the 2-back working memory test for the number of voxels (voxels that differ significantly from reference wave form) in Posterior Parietal Cortex (PPC). A responder in the 2-back working memory test was defined as a patient showing a response latency of less than 713 ms at endpoint. This is based on baseline data from the matched control population in a functional imaging study in patients with obstructive sleep apnea. The change from Baseline to Endpoint in the number of activated voxels for each treatment group among the responders is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Number of Contiguous Activated Voxels Meeting Predefined Threshold in the Thalamus on fMRI by 2-Back Working Memory Test-Change From Baseline; Subgroup-Responders in 2 Back Memory Test	This is a subgroup analysis of responders on the 2-back working memory test for the number of activated voxels (voxels that differ significantly from reference wave form) in the thalamus. A responder in the 2-back working memory test was defined as a patient showing a response latency of less than 713 ms at endpoint. This is based on baseline data from the matched control population in a functional imaging study in patients with obstructive sleep apnea. The change from Baseline to Endpoint in the number of activated voxels for each treatment group among the responders is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Number of Contiguous Activated Voxels Meeting Predefined Threshold in the DLPFC on fMRI by 2-Back Working Memory Test-Change From Baseline; Subgroup-Non Responders in 2 Back Memory Test	This is a subgroup analysis of non-responders on the 2-back working memory test for the number of voxels meeting the predefined threshold (voxels that differ significantly from reference wave form) on functional magnetic resonance imaging (fMRI). A non-responder in the 2-back working memory test was defined as a patient showing a response latency of 713 ms or greater at endpoint. The change from Baseline to Endpoint in the number of activated voxels in the dorsolateral prefrontal cortex (DLPFC)for each treatment group among the non-responders is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Number of Contiguous Activated Voxels Meeting Predefined Threshold in the ACC on fMRI by 2-Back Working Memory Test-Change From Baseline; Subgroup-Non Responders in 2 Back Memory Test	This is a subgroup analysis of non-responders on the 2-back working memory test for the number of voxels meeting the predefined threshold (voxels that differ significantly from reference wave form) on functional magnetic resonance imaging (fMRI). A non-responder in the 2-back working memory test was defined as a patient showing a response latency of 713 ms or greater at endpoint. The change from Baseline to Endpoint in the number of activated voxels in the anterior cingulate cortex (ACC)for each treatment group among the non-responders is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Number of Contiguous Activated Voxels Meeting Predefined Threshold in the PPC on fMRI by 2-Back Working Memory Test-Change From Baseline; Subgroup-Non Responders in 2 Back Memory Test	This is a subgroup analysis of non-responders on the 2-back working memory test for the number of voxels meeting the predefined threshold (voxels that differ significantly from reference wave form) on functional magnetic resonance imaging (fMRI). A non-responder in the 2-back working memory test was defined as a patient showing a response latency of 713 ms or greater at endpoint. The change from Baseline to Endpoint in the number of activated voxels in the posterior parietal cortex (PPC)for each treatment group among the non-responders is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Number of Contiguous Activated Voxels Meeting Predefined Threshold in the Thalamus on fMRI by 2-Back Working Memory Test-Change From Baseline; Subgroup-Non Responders in 2 Back Memory Test	This is a subgroup analysis of non-responders on the 2-back working memory test for the number of voxels meeting the predefined threshold (voxels that differ significantly from reference wave form) on functional magnetic resonance imaging (fMRI). A non-responder in the 2-back working memory test was defined as a patient showing a response latency of 713 ms or greater at endpoint. The change from Baseline to Endpoint in the number of activated voxels in the thalamus for each treatment group among the non-responders is presented.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Change From Baseline to Endpoint in the BOLD Signal Intensity in the Dorsolateral Prefrontal Cortex (DLPFC) at Resting State	At resting state, this is an analysis of the change from Baseline to Endpoint in the blood oxygen level dependent (BOLD) signal intensity in the dorsolateral prefrontal cortex (DLPFC).	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Change From Baseline to Endpoint in the BOLD Signal Intensity in the Anterior Cingulate Cortex (ACC) at Resting State	At resting state, this is an analysis of the change from Baseline to Endpoint in the blood oxygen level dependent signal (BOLD) intensity in the anterior cingulate cortex (ACC).	Baseline and Endpoint (Week 2 or last observation after Baseline)	No
Secondary	Change From Baseline in the BOLD Signal Intensity in the Posterior Parietal Cortex (PPC) at Resting State	At resting state, this is an analysis of the change from Baseline to Endpoint in the blood oxygen level dependent (BOLD) signal intensity in the posterior parietal cortex (PPC).	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Change From Baseline to Endpoint in the BOLD Signal Intensity in the Thalamus at Resting State	At resting state, this is an analysis of the change from Baseline to Endpoint in the blood oxygen level dependent (BOLD) signal intensity in the thalamus.	Baseline and Endpoint (Week 2 or last observation after Baseline)	No
Secondary	Change From Baseline to Endpoint in the Number of Voxels Meeting Predefined Threshold in the Dorsolateral Prefrontal Cortex (DLPFC) at Resting State	At resting state, this is an analysis of the change from Baseline to Endpoint in the number of contiguous activated voxels (voxels that differ significantly from reference wave form) on functional magnetic resonance imaging (fMRI) in the dorsolateral prefrontal cortex.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Change From Baseline to Endpoint in the Number of Voxels Meeting Predefined Threshold in Anterior Cingulate Cortex (ACC) at Resting State	At resting state, this is an analysis of the change from Baseline to Endpoint in the number of contiguous activated voxels meeting pre-defined threshold (voxels that differ significantly from reference wave form) on functional magnetic resonance imaging (fMRI) in the anterior cingulate cortex (ACC).	Baseline and Endpoint (Week 2 or last observation after Baseline)	No
Secondary	Change From Baseline to Endpoint in the Number of Voxels Meeting Predefined Threshold in Posterior Parietal Cortex (PPC) at Resting State	At resting state, this is an analysis of the change from Baseline to Endpoint in the number of activated voxels meeting predefined threshold (voxels that differ significantly from reference wave form) on functional magnetic resonance imaging (fMRI) in the posterior parietal cortex (PPC).	Baseline and Endpoint (Week 2 or last observation after Baseline)	No
Secondary	Change From Baseline to Endpoint in the Number of Voxels Meeting Predefined Threshold in the Thalamus at Resting State	At resting state, this is an analysis of the change from Baseline to Endpoint in the number of activated voxels meeting predefined threshold (voxels that differ significantly from reference wave form) on functional magnetic resonance imaging (fMRI) in the thalamus.	Baseline and Endpoint (Week 2 or last observation after baseline)	No
Secondary	Change From Baseline to Endpoint (2 Weeks or Last Observation After Baseline) in the Mean Response Latency in the Psychomotor Vigilance-Like Test	During anatomic scanning (and prior to functional runs when anatomic scanning was not performed), a modified continuous 10 minute attention task ("Psychomotor Vigilance Test [PVT]-like task," nearly identical to the PVT but for absence of performance feedback) was run to obtain a measure of vigilance in the scanner—in this instance, the "+" symbol appeared at random (mean inter trial interval of 5 seconds, range 2 - 10 seconds) but disappeared when subject pressed a button. Subject performance speed was measured. Change in subject performance speed from Baseline to Endpoint is presented.	Baseline and Endpoint (Week 2 or last observation after Baseline)	No