Flumazenil for the Treatment of Primary Hypersomnia

A Ten Subject, Double-Blind, Placebo-Controlled Trial of Single Day Dosing of Sublingual Flumazenil in Individuals With Primary Hypersomnia or Excessively Long Total Sleep Time and Excess Endogenous Potentiation of GABA-A Receptors

The term 'hypersomnia' describes a group of symptoms that includes severe daytime sleepiness and sleeping long periods of time (more than 10 hours per night). Sometimes, hypersomnia is caused by a problem with the quality of sleep occurring at night, for instance when nighttime sleep is disrupted by frequent breathing pauses. In other cases, however, hypersomnia occurs even when nighttime sleep is of good quality. These cases of hypersomnia are presumed to be a symptom of brain dysfunction, and so are referred to as hypersomnias of central (i.e., brain) origin, or primary hypersomnias. The causes of most of these primary hypersomnias are not known. However, our group has recently identified a problem with the major brain chemical responsible for sedation, known as GABA. In a subset of our hypersomnia patients, there is a naturally-occurring substance that causes the GABA receptor to be hyperactive. In essence, it is as though these patients are chronically medicated with Valium (or Xanax or alcohol, all substances that act through the GABA system), even though they do not take these medications. Current treatment of central hypersomnias is limited. For the fraction of cases with narcolepsy, there are FDA-approved, available treatments. However, for the remainder of patients, there are no treatments approved by the FDA. They are usually treated with medications approved for narcolepsy, but sleep experts agree that these medications are often not effective for this group of patients. Based on our understanding of the GABA abnormality in these patients, we evaluated whether flumazenil (an medication approved by the FDA for the treatment of overdose of GABA medications or the reversal of GABA-based anesthesia) would reverse the GABA abnormality in our patients. In a test tube model of this disease, flumazenil does in fact return the function of the GABA system to normal. The investigators have treated a few patients with flumazenil and most have felt that their hypersomnia symptoms improved with this treatment. To determine whether flumazenil is truly beneficial for primary hypersomnia, this study will compare flumazenil to an inactive pill (the placebo). All subjects will receive both flumazenil and the placebo at different times, and their reaction times and symptoms will be compared on these two treatments to determine if one is superior. Currently, flumazenil can only be given through an injection into a vein (i.e., intravenously). This study will evaluate this intravenous dosing as well as a new form of flumazenil, which is taken as a lozenge to be dissolved under the tongue. If this study shows that flumazenil is more effective than placebo in the treatment of hypersomnia, it will identify a potential new therapy for this difficult-to-treat disorder.

The PVT measures the reaction time to button press following the presentation of a visual stimulus, reported here as the median reaction time for multiple presentations during the 10 minute task. The measure used was the change in median reaction time from baseline to drug administration, where the median reaction time at each of the time points (below) was averaged to provide a single on-treatment value for median reaction time. The measure was then calculated as baseline value - treatment value, such that higher numbers denote improvement from baseline.

10, 30, 60, 90, 120, and 150 minutes after drug administration (averaged for all time points for each subject)

A PVT lapse is defined as a reaction time exceeding 500 msec following the presentation of a single stimulus, which are then summed for the entire 10 minute PVT testing period. The measure used was the change in the frequency of lapses from baseline to drug administration (calculated as baseline value - average value with study drug, where higher numbers denote improvement from baseline).

10, 30, 60, 90, 120, and 150 minutes after drug administration (averaged for all time points for each subject)

The PVT duration of lapse domain is defined as the reciprocal of the reaction time averaged across the slowest 10% of responses. The measure used was the change in duration of lapse domain from baseline to drug administration (calculated as baseline value - average value with study drug, where lower numbers denote improvement from baseline).

10, 30, 60, 90, 120, and 150 minutes after drug administration (averaged for all time points for each subject)

The optimum response times is defined as the reciprocal of the reaction time averaged across the fastest 10% of responses. The measure used was the change in optimum response time from baseline to following drug administration (calculated as baseline value - average value with study drug, where lower numbers denote improvement from baseline).

10, 30, 60, 90, 120, and 150 minutes after drug administration (averaged for all time points for each subject)

The false response frequency is defined as the number of button presses when no stimulus is presented. The measure used was the change in false response frequency from baseline to drug administration (calculated as baseline value - average value with study drug, where higher numbers denote improvement from baseline).

10, 30, 60, 90, 120, and 150 minutes after drug administration (averaged for all time points for each subject)

PVT Additional Measure #5, Change in Visual Analog Scale Rating of Sleepiness at the Completion of PVT

At the end of the 10 minute PVT testing period, subjects were asked to rate their current level of sleepiness along a line, which was transformed into a numeric value from 1-10, such that high levels indicated more severe subjective sleepiness. The measure used was the change in this rating from baseline to drug administration (calculated as baseline value - average value with study drug, where higher numbers denote improvement from baseline).

10, 30, 60, 90, 120, and 150 minutes after drug administration (averaged for all time points for each subject)

The Stanford Sleepiness Scale (SSS) is a subjective rating of sleepiness, with score ranging from 1 to 7, where higher values reflect more severe sleepiness. The measure used was change in SSS from baseline to drug administration (calculated as baseline value - average value with study drug, where higher numbers denote improvement from baseline).

10, 30, 60, 90, 120, and 150 minutes after drug administration (averaged for all time points for each subject)

EEG signals reflect the state of excitability of the cerebral cortex and correlate highly with levels of behavioral arousal. This is quantifiable as 'power' of the signal (microvolts squared/signal frequency). The EEG signals will be acquired and stored for off-line power analysis and comparison between treatment conditions.

Inclusion Criteria: Hypersomnia (meeting clinical criteria for idiopathic hypersomnia with or without long sleep time, narcolepsy lacking cataplexy, or symptomatic hypersomnia not meeting International Classification of Sleep Disorders 2 (ICSD-2) criteria inclusive of habitually long sleep periods of > 10 hours/day) evidence for GABA-related abnormality, as demonstrated by our in-house, in vitro assay age > 18 high performance liquid chromatography/liquid chromatography tandem mass spectrometry verification of the absence of exogenous benzodiazepines (BZDs). Exclusion Criteria: Contraindications to use of flumazenil (pregnancy, hepatic impairment, seizure history, pre-menstrual dysphoric disorder, traumatic brain injury, cardiac disease (left ventricular diastolic dysfunction), or cardiac dysrrhythmia. Current use of a BZD or BZD-receptor agonists moderate or severe sleep apnea (RDI > 15/hr), severe periodic limb movement disorder (PLMI > 30/hr) diagnosis of narcolepsy with cataplexy, as determined by ICSD-2 criteria and confirmed by absence of cerebrospinal fluid (CSF) hypocretin metabolic disorders such as severe anemia, adrenal insufficiency, severe iron deficiency, vitamin B12 deficiency, or hypothyroidism that may explain symptoms of hypersomnia