Efficacy of modafinil compared to γ-hydroxybutyrate in the treatment of narcolepsy

Introduction

Narcolepsy is a chronic disease state which causes overwhelming daytime sleepiness, regardless of the circumstances or settings, and sometimes cataplexy. Cataplexy is the sudden loss of muscle tone, often precipitated by a strong emotion. Narcolepsy Type 1 (NT1) is defined as presenting with cataplexy, while narcolepsy without cataplexy is termed Narcolepsy Type 2 (NT2). Other common manifestations include sudden sleep attacks, insomnia episodes (inability to achieve or maintain sleep during normal sleeping hours), hypnagogic and hypnopompic hallucinations, and altered sleep stage architecture, specifically Rapid Eye Movement (REM) sleep. Narcolepsy is often debilitating, but typically does not pose a direct threat to one’s life. Rather, daily routines (i.e., driving, being productive at work, etc.), normal functioning and mental health are often greatly impacted (Thorpy and Dauvilliers, 2015).

Many co-morbidities have been strongly associated with both NT1 and NT2, such as obesity, type 2 diabetes mellitus, mood/psychiatric disorders, eating disorders, and attention deficit hyperactivity disorder. Additionally, associations also exist with other sleep and sleep-related disorders, such as restless-leg syndrome and REM parasomnias (Barateau and Dauvilliers, 2019). In a meta-analytic study of more than eight-million subjects, Scheer et al. (2019) found that 0.079% of the general United States public met criteria for narcolepsy, while incidence rates for NT2 are generally several-fold less than that of NT1. Additionally, females are more likely to be affected than males at a ratio of 3:2 (Scheer et al., 2019).

Although the specific etiology of narcolepsy is largely unknown, the literature clearly asserts that narcolepsy is characterized by specific physiological changes. Of these changes, the reduced levels of hypocretin (also known as orexin) in the lateral hypothalamus is the most understood and studied association (Takenoshita et al., 2018). Hypocretin is a peptide hormone that is involved with wakefulness regulation and REM sleep stages. Cerebrospinal fluid levels of < 110 pg/ml are consistent with a diagnosis of narcolepsy (Stephansen et al., 2018). It has been hypothesized that loss of hypocretin-producing cells could be due to autoimmune phenomena (Latorre et al., 2018), genetics (Ollila et al., 2015), human leukocyte antigen (HLA) haplotypes (Stephansen et al., 2018), modern pharmacological therapies such as influenzae vaccines (Miller et al., 2013), or the influenza virus itself (Bonvalet et al., 2017).

Many treatment strategies have been developed for this chronic disease state with some promising results, although further research is required to identify the most robust ‘gold standard’ of therapy. Currently, it is generally accepted that lifestyle changes (diet, regular sleep schedule, avoiding sleep deprivation, etc.) and pharmacological therapies are the most promising means for providing relief to patients with both NT1 and NT2 (Barateau and Dauvilliers, 2019). Examples of therapies for this purpose include γ-hydroxybutyrate (GHB, also known as sodium oxybate), modafinil (or its R-enantiomer, armodafinil), amphetamines (increases dopamine/norepinephrine in the CNS), methylphenidate (increases dopamine/norepinephrine in the CNS), pitolisant (selective histamine agonist at H₃ receptors), orexin agonists, immune-based therapies, and more (Barateau and Dauvilliers, 2019).

Modafinil is the current standard for treatment of excessive daytime sleepiness in narcolepsy (Sahni et al., 2019), although its mechanism is complex and largely unknown. The leading theory suggests that modafinil blocks the dopamine reuptake transporter (DAT) at the presynaptic neuron, providing amplified dopaminergic stimulation (Karabacak et al., 2015). Other theories include flux alterations of neurotransmitter exocytosis into the synaptic cleft (Ranjbari et al., 2021), and norepinephrine and/or serotonin reuptake inhibition at the presynaptic neuron (Murillo-Rodríguez et al., 2018). Since modafinil is a stimulant, it is typically taken in the morning and/or early afternoon.

On the other hand, GHB is an oral medication that is typically taken once before bedtime, and once 2.5 to 4 hours after upon briefly waking up. The total sum of the two doses typically ranges from 6 to 9 grams (Ohayon et al., 2019). It is also noteworthy to mention that GHB is often combined with modafinil for the treatment of narcolepsy (Erman et al., 2017), and that GHB is currently the only FDA-approved drug for the treatment of cataplexy associated with narcolepsy (Thorpy and Bogan, 2020). This drug agonizes both the Gamma-hydroxybutyric Acid B (GABA_B) receptor and the GHB-specific receptor (Trombley et al., 2020). Since stimulation of the GABA_B receptor causes cellular hyperpolarization in the CNS (through G-protein-mediated opening of inward rectifying potassium channels), GHB has a depressive/sedative effect on the brain. Conversely, when the GHB-specific receptor is stimulated (e.g., by GHB), excitatory effects are elicited, as seen in an animal-model study by Trombley et al. (2020), which found increased extracellular glutamate following selective stimulation of these receptors in neural tissue. Although GHB is more well-known for its depressive and sedative effects (e.g., at high concentrations through GABA_B stimulation), concentration-dependent differences in receptor affinity allow GHB to cause excitatory effects at lower concentrations through GHB-specific receptor stimulation (Trombley et al., 2020). Similar to modafinil, the current understanding of GHB’s complex mechanism of specifically treating narcolepsy is unknown. It has previously been hypothesized that GHB consolidates sleep by reducing fragmentation of the sleep stages (Roth et al., 2017).

One area of interest is the effect of pharmacotherapy on the sleep architecture and sleep staging. Before a patient falls asleep, they will be exhibiting β-waves on the electroencephalogram (EEG) portion of the PSG. When the patient dozes off (eyes shut, but still awake), this is termed N1, and the brain exhibits α-waves on EEG. N2 consists of theta-waves and contains sleep spindles and K-complexes (Latreille et al., 2020), while N3 (also called deep sleep, stage 3 sleep, slow-wave sleep) presents with larger 𝛿-waves (Hasan et al., 2015). Stephansen et al. (2018) explains that patients with narcolepsy often present with sleep architectural changes such as a shortened sleep latency, poor sleep consolidation and rapid/frequent shifts into REM sleep (Stephansen et al., 2018). Patients may also experience “sleep-wake dissociation,” in which the patient is partially awake and partially in REM sleep, which may be related to certain comorbidities. Examples include sleep paralysis and cataplexy (REM muscle paralysis in the setting of wakefulness). Dreaming while awake is another example, which has been termed hypnogogic/hypnopompic hallucinations (Stephansen et al., 2018).

This is an important area of research because refinement of treatment is paramount to the well-being and quality of life for patients with NT1/NT2. This review of the literature attempts to identify and compare efficacies of GHB and modafinil. As such, it is hypothesized that GHB provides greater functional efficacy than modafinil in the treatment of narcolepsy. It is also important to consider the adverse effects of these drugs, because if a patient stops using any given drug, there is no chance for the drug to be functionally efficacious. Hopefully, clarifying this knowledge will encourage physicians to prescribe the most appropriate medications to patients with narcolepsy, while minimizing adverse effects and reactions.

Key Terms: narcolepsy (NT1/NT2), γ-hydroxybutyrate (GHB) (also known as sodium oxybate, Xyrem), modafinil, REM sleep, N2, N3, N4, sleep latency, cataplexy, Epworth Sleepiness Scale (ESS), Maintenance of Wakefulness Test (MWT), Polysomnography (PSG), Pittsburgh Sleep Quality Index (PSQI)).

Methods

Database used: PubMed

Search stem: (oxybate OR GHB) AND modafinil

The initial search yielded 122 results. After removing one duplicate, 121 articles remained. Certain study designs, such as reviews, systematic reviews, case reviews, clinical reviews and meta-analyses (n = 77) were excluded, leaving 44 articles. Non-human studies (n = 3) were excluded, leaving 41 articles. No date ranges were set, in attempt to collect all journal entries making the direct comparison between the two drugs of interest. Thus, it follows that studies were only included if they directly compared both GHB and modafinil in the treatment of narcolepsy. Studies (n = 36) not fitting this criterion were excluded. This final intended parameter left five remaining articles, which were to be used for this review. One of the five papers was excluded upon further examination because of insufficient data reporting of the study’s modafinil statistics (Husain et al., 2020). To clarify, modafinil was given to the various applicable study groups, but authors failed to present sufficient numerical data, analysis, and discussion in their entry. For this reason, this journal entry was also excluded, leaving a grand total of four articles to be incorporated in the review.

Results

Of the papers included in this review, two were retrospective placebo-controlled RCTs (Black et al., 2016 & Dauvilliers et al., 2017), one was a double-blinded placebo-controlled RCT (Black et al., 2009), and one was a double-blinded multicenter RCT design (Black and Houghton, 2006). Table 1 describes the study design of the four papers selected for this review of the literature.

Black and Houghton (2006) conducted a double-blind, placebo-controlled multicenter RCT to assess the effects of GHB, modafinil and the combination of both drugs on patients with narcolepsy. Multiple sites (n = 44) across the globe were involved, including the United States, Canada, France, Germany, Czech Republic, the Netherlands, Switzerland, and the United Kingdom. To be included, each subject (n = 270) was previously diagnosed with narcolepsy and treatment with modafinil was present for purposes of baseline assessments. Pre-intervention modafinil dosage ranged from 200 mg to 600 mg per day (Black and Houghton, 2006).

Patients in the Black and Houghton (2006) study were divided into one of four groups: [1] GHB alone (GHB group), [2] modafinil alone (modafinil group), [3] GHB and modafinil together (GHB plus modafinil group), or [4] double-placebo group (placebo group). Before the eight-week intervention phase occurred, there was a two-week baseline period, during which patients continued to take their previously prescribed dosages of modafinil. During this phase, MWT, ESS and PSG data were collected for baseline analysis. At the beginning of the intervention phase, 6 grams of GHB was given to each subject in the GHB or GHB plus modafinil groups, for a total of four weeks. During weeks six through ten, the dosage was elevated to 9 grams of GHB per night. The principal outcome measures were the MWT, ESS, and sleep diaries to assess inadvertent naps and sleep attacks (Black and Houghton, 2006).

Compared to the mean baseline modafinil MWT (daytime sleep latency) score (and standard deviation) of 9.74 ± 6.57 minutes, the placebo group scores significantly decreased (worsened) to 6.87 ± 6.14 minutes at the end of the eight-week intervention (p < 0.001). This is likely due to the removal of modafinil during the intervention. However, the GHB group showed no decrease in sleep latency compared to its baseline score. For the GHB plus modafinil group, an additive effect was found, such that daytime sleep latency increased (improved) from a baseline of 10.43 ± 6.77 minutes to 13.15 ± 6.91 minutes (p < 0.001). It is noted that majority of change occurred during the first four weeks of intervention (Black and Houghton, 2006). Compared to baseline, ESS scores were significantly decreased (improved) at the end of week eight, for only the GHB group, 15 to 12 (p < 0.001), and the GHB plus modafinil group, 15 to 11 (p < 0.001). Standard deviation or standard error data was not provided for the remainder of this study’s applicable results.

These authors also found that at week eight, compared to placebo and modafinil groups, the GHB group and the GHB plus modafinil group presented with significantly lower instances of sleep attacks. The GHB group scored 10.05 sleep attacks per week at baseline, but only 7.10 at the conclusion of the intervention, while the GHB plus modafinil group scored 11.82 and 5.55, respectively. These differences were both significant (for both, p < 0.001) (Black and Houghton, 2006). The placebo group and the modafinil group remained unchanged for this parameter.

Black et al. (2009) investigated the effects of modafinil and GHB on nocturnal sleep cycle disruption, thus directly affecting and predicting the presence and strength of narcolepsy symptoms. This double-blinded, placebo-controlled study included 278 test subjects, each of whom had previously been taking modafinil due to a narcolepsy diagnosis. Pre-intervention modafinil dosages ranged from 200 – 400 mg per day. To prevent pregnancy and its potentially confounding complications, female patients were only accepted if they agreed to using birth-control, were surgically sterile, or held post-menopausal status. Patients were excluded if they had used GHB within the past month (e.g., for another clinical trial), had a diagnosis of sleep apnea (or positive screening at baseline testing in the current study), substance abuse disorder, seizures, or other neurological reasons (Black et al., 2009).

Following a two-week baseline period, each patient was randomized into one of four possible groups for the interventional period of eight total weeks. These groups’ drug regimen included [1] GHB placebo and modafinil placebo (placebo group), [2] GHB only (GHB group), [3] modafinil only (modafinil group), or [4] both GHB and modafinil (GHB plus modafinil group). Patients randomized into the modafinil group or the GHB plus modafinil group did not change their personalized dose of modafinil throughout the study, while patients in the placebo group and the GHB group were not prescribed any modafinil throughout the entire interventional phase. In the GHB group and GHB plus modafinil group, 6 grams of GHB was given nightly for the first four weeks of intervention, while nine grams were administered during each night of the final four weeks. This study used the MWT, PSG, sleep diaries and ESS for assessment of narcolepsy symptoms. Specifically, the MWT and PSG data were collected at three points: once at the conclusion of the baseline (pre-interventional) phase, once after week four and once following week eight of the interventional phase (Black et al., 2009).

PSG results at weeks four and eight showed no change in total sleep time in any of the test groups. At week eight, compared to baseline, two of the study groups showed greater proportions of N3 sleep, rather than lighter stages of sleep. More specifically, both the GHB group and the GHB plus modafinil group had an increased median of 43.5 minutes (p < 0.001) and 24.25 minutes (p < 0.001) of N3 sleep, respectively. Standard error and standard deviation data were not reported for this study’s results. By virtue of more N3 sleep, both groups showed a decrease of total time spent in N1 and REM sleep. At baseline, the GHB group spent 16 (p < 0.001) and 38.5 (p < 0.001) more minutes per night in N1 and REM, respectively, than week eight. At baseline, the GHB plus modafinil group spent 17 (p = 0.004) and 26.5 (p < 0.001) more minutes in N1 and REM, respectively, than week eight. No significant sleep architectural differences were found when the placebo group or the modafinil group were compared to their respective baseline scores (Black et al., 2009).

Results for the MWT after eight weeks of intervention also showed certain significant results. It was found that compared to baseline, when the placebo group patients started the intervention (modafinil was removed altogether), there was significant decrease in sleep latency. Numerical data was not provided, except for a p-value of < 0.001. Conversely, as compared to their baseline, the GHB plus modafinil group had a significantly elevated median sleep latency period. Numerical data was not provided, except for a p-value of < 0.001. Patients assigned to either the modafinil group or the GHB group failed to demonstrate any significant change in MWT sleep latency, compared to baseline therapy (Black et al., 2009).

Pertaining to the ESS, results showed that the placebo group’s week eight scores were significantly worse than baseline modafinil treatment scores. Exact numerical data was not presented, except for a p-value of 0.011. Lastly, after week eight, there were significant improvements for the GHB group and the GHB plus modafinil group, compared to their baseline. Exact numerical data was not presented, except for p-values of < 0.001 and < 0.001, respectively. There were no significant results for the modafinil group for the ESS (Black et al., 2009).

In a separate study, Black et al. (2016) conducted a retrospective analysis of a Phase-III RCT, which investigated efficacy of: [1] GHB alone (GHB group), [2] modafinil alone (modafinil group), [3] a combination of GHB and modafinil (GHB plus modafinil group), and [4] double-placebo (placebo group). Furthermore, the modafinil group received a GHB placebo, and the GHB group received a modafinil placebo (instead of the pre-interventional ‘authentic’ modafinil). The placebo group received both GHB placebo and modafinil placebo. The study followed a double-dummy protocol, allowing for proper blinding of the treatment regimes. Inclusion criteria consisted of adult age and a diagnosis of narcolepsy as defined by the criterion set by the International Classification of Sleep Disorders, 2^nd Edition (Thorpy, 2012). Patients with both NT1 (n=95) and NT2 (n=127) were included but studied separately. Additionally, each patient took a stable dose of modafinil prior to the onset of the pharmacological intervention (range: 200-600 mg/day). Subjects were GHB-naïve (Black et al., 2016).

Each subject took their respective medication regimen for a total of 56 days. Patients who were assigned to take modafinil (modafinil group and GHB plus modafinil group) remained with their respective personal doses of modafinil, although patients assigned to take GHB (GHB group & GHB plus modafinil group) followed a graded dosage scheme throughout the study. For the first 28 days, a total dosage of six grams per night were ingested. For the following 28 days, a total of 9 grams per night were taken. All administration of GHB (and GHB placebo) was given in two equal doses: the first dose was given at bedtime, and the second dose was administered 2.5-4 hours later. For example, on the 35^th day, each patient received two separate doses of 4.5 grams of GHB (or GHB placebo). This study measured changes over time in narcolepsy symptoms, by administering the ESS and MWT at various times throughout the study. These tests occurred at the beginning of week one (baseline) and at the end of weeks four and eight (Black et al., 2016).

Comparing to baseline ESS to week 8 ESS, Black et al. (2016) found significant change for the GHB plus modafinil group for both NT1, -3.8 (p = 0.002) and NT2, -2.8 (p = 0.015). Standard errors were provided graphically, but numerical values were not reported in this study. In patients with NT2, there were also significant improvements for the GHB group compared to baseline, -3.0 (p = 0.021). After eight weeks, average placebo group ESS scores for both NT1 and NT2 were +0.8 and +0.8, respectively. Compared to this placebo, NT1 patients scored significantly less for both the GHB group, -2.9 (p = 0.011), and the GHB plus modafinil group, -3.8 (p = 0.002). Similarly, for NT2 patients, significant reductions in score were -3.0 (p = 0.021), and -2.8 (p = 0.015), respectively. The modafinil group did not yield statistically significant results whilst being compared to the placebo group for either NT1 patients or NT2 patients (Black et al., 2016).

For NT1 and NT2 patients in the placebo group, MWT results showed that mean sleep latencies were 2.58 (p < 0.05) and 2.90 minutes (p < 0.05) less at week eight than at baseline. Relative to the placebo results, the GHB plus modafinil groups stayed awake for 3.34 (p < 0.05) and 2.16 minutes (p < 0.001) longer at week eight than at baseline. The NT2 GHB group also reported longer mean wakefulness periods of 0.45 minutes (p = 0.007), under these same comparison parameters (Black et al., 2016).

In another study, Dauvilliers et al. (2017) reported similar findings to those of Black et al. (2016). The purpose of this study was to investigate the effects of the same three narcolepsy treatment modalities (and placebo) on sleep stage shifts and disrupted nighttime sleep (DNS). Both parameters directly relate to subjective feelings of sleepiness during the daytime hours. Selection of patients (n= 222) and study group allocation were identical to the Black et al. (2016) study. However, PSG results were analyzed for sleep architecture, rather than ESS or MWT. Specifically, the number of shifts from N2/N3/REM to N1/awake, REM to N1/awake, and N1/awake to REM were measured. Additionally, question #6 of the PSQI was used to grade individual sleep quality, both preceding (baseline) and during (weeks four and eight) the intervention (Dauvilliers et al., 2017).

Results were reported as least-square mean change (LSMC). At week eight, this study found a significant decline, from baseline, in the number of shifts from N2/N3/REM to N1/awake in both the GHB group, -16.5 (p < 0.001), and GHB plus modafinil group, -13.7 (p < 0.001). Again, standard errors were provided graphically, but numerical values were not reported for any results of interest for this study. Furthermore, both changes were greater than those demonstrated by the placebo group, with an insignificant LSMC of -0.6. There were no significant results for this shift in the modafinil group at week eight compared to baseline, or when compared to the placebo group. There was also a significant decline in the number of transitions from REM to N1/awake for the GHB group and GHB plus modafinil group when compared to baseline. LSMC score changes were -6.0 (p < 0.001), and -6.8 (p < 0.001), respectively. The modafinil group, -0.5, did not differ significantly from the placebo group, -0.6. Finally, from baseline to week eight, among the patients who had shifts from N1/awake to REM, both the GHB group, -2.5 (p = 0.083) and GHB plus modafinil group, -2.2 (p = 0.002), showed significantly less frequent shifts per hour compared to placebo, +0.2 (Dauvilliers et al., 2017).

These authors found that quality of sleep, as measured by the PSQI, improved significantly among the GHB group, -0.52 (p < 0.001 from baseline, p < 0.05 from placebo), and the GHB plus modafinil group, -0.49 (p < 0.001 from baseline, p < 0.05 from placebo), but not with modafinil alone, -0.04. For comparison, placebo LSMC at week eight was -0.07 (Dauvilliers et al., 2017).

Discussion

The purpose of this literature review was to explore and compare the efficacy of modafinil and GHB for the treatment of NT1 and NT2. The four previously mentioned journal articles will now be discussed and analyzed in depth to determine supporting and/or refuting evidence for the hypothesis. Additionally, weaknesses and limitations will be considered for each study, as well as areas for potential further research.

Black and colleagues (2006), help prove the hypothesis in certain ways. For example, these authors found that compared to modafinil baseline therapy, the placebo group MWT scores were significantly lower (worse) at week eight, suggesting that the modafinil was providing beneficial effects before it was stripped away. In comparison, for the GHB group, significant score reductions at week eight were not found. Since modafinil was also stripped from this group, it is suggested that GHB can ‘replace’ the missing effects of modafinil.

Additionally, GHB was the only drug that showed efficacy in preventing sleep attacks (cataplexy) and inadvertent daytime naps. This is because both groups which contained GHB treatment showed markedly reduced frequencies of these symptoms when compared to their respective baseline. Neither the modafinil group nor the placebo group showed signs of reductions. This finding is completely consistent with the hypothesis because while GHB shows at-least some efficacy, modafinil fails to show any efficacy for preventing these specific symptoms of narcolepsy.

These authors found that “As determined by MWT measures, sodium oxybate (GHB) alone was as effective as modafinil for the treatment of EDS and significantly more efficacious, as measured by ESS and the number of weekly inadvertent naps or sleep attacks …” (Black and Houghton, 2006). Further, specifically for the MWT, the strongest positive results among any group were for the GHB plus modafinil group. This group received GHB in addition to the existing modafinil regimen, suggesting an additive effect of the two drugs. Sleep attack assessment results also point towards an additive effect. In contrast, additive effects were not found for the ESS tests.

Since this research took place in multiple centers, including 44 states in the USA and an additional seven countries across the globe, this study design allowed for great generalizability for different races, cultures, and climates. However, using multiple international study centers carries its own limitation; each finding was not simply measured, interpreted, and analyzed by one sole researcher, potentially resulting in communication complications (e.g., language barriers). Further, lack of equipment standardization among the different international sites could lead to measurement precision issues (rather than accuracy) of the results.

The study by Black et al. (2009) also evaluated the efficacy of GHB, modafinil and the two drugs in combination for the treatment of narcolepsy. Both groups who took GHB presented with improved amount of time spent in deeper stages of sleep, while modafinil did not change, or seemed to hinder the amount of time spent in deeper stages of sleep.

MWT results for the placebo group point towards modafinil’s positive effects on daytime wakefulness. By removing modafinil, patients unfortunately fell asleep faster during the MWT. At week eight, compared to modafinil-only baseline, the GHB plus modafinil group showed significant improvements in the MWT, while the modafinil alone and GHB alone did not produce significant results. Results from this study are partially in line with the hypothesis. This study would suggest that the most efficacious outcomes are yielded when both GHB and modafinil are used together. Overall, while modafinil has strong effects on wakefulness, it seems that GHB’s sleep architectural-altering properties provide the strongest efficacy.

A limitation of this study is that it does not discern between the different doses of GHB. During the first four weeks of the intervention, patients took 6 g of GHB, while during weeks five through eight, patients took 9 g. This limitation also holds true among each of the four studies in this review. Future studies could investigate differences among doses of GHB (and modafinil) to further refine the search for a better standard of treatment. Black et al. (2009) also speculate that nocturnal EEG changes in the GHB groups (PSG) could truly not be secondary to genuine sleep architecture alterations, but potentially to anaesthetic-like properties of the drug or something different all-together. One final limitation of the study design is that pregnant women or women who did not accept taking birth control were not permitted to participate in this study, limiting the generalizability of the results. Further research is clearly needed to understand the complete mechanism of GHB on sleep architecture.

At the beginning (baseline) of the Black et al. (2016) study, it was also found that subjects with NT1 scored more poorly on the ESS and MWT, as compared to NT2 scores. These differences are important because it helps us recognize the spectrum of disease state presented by a narcolepsy diagnosis. However, once the drugs were added, many results for both NT1 and NT2 yielded very similar results, suggesting that GHB and modafinil have similar effects across this spectrum of narcolepsy symptoms. The authors effectively displayed and discussed how GHB and modafinil effect the ESS and the MWT. GHB alone, and GHB in combination with modafinil resulted in improvements in excessive daytime sleepiness (ESS and MWT) relative to the placebo. Modafinil on its own however did not yield any significant results in this parameter. Also, the fact that the GHB plus modafinil group showed the strongest improvement suggests that these drugs have an additive effect. Overall, these findings are consistent with the hypothesis of this review, and consistent with findings of the other reviewed articles. This study earns merit, as discussed above, because it objectively stratifies for presence of cataplexy, allowing for greater generalizability of results across the narcolepsy spectrum. However, certain limitations of this study also exist. The authors explain that differences in narcolepsy diagnosis measures among patients could have led to misclassification of patients with NT1, into the NT2 group, and vice-versa. It is explained that different methods of diagnosing NT1 versus NT2 (e.g., sleep-onset REM periods on PSG, hypocretin levels, HLA-DQB1*06:02 positivity, etc.) can lead to this misclassification. The authors do however acknowledge that this is an unlikely phenomenon. If misclassification did occur, it is also unlikely to majorly affect the findings since the sample size (n = 222) is relatively high.

In the paper written by Dauvilliers et al. (2017), various sleep architectural changes were measured using PSG. It was found that GHB (with and without modafinil combination) created more condensed and efficient sleep for patients in this study arm. For the N2/N3/REM to N1/awake shift, patients taking GHB alone demonstrated the strongest decline of incidence. In the group who took GHB and modafinil, significant reductions were also found, but to a lesser degree. This would make one think that these results are consistent with the hypothesis. Although this is likely true, it must be considered that modafinil is a wakefulness promoting drug, and it does not produce any sedative effects. It is most likely that the addition of modafinil to a GHB regimen creates stimulation during the day but has a somewhat detrimental residual effect on sleep architecture during the night. Although GHB and modafinil have opposing mechanisms, the summation of their gross effects (with proper circadian rhythmic timing) is to reduce the severity of narcolepsy symptoms during the day. It is also important to consider that higher concentrations of GHB stimulate the GABA_B receptors, creating an overall neuro-depressive effect, while lower concentrations will stimulate the GHB-specific receptors, causing an excitatory effect. One should understand that since GHB possesses multiple concentration-dependent mechanisms, it is possible that alterations in the equilibrium of this system could be an important player in future research moving forward.

The authors effectively combined these objective measures (PSG) and subjective measures (PSQI) allowing for the reader to speculate that the sleep architectural changes translate into the subjective feelings of a good night’s rest. By manipulating the sleep architectural changes, we are potentially presented with an exciting way to improve sleep quality in patients with narcolepsy.

As seen with each of the other three studies in this review, Dauvilliers et al. (2017) study length was too short to generalize results beyond a period of two months. Future studies should evaluate whether these promising effects remain robust throughout a period of one year (or more). In general, a large proportion of drugs display phenomena such as tolerance or sensitization and pharmacodynamic changes, so this is a crucial topic to investigate regarding GHB and modafinil. Unlike the study done by Black et al. (2016), this study was limited in that it did not focus on the presence of cataplexy as a stratification parameter. In future studies, it would be important to assess the differences of sleep architecture in patients with NT1 and NT2.

In general, these four studies exhibit certain similar limitations due to similar study designs, since all came from related laboratory groups with overlapping authors, therefore weakening any clear scientific conclusion. For example, the results for modafinil may have been overrepresented due to a selection bias. Since pre-trial modafinil therapy was required as a selection criterion for each of the studies, it is likely that more patients who respond positively to modafinil were included as study subjects. This is because prospective patients who had previously failed to experience the benefits of modafinil had simply stopped taking the drug altogether before they could have been considered for inclusion. It is also possible that results for modafinil were underrepresented, as discussed by Black and Houghton (2006), because patients having previously taken modafinil before the start of the trial. Extending this idea, there is a possibility of some patients not being sufficiently medicated with modafinil before the commencement of the trial. This potential under-dosing could spill over into the RCTs and under-represent any positive findings for the modafinil group. The root of this problem is the patients themselves (or their third-party physicians) deciding the “proper” dose of modafinil (varying from 200-600 mg/day). Ideally, researchers of the reviewed journals would have assessed the modafinil dosages for each patient to correct any obscurities/incongruencies before the commencement of the trials. Future extensions could also explore additional drug combinations (i.e., pitolisant, methylphenidate, etc.) and dosage timelines.

Conclusion

This review of the literature uncovered four primary journal articles to assess efficacy of GHB and modafinil. Through the conglomeration and analysis of the subjective and objective findings, modafinil and GHB both proved effective for treating excessive daytime sleepiness associated with narcolepsy. While GHB seems to alter the sleep architecture, creating less disruption and more sleep-stage consolidation, modafinil works as a stimulant to augment arousal during the day. Collectively, these four studies suggest that combinations of these drugs may create the best possible treatment for patients with excessive daytime sleepiness due to narcolepsy. It is also somewhat suggested that GHB monotherapy is superior to that of modafinil, although more research is needed to confirm that conclusion.