According to Bastuji, Jouvet himself experienced excessive daytime sleepiness and fatigue in the 1980s but could not be treated with modafinil, because of strict conditions on its use.16 In France, the drug became commercially available in 1994. Initially, it could only be prescribed by a neurologist and was only available through hospital pharmacies.17

  Jouvet’s research established an apparent distinction between modafinil and conventional stimulants.18 Giving modafinil to animals kept them awake longer but did not lead to agitation or rebound sleep afterward. Jouvet found the drug did not activate the same neurons as amphetamines and did not seem to act via dopamine. These would become key selling points.

  SELECTIVE ACTIVATION

  The Pennsylvania-based company Cephalon licensed the rights to modafinil from Lafon in 1993 and organized the clinical trials necessary for FDA approval. These studies set the standards for other wake-promoting medications. The sleep medicine field had established objective or physiological measures of sleepiness, and these came into play for modafinil’s clinical trials. FDA officials have occasionally expressed a wish for “real-world” sleepiness outcomes, such as the frequency of unplanned naps, but clinical trials have generally not relied on those types of measures for approval.19

  For modafinil, the primary measures of efficacy were the Maintenance of Wakefulness Test (MWT), which was more responsive to treatment than the MSLT, along with the subjective Epworth Sleepiness Scale (ESS).20 Headache and nausea were the most common side effects. Although modafinil did not eliminate daytime sleepiness, most people in these studies saw a significant decline in symptoms. In these studies, most of the more than five hundred participants had narcolepsy with cataplexy: more than 85 percent in one, 70 percent in the other.

  In addition to demonstrating efficacy in narcolepsy, Cephalon was required by the FDA to assess modafinil’s abuse potential. For people accustomed to stimulants, it could weakly substitute for more potent drugs; it produced more inhibition of sleep and evoked less pleasure.21 Modafinil’s physical properties also made it difficult to inject or smoke. Taken together, these factors explain why it received a Schedule IV designation from the Drug Enforcement Administration—with fewer restrictions on phone-in prescriptions and refills, in comparison with methylphenidate and amphetamines.22

  Modafinil became commercially available in the United States in 1999. Its emergence excited researchers who were probing which parts of the brain were activated by the drug. “This is the first drug in history to selectively promote wakefulness in a way analogous to how the brain normally wakes up,” the neuroscientist Dale Edgar told the New York Times.23

  Contradicting Jouvet’s previous research, Edgar and others at Stanford revealed that modafinil weakly increased dopamine levels in the brain.24 The findings continued to show a distinction between modafinil and conventional stimulants. In mice, modafinil appeared to selectively activate cells in the hypothalamus, including those that produce hypocretin, while avoiding cells in reward centers activated by amphetamines.25 Pharmacologists currently regard modafinil as an “exceptionally weak, but apparently very selective” dopamine transporter inhibitor.26

  FORK IN THE ROAD

  With their eyes on a larger market, Cephalon executives began exploring what would be necessary to expand the drug’s indication to include other conditions, such as sleep apnea and neurological disorders. At one point, FDA officials suggested that they might be willing to grant “a general claim for the treatment of EDS [excessive daytime sleepiness] if it could be shown that PROVIGIL had an effect on this symptom regardless of the clinical setting in which it occurred.”27

  This was a fork in the road, with significant implications. Being able to treat EDS without regard to underlying cause would include being able to treat EDS with an unknown neurological cause. Also, it highlights the distinction between symptom and disease. Drugs for transient symptoms such as headache, joint pain, or mild insomnia are available over the counter, and several prescription medications are approved for insomnia without specifying that a cause must be identified.

  However, for sleepiness, the regulators’ stance was that anything beyond caffeine or dietary supplements should require a prescription, and the FDA decided a more specific diagnosis than EDS would be required as well. The agency told Cephalon that it would need to demonstrate efficacy in well-defined groups of people. Some officials could see what was coming. The documents prepared for the FDA’s advisory committee contain this comment from Paul Leber, director of Neuropharmacological Drug Products: “One concern, however, is that ‘hype’ about modafinil’s capacity to promote arousal and vigilance without the untoward stimulant-like effects of amphetamine may promote its use in a number of non-orphan conditions, notably in children with ADHD. If this were to occur, the major use of modafinil might well become an off label use.”28

  Cephalon was gearing up to promote Provigil more broadly. A January 2002 warning letter from the FDA said that the company’s brochures and advertisements in medical journals were misleading, giving the impression that “Provigil can be used to improve wakefulness in all patients presenting with symptoms of daytime sleepiness, characteristic of generalized sleep disorders, whether or not they have narcolepsy.”29

  Critics have labeled Cephalon’s marketing as a classic example of disease mongering: conflating the symptom of excessive daytime sleepiness, which many people experience, into a disease requiring a remedy. One Arizona doctor complained in 2009 that “excessive sleepiness was a little-known term to primary care physicians” until modafinil’s introduction.30

  A LARGE, NATIONAL EXPERIMENT

  By 2001, sales of modafinil had shot up to $150 million, mostly going to people who didn’t have narcolepsy.31 Articles appeared in magazines such as the New Yorker about modafinil’s potential, suggesting that it may be possible to drastically limit someone’s need for sleep.32 “There are no warts on this drug,” Frank Baldino Jr., Cephalon’s hard-driving CEO—who reportedly used it for jet-lag—told the New York Times. “The only question is how big we can make it.”33

  At the time, Cephalon was conducting studies of modafinil in both obstructive sleep apnea and shift work sleep disorder. The company donated money for an endowed professorship at Harvard for one of the leaders of the shift work study, Charles Czeisler.34 Because standard CPAP treatment of sleep apnea can leave people with residual sleepiness, sleep apnea was an obvious choice, but shift work sleep disorder was more curious as an indication. It was circadian misalignment imposed by someone’s job. Do truck drivers or night-shift nurses necessarily have a sleep disorder?

  For shift work, a case for symptom management could be made based on public safety. At night, participants in the shift work studies had the same level of objective sleepiness as people with narcolepsy or IH do during the day. That is, they fell asleep in an average of two minutes during a nighttime version of the MSLT. By this measure, modafinil made them somewhat less sleepy, nudging them up to four minutes. It cut in half the number who reported accidents or near-accidents while commuting home. Yet the authors noted: “Although modafinil improves the measured levels of performance, it is far from what is needed for these patients to function at a normal level.”35

  Several in the sleep medicine community expressed caution about modafinil’s promotion. “What I worry about more is the issue of somebody who wants to succeed in business and work 100-hour weeks, and the way to do that is to pop a modafinil,” said the University of Pittsburgh sleep researcher Timothy Monk, who was studying the drug’s temporary use by astronauts. “That is a cause for concern, given that we don’t know the effects of only having four hours of sleep and then avoiding the negative sleepiness with a pill.”36

  A series of studies conducted by U.S. Army researchers concluded that modafinil does not appear to offer advantages over caffeine for improving performance and alertness in healthy adults lacking sufficient sleep.37 In a similar vein, a 2003 editorial from the University of Pennsylvania sleep researcher Allan Pack argued against the prospect of a broad FDA approval for excessive sleepiness.38 Simply giving someone a pill for sleep apnea’s symptoms without treating their respiratory problems would put them at higher risk of heart attack and stroke, he wrote. He added that in healthy people, encouraging them to sleep less with the help of modafinil also had risks. “If modafinil is used for this purpose, we will be engaged in a large, national experiment without data to support it,” Pack wrote.

  That fall, Cephalon almost got what its executives wanted. An FDA advisory committee split evenly on whether there was enough evidence for a broad label for excessive sleepiness, and the agency chose the narrower path. (Emmanuel Mignot was among the “Yes” votes on this question.)39 “We’re talking about treating a symptom without understanding the many possibilities that lead to that symptom,” said Advisory Committee Chair Claudia Kawas, who voted “No.”

  This episode showed what regulators and some doctors were not quite ready for. If Cephalon had succeeded, that would have made it easier for people with idiopathic hypersomnia—and multiple sclerosis and other conditions—to obtain modafinil prescriptions and have them covered by insurance.

  Another boundary was set in 2010 when the FDA declined to approve armodafinil (Nuvigil), a variation of modafinil, for the acute, transient condition of jet-lag.40 Cephalon brought armodafinil forward to get around modafinil’s expiring patent. Modafinil consists of a mixture of two mirror-image chemicals, while armodafinil is just one of the two, and armodafinil is supposed to last longer in the body. Neither drug shifts sleep-wake cycles—they only help someone stay awake.

  SKIN IN THE GAME

  Starting in the 1990s, whistleblowers and later the federal government pursued several cases against pharmaceutical companies for off-label promotion, extracting large settlements and fines. Several major pharmaceutical companies—including Pfizer, Eli Lilly, Bristol Myers Squibb, AstraZeneca, Abbott, and Johnson & Johnson—ran into this type of trouble, indicating how widespread the practice of off-label promotion was.

  While Cephalon’s promotion of an opioid lollipop received more press attention, modafinil was part of a set of whistleblower complaints alleging that the company was courting off-label prescriptions. Its sales representatives allegedly used exploratory studies of modafinil in several conditions, such as depression, multiple sclerosis, and Parkinson’s, as promotional material and aggressively pushed for modafinil prescriptions, especially among psychiatrists.41 The complaints were eventually settled in 2008 for more than $425 million.42

  Along the way, Cephalon’s efforts to have modafinil approved for children and adolescents with ADHD (attention deficit hyperactivity disorder) ran aground in 2006 because of reports of serious skin rashes. Three cases of severe skin rashes occurred among 1,236 children and teens participating in clinical trials, according to FDA briefing documents.43 These were severe enough to notify a doctor, but they did not require medical intervention or hospitalization. At least one was considered by dermatologists to be Stevens-Johnson syndrome, a rare life-threatening skin disease involving widespread red or itchy blisters. Eight other patients discontinued modafinil because of rash or hypersensitivity. Cautiously, FDA advisors voted that modafinil should not be approved for pediatric use.44

  Several other drugs have been linked to Stevens-Johnson syndrome, including NSAIDs (nonsteroid anti-inflammatory drugs), antibiotics, and antiepilepsy medications. It is suspected to be like an allergy—the immune system overreacts to the drug or one of its metabolites. For some drugs, the condition is linked to certain HLA alleles, suggesting an analogous mechanism to narcolepsy type 1 and other autoimmune diseases.45

  In 2007, the FDA issued a safety alert because of reports of serious skin rashes developing in reaction to modafinil in adults. In postmarketing monitoring for modafinil, six cases of severe skin reactions were reported to the FDA.46 The skin rash reactions, together with concern about adverse reactions related to psychiatric disorders, prompted the United Kingdom to request a European Medicines Agency review.47 The data subsequently compiled by the EMA was more serious than what had been reported to the FDA. The EMA cited sixteen cases of Stevens-Johnson and related skin reactions occurring in a postmarketing setting, three of which were fatal.48

  BENEFIT-RISK RATIO

  In 2010, the European Medicines Agency withdrew approval for modafinil for sleep apnea, shift work sleep disorder, and idiopathic hypersomnia. The decision meant doctors in European countries could only prescribe modafinil for IH off label, and it would not be reimbursed by national health insurance programs. Previously modafinil was explicitly approved for IH in four European countries: France, Poland, Sweden, and Norway. Sleep specialists in Europe quietly disagreed with the EMA’s decision and continued to consider modafinil a first-line treatment for IH.49 The decision was also criticized with regard to sleep apnea: “Therefore, the removal of modafinil for this indication appears to have been an unfortunate decision by the EMA, particularly when the major pharmaceutical alternatives are amphetamine derivatives, which do not have a better safety profile and are well-known potential drugs of abuse.”50

  One aspect of the EMA’s review deserves emphasis. When discussing clinical studies, the reviewers prioritized objective measurements of patients’ sleepiness, based on tests like the MWT, rather than subjective questionnaire-type data. When the objective effects were small, such as in sleep apnea, the benefits were considered not “clinically relevant” and not worth the risks of psychiatric adverse reactions and skin rashes. Only in narcolepsy was the benefit/risk ratio acceptable, the agency said.

  The studies submitted to the EMA only included a handful of IH patients, so the reviewers concluded that the condition was very rare. The scarcity of information spurred clinical researchers in France and Germany to perform two small clinical trials of modafinil that included people with idiopathic hypersomnia. In both of these studies, several people were already being treated with modafinil or methylphenidate before going through a washout period. The French study included a driving test, when participants drove a 230-kilometer highway trip, accompanied by a professional driving instructor.51 It showed that modafinil could improve driving performance in people with narcolepsy type 1 and IH, cutting in half the number of inappropriate line crossings.

  The study in Germany was the first placebo-controlled trial anywhere to focus on IH, although it recruited people with IH diagnoses who did not require long sleep. A few of the thirty-three participants slept only five or six hours on weeknights, while sleeping longer on weekends.52 Modafinil improved participants’ subjective sleepiness, as measured by the Epworth Sleepiness Scale, but there was no significant difference in sleep latency, as measured by the MWT. The study’s organizer, Geert Mayer, said he sent the results to the EMA and never received a response. Reversing the EMA’s decision on modafinil for IH may require a larger number of patients—or a group with more uniform sleep habits. A larger Japanese study of modafinil published in 2021, also focusing on people with IH without long sleep times, did show a significant effect on participants’ sleep latency.53

  While modafinil has been studied extensively for its ability to enhance cognitive performance in healthy adults, there is very limited data on its cognitive or performance-related effects for people with IH, such as the French driving study. One recent review concluded: “Data with these stimulants is far from positive if we consider that effects are small, in experiments that do not accurately reflect their actual use in the wider population. There is a user perception that these drugs are effective cognitive enhancers, but this is not supported by the evidence so far.”54

  NEWER OPTIONS

  Even as newer drugs are approved for IH, modafinil will continue to be considered as a first-line option, because neurologists and sleep specialists have the most experience with it and because modafinil and armodafinil are currently available as inexpensive generics. That hierarchy could change as prescribers gain more experience with other medications (table 14.1). Two newer wake-promoting medications became available in the United States in 2019: Jazz Pharmaceuticals’ Sunosi (solriamfetol) and Harmony Biosciences’s Wakix (pitolisant). Their applications have not raised grand policy questions at the FDA in the way that modafinil did—so far. That’s because modafinil had established a precedent, and neither manufacturer sought to expand the boundaries of what their medications could be prescribed for.55 The FDA approved Sunosi for the indications of narcolepsy type 1 or type 2 and obstructive sleep apnea, while Wakix was approved only for narcolepsy.

  Sleep specialists seemed to view solriamfetol as an extension of drugs they were already familiar with, since it works by enhancing signals from dopamine and norepinephrine.56 Like modafinil, solriamfetol did appear to increase blood pressure slightly, a potential drawback for cardiovascular health. The Korean company SK Biopharmaceutical had originally developed solriamfetol as an antidepressant and licensed it to Johnson & Johnson, but an initial clinical trial in depression was not successful.57 After reporting a side effect of insomnia, the Korean firm reframed solriamfetol for sleep disorders and sought out Jazz as a partner, leading to the drug’s eventual approval.

  In contrast, pitolisant was distinct from many previous wake-promoting medications, because it did not act through dopamine. Instead, it increases the availability of the neurotransmitter histamine, representing the opposite of the drowsiness induced by antihistamines. Pitolisant is a histamine H3 receptor antagonist, and its main targets are the histamine-producing neurons of the tuberomammillary nucleus. Other groups of neurons are stimulated too, because the drug weakens an inhibitory receptor present on neurons that do not make histamine themselves. The effect on other functions of histamine—in allergic reactions and in the stomach and heart—is limited, because the H3 receptor is mostly found in the brain.58