Treat the Symptom, Not the Cause? Pitolisant for Sleepiness in Obstructive Sleep Apnea

Pitolisant, an antagonist/inverse agonist of histamine H3 receptors, is a novel wake-promoting medication. It was recently approved by the U.S. Food and Drug Administration (FDA) for the treatment of sleepiness due to narcolepsy and has been available in the United States since November 2019. Because it is so new, and because it is a first-in-class drug, sleep medicine clinicians are early in the process of developing acumen about its applications (which patient, which dose, and when). Controlled clinical trials are very welcome in this regard.

In this issue of the Journal, Dauvilliers and colleagues (pp. 1135–1145) report data from a trial assessing the use of pitolisant for a non–FDA-approved indication, treatment of sleepiness in people with obstructive sleep apnea (OSA) who refuse first-line treatment with continuous positive airway pressure (CPAP) (1). This industry-sponsored trial randomized 268 people with OSA and an average apnea–hypopnea index (AHI) of 49, half of whom had comorbid cardiovascular disease, to pitolisant or placebo in a 3:1 ratio. This study has multiple strengths. It was rigorously designed, with central electronic randomization, matched pitolisant and placebo capsules, blinding of participants and outcome assessors, adequate power for the primary outcome, two independent statisticians, and a multisite design across 28 sites in 10 countries. Pitolisant was significantly better than placebo at reducing subjective sleepiness. It was also well tolerated, with similar rates of treatment-related adverse events observed for pitolisant and placebo, and only rare prolongation of QT intervals. Despite these strengths, the work by Dauvilliers and colleagues leaves several key unanswered questions for clinicians considering prescribing pitolisant.

First and foremost is the question of whether it is even appropriate to medicate OSA-induced sleepiness without treating the underlying OSA. OSA is a multisystem disease whose treatment may have health benefits beyond symptom control, in particular, reductions in blood pressure (2). As a result, the goal of OSA treatment is not only resolution of symptoms but also normalization of disease metrics such as the AHI (3). Although first-line CPAP is sometimes not tolerated, multiple other treatment options are recommended in clinical practice guidelines depending on the clinical situation, including mandibular advancement devices, surgical procedures in carefully selected patients, and weight loss (3–6). If sleepiness is improved by medication, it seems likely that people with OSA will have less motivation to pursue careful CPAP troubleshooting or effective CPAP alternatives, and the potential opportunity to improve their health will be lost. This is akin to treating insufficient sleep syndrome with a wake-promoting medication. The sleepiness and fatigue caused by insufficient sleep could be improved with wake-promoting medications, but the other physiologic benefits of sleep would be lost. Extending sleep durations can be very challenging because of family, work, and social obligations, but the benefits of sleep are worth this hard work. Similarly, however safe pitolisant may be, it is unlikely to be healthier than restoring normal breathing during sleep.

There is also a question regarding the clinical importance of this statistically significant benefit. Despite widespread use of the Epworth Sleepiness Scale (ESS) in clinical trials and clinical practice, there is still uncertainty about how much the ESS scores must change to represent a meaningful improvement. Dauvilliers and colleagues prespecified a minimal important difference of 3 points (1). The observed benefit of pitolisant was a modest 2.8 points and thus did not meet this threshold, raising the possibility that the average benefit of pitolisant for OSA is not meaningful. However, it can be challenging to determine minimal important differences (7), and other work has suggested that the minimal important difference in ESS for patients with OSA falls between 2 and 3 points (7, 8). Studies assessing other pharmacologic treatments of OSA sleepiness have shown similar reductions in ESS compared with placebo (1.7–4.5 points more than placebo with solriamfetol [9], 3.0 points more with modafinil [10], and 2.8 points more with armodafinil [10]), although some patients in these studies were also treated with CPAP, potentially creating a treatment ceiling effect. Supporting the argument that pitolisant’s effect may be clinically significant, Dauvilliers and colleagues found improvements in several subjective secondary outcomes, including fatigue and patient and clinical global impressions.

What about objective measures of sleepiness? In Dauvilliers and colleagues’ study, the Oxford Sleep Resistance Test was the only objective measure of sleepiness, and scores were not significantly improved with pitolisant (1). Similarly, objective cognitive performance was not improved. This differs from the use of pitolisant in narcolepsy, which was shown to result in a modest improvement of 1.5 minutes on the maintenance of wakefulness test (11). The absence of an objective benefit does not necessarily imply that pitolisant is ineffective for OSA sleepiness, because objective and subjective tests measure different aspects of the construct of sleepiness (12). However, objective tests such as the maintenance of wakefulness test may better predict important outcomes, such as driving safety (13). It is possible, but speculative, that sleepiness-related safety concerns may remain even after subjective sleepiness improves in patients with OSA treated with pitolisant, which would have important implications for decisions about whether or not to use this treatment strategy.

Finally, the maximum dose of pitolisant tested in Dauvilliers and colleagues’ study was 20 mg, reached by approximately 85% of the participants (1). This is half the maximum dose tested in narcolepsy studies (11, 14) and lower than the maximum FDA-approved dose for narcolepsy of 35.6 mg. It is unclear whether higher doses might yield a more impressive benefit for sleepiness, and if so, at what cost of increased side effects.

Despite these remaining questions, in combination with prior work (11), this study by Dauvilliers and colleagues now establishes pitolisant as a treatment for sleepiness across at least three different pathophysiologic causes: hypocretin deficiency, OSA, and the as-yet-unidentified mechanism of narcolepsy type 2. Although this does not imply that it will be similarly effective for all causes of excessive daytime sleepiness, further studies of pitolisant are clearly warranted. This is particularly true for hypersomnolence disorders in which current treatment options are limited, such as idiopathic hypersomnia and hypersomnolence associated with medical or psychiatric disease, where preliminary clinical observations suggest that pitolisant can reduce sleepiness in people with idiopathic hypersomnia (15) or Prader-Willi syndrome (16). As such, this demonstration of a benefit from pitolisant for OSA-related sleepiness is an important step in the development of effective treatments across a wide range of sleepiness-producing disorders.