Effects of Pitolisant, a Histamine H3 Inverse Agonist, in Drug-Resistant

Sleep Medicine 15 (2014) 681–687

Contents lists available at ScienceDirect

Sleep Medicine

journal homepage: www.elsevier.com/locate/sleep

Original Article Effects of pitolisant, a histamine H3 inverse agonist, in drug-resistant idiopathic and symptomatic hypersomnia: a chart review ⇑ Smaranda Leu-Semenescu a,b,c, , Nandy Nittur b, Jean-Louis Golmard d, Isabelle Arnulf a,b,c a National Reference Centre for Narcolepsy and Idiopathic Hypersomnia, France b AP-HP, Hôpital Universitaire Pitié-Salpêtrière, Service des Pathologies du Sommeil, Paris, France c Pierre and Marie Curie University, Centre de Recherche de l’Institut du Cerveau et de la Moelle épinière, Inserm U 1127, CNRS UMR7225, Paris, France d Department of Biostatistics, Hôpital Universitaire Pitié-Salpêtrière, ER4, Pierre and Marie Curie University, Paris, France article info abstract

Article history: Objective: To evaluate the benefits and risks of pitolisant (a wake-enhancing drug that increases the his- Received 13 September 2013 tamine release in the brain by blocking presynaptic H3 histamine reuptake) in patients with idiopathic Received in revised form 10 December 2013 (IH) and symptomatic (SH) hypersomnia plus sleepiness refractory to available stimulants (modafinil, Accepted 7 January 2014 methylphenidate, mazindol, sodium oxybate, and D-amphetamine). Available online 18 March 2014 Methods: Through retrospective analyses of patient files, the benefit (the score from the Epworth Sleep- iness Scale [ESS], authorization renewal) and tolerance (side-effects) of pitolisant were assessed. Keywords: Results: A total of 78 patients with IH (n = 65%, 78% women) and SH (n = 13%, 54% women) received pitol- Idiopathic hypersomnia isant 5–50 mg once per day over the course of five days to 37 months. The median (interquartile range) Pitolisant BF2.649 ESS scores of patients with IH decreased from 17 (15.5–18.5) to 14 (12–17). There were 36% responders H3 receptor antagonist (ESS fall of P3). The improvement in ESS score (À1.9 ± 2.6) was different from 0 in IH without long sleep Histamine time (P < 0.002) and in IH with a long sleep time (P < 0.0001), but not in SH. Forty-four (63%) patients Sleepiness with IH and 12 (77%) patients with SH stopped pitolisant, mostly due to a lack of efficacy. Side-effects included gastrointestinal pain (15.4%), increased appetite and weight gain (14.1%), headache (12.8%), insomnia (11.5%), and anxiety (9%), as well as exceptional reports of depression and persistent genital arousal. Conclusion: Pitolisant had a long-term favorable benefit/risk ratio in 23–38% of drug-resistant patients with IH and SH, suggesting that histamine neurons can be stimulated in severe idiopathic and symptomatic hypersomnia. Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction abnormally stimulated by an unknown endogenous substance in some cases [4]. Cerebrospinal fluid histamine levels are reduced Idiopathic hypersomnia (IH) is a rare (0.002–0.02% of the popu- [5,6] or unchanged in instances of IH [7], but they are also reduced lation) and disabling disease that affects mostly young subjects [1]. in narcolepsy with or without hypocretin deficiency, suggesting In addition to suffering from severe daytime sleepiness, most that low histamine levels signal rather than cause somnolence of patients report major sleep inertia, impaired daytime alertness a central origin. In addition to these idiopathic central hypersom- (the feeling of never being really awake), and altered cognition nias, some hypersomnias that are symptomatic of diencephalic with automatic behaviors [2,3]. Two forms of the disease are recog- lesions (e.g. neurolupus and craniopharyngioma) and medical dis- nized: one with and one without long (>10 h) night-time sleep. The eases have been described [1]. cause of this disease is unknown. The hypocretin system, which is Because IH is rare, no randomized controlled therapeutic trials deficient in narcolepsy/cataplexy, is unaffected in patients with IH, have been performed. As a consequence, no stimulant has received whereas the gamma-aminobutyric acid (GABA) system may be current approval from drug agencies for the purposes of treating IH. At present, clinicians treat IH-associated sleepiness with the same treatments as those used in narcolepsy [8]. Modafinil has a good benefit/risk ratio in IH with or without long sleep time, similar ⇑ Corresponding author at: Service des Pathologies du Sommeil, Hôpital Pitié Salpêtrière, 47-83 boulevard de l’Hôpital, 75651 Paris Cedex 13, France. Tel.: +33 to its effects in narcolepsy/cataplexy [9]. However, modafinil and 142167702; fax: +33 1 42167700. methylphenidate provide, at best, only moderate improvement in E-mail address: [email protected] (S. Leu-Semenescu). sleepiness in patients with IH [10]. Reports of sodium oxybate use http://dx.doi.org/10.1016/j.sleep.2014.01.021 1389-9457/Ó 2014 Elsevier B.V. All rights reserved. 682 S. Leu-Semenescu et al. / Sleep Medicine 15 (2014) 681–687 in IH are exceptional [10]. Plus, some patients have a severe sleepi- 2010 to March 2013 and who used the drug for at least a single ness refractory to these drugs and require another therapeutic day. To be eligible for an individual ATU of pitolisant, delivered option. Dextroamphetamine has been approved by the US Food by the French Drug Agency (Agence Nationale de Sécurité du and Drug Administration for the treatment of narcolepsy and is also Médicament et des produits de santé), the patients must have suf- used in patients with IH in the UK and the USA [10,11]. Pemoline has fered symptoms resistant to stimulants that were approved for been withdrawn from the market due to concerns about the risk of narcolepsy given at the highest tolerable dosage. In France, modafi- liver toxicity [10,12]. Mazindol and D-amphetamine can be obtained nil, methylphenidate and sodium oxybate are marketed for narco- in rare cases in France on a named patient basis after approval by the lepsy and can be used by analogy in primary and secondary French Drug Agency [13]. Although all of these treatments are active, hypersomnias of central origin. Mazindol [13] and D-amphetamine further medication improvements are needed in terms of efficacy, are imported on an individual basis and are delivered by the hos- ease of use, and safety. pital pharmacy after the ATU has been obtained. These criteria Pitolisant (BF 2.649, previously named tiprolisant; Bioprojet for eligibility are strict and must be approved by two anonymous Ltd., Paris, France) is a new histaminergic stimulant. This inverse experts in the field after they have reviewed the medical case. agonist of the histamine H3 autoreceptor increases histamine The individual ATU procedure is aimed at making some drugs release in the hypothalamus and cortex. Pitolisant increases wake- that are currently absent from the national market available to a fulness and decreases narcoleptic episodes in orexin knockout limited number of patients with resistant, severe and orphan dis- mice. In a single-blind trial, it reduced excessive daytime sleepi- eases for the sake of human compassion. This procedure requires ness in patients with narcolepsy/cataplexy [14]. In a recent large, the treating physician (who should be experienced in the field randomized controlled study of pitolisant (versus modafinil and and a member of a sleep center in a university hospital) to com- versus placebo) in 95 patients with narcolepsy, the Epworth Sleep- plete a detailed, anonymous form listing the patient’s age, sex, iness Scale (ESS) score improved by À5.8 with pitolisant, a weight, diagnosis, medical history, severity of the disease, and list decrease which was non-inferior to modafinil [15]. We first of maximum dosages and effects (benefit and side-effects) of drugs reviewed the charts of four teenagers with resistant narcolepsy already used to alleviate sleepiness along with the reason for who benefited from pitolisant treatment [16]. Since 2010, because requesting the authorization. The form is sent to the drug agency the drug had been tested in phase III trials and one long-term via the local hospital pharmacy and examined by two anonymous open-extension trial for narcolepsy, it was available on a named expert referees (one physician and one pharmacist) who decide patient basis to patients with IH experiencing severe sleepiness whether to authorize the administration of the drug for three refractory to available stimulants after individual authorization months. In particular, they require the knowledge of whether the from the French Drug Agency. We studied the benefits and risks maximum stimulant dosages have been tried in these patients of pitolisant in our patients with IH and symptomatic hypersomnia and whether intolerable side-effects were observed. Their decision over the long term using a chart review process. is based on a major resistance to marketed drugs given at the highest tolerated dosage. If refractory sleepiness is the motive for requesting pitolisant, then the symptoms should be refractory to 2. Methods modafinil 600 mg/day, methylphenidate 80 mg/day, and sodium oxybate 9 g/day alone or in combination. The liability is shared 2.1. Subjects between the French Drug Agency and the physician prescribing the non-marketed drug (who engages his or her liability with a The files of all patients diagnosed with IH and symptomatic written signature). Every three months, the physician must report hypersomnia at the University Hospital of Pitié-Salpêtrière (Paris) on the benefits and side-effects of the drug to the agency through a and treated with pitolisant were retrospectively collected. The form and dependening on this information, can obtain a renewal of study was approved by the local institutional review board. All of the authorization for use. The hospital pharmacy buys the drug the patients had undergone an attended nocturnal polysomnogra- from the pharmaceutical company and sells it directly to the phy followed by five multiple sleep latency tests (MSLTs) the next patient, and the price is paid by the national health insurance. day. In most cases, the tests were followed by another 24 h long- The patient (and, if a minor, their parents) receives the information term sleep monitoring, as previously described [2]. The patients and accepts it verbally (no signed consent is required in the ATU had been previously diagnosed with IH because they met the fol- according to the French law). They must report to the physician lowing criteria: (i) complaint of excessive daytime sleepiness every three to six months for renewal of the ATU. occurring daily for more than three months and (ii) hypersomnia In France, pitolisant was given through Phase II and Phase III tri- that was not better explained by another sleep disorder, medical als organized by the pharmaceutical company Bioprojet to patients or psychiatric condition, medication use or substance abuse, or with narcolepsy from 2007 to 2013, but never to patients with IH. behaviorally induced insufficient sleep syndrome. In addition, the Because of a favorable side-effect profile and recent concerns about IH patients without long sleep time had mean sleep latencies dur- the use of amphetaminergic and pseudo-amphetaminergic stimu- ing the MSLT of <8 min and no more than one sleep onset REM lants having anorectic properties (reviewed in [13]), pitolisant (rapid eye movement) period. The IH patients with a long sleep was authorized for individual use via the ATU process in IH refrac- time had a total sleep time >660 min (i.e. 11/24 h) during long- tory to usual stimulants beginning in April 2010. term sleep monitoring or a night-time sleep time >600 min plus mean daytime sleep latencies of <8 min for those who had no 2.3. Clinical measures and outcome long-term sleep monitoring. We also included patients with hypersomnia symptomatic of a concomitant disease who met the same The IH and symptomatic hypersomnia (SH) criteria were clinical and polysomnographical criteria as patients with IH and checked. Demographic and clinical measures were collected retro- who also used pitolisant. spectively, including age, sex, weight, diagnosis, date of disease onset, date of diagnosis, disease course at time of investigation, 2.2. Process for being authorized to be given pitolisant baseline score on the ESS performed when the patients were not on any treatment, and time of sleep monitoring. The treatment his- The files were collected for patients who benefited from at least tory was also collected, including date, duration, type, and maxi- one temporary authorization for use (ATU) of pitolisant from April mum dosage of each treatment (and combination when S. Leu-Semenescu et al. / Sleep Medicine 15 (2014) 681–687 683 appropriate) and spontaneously reported side-effects. The ESS Table 1 value was obtained under pitolisant at the maximum dosage. The Demographic and clinical characteristics of patients with idiopathic (IH) versus symptomatic (SH) hypersomnia. residual ESS was measured near the end of pitolisant treatment for those patients who switched to a different treatment. If the Patients with hypersomnia IH SH P treatment was stopped, the patients and clinicians provided the N 65 13 reasons for termination (adverse effects, lack of efficacy, tolerance, Age (years) 35 [27–44] 51 [34–64] 0.007 pregnancy, contraception or no need for treatment). Clinicians Women (%) 72.30 53.84 0.20 Age at symptom onset (years) 20 [15–32] 42 [35–47] 0.001 evaluated habituation to the drug (need to increase the dose with Age at diagnosis (years) 28 [21–37] 48 [38–58] 0.001 time or decrease benefit with time), abuse, and whether any loss of Time between symptoms and 5 [2–10] 5 [2–11] 0.85 efficacy had occurred; they also noted adverse effects that might diagnosis (years) possibly or most likely have been related to pitolisant. These Disease course (years) 8 [5–15] 12 [6–16] 0.73 effects were later classified into categories depending on the sys- Epworth Sleepiness Scale tem affected (nervous system, psychiatric, cardiovascular system, Score at diagnosis 17 [15–19] 18 [18–20] 0.08 or gastrointestinal system) or were classified as general adverse Score at baseline, before pitolisant 17 [15.5–18.5] 16 [13–18] 0.49 HLA DQB1⁄06:02 positive (%) 7.69 7.69 0.75 events. Data are median [interquartile range] unless otherwise indicated. 2.4. Night-time and daytime sleep measures (n = 1) or fibromyalgia (n = 1). Some notable events preceded the onset of IH in a few cases, including Guillain–Barré syndrome in We collected the baseline sleep measures from the clinical files. one patient (n = 1; IH persisted eight years after the end of the syn- All of the patients had been monitored in the Sleep Disorder Unit of drome), vaccination against the human papillomavirus (GardasilÒ; the Pitié-Salpétrière Hospital. Polysomnographic recordings con- Sanofi-Pasteur-MSD, Lyon, France) two weeks before sleepiness sisted of an electroencephalograph, electro-oculograms, levator onset (n = 1), and surgery (n = 2, one for thyroidectomy and one for menti and bilateral tibialis anterior surface electromyographs, suprarenalectomy). Comorbidities included depression secondary nasal pressure trough cannulae, respiratory efforts using thoracic to hypersomnia (n = 16, including 15 with IH and one with SH; nota- and abdominal belts, position, tracheal sounds, pulse rate, and bly depressive symptoms occurred several years after sleepiness transcutaneous oximetry. Sleep stages, periodic leg movements, onset), hypertension (n = 6), treated hypothyroidism (n = 5), ade- and respiratory events were scored visually by neurologists based quately treated sleep apnea syndrome (n = 5, including four with on standard criteria. The MSLT was performed in all patients at IH and one with SH), obsessive–compulsive disorder (OCD; n = 5), 08:00, 10:00, 12:00, 14:00, and 16:00. These recordings were ter- Crohn’s disease (n = 1), and impulsivity (n = 1). minated after 20 min if no sleep had occurred or after 15 min of Patients with IH were of a younger age (at symptom onset, time sleep time following the first epoch of sleep. In addition, 61/78 of diagnosis and time of study) compared with patients with SH; (78.2%) patients underwent long-term (24 h) sleep monitoring on there were no other demographic or clinical differences compared the second day. The respiratory sensors were removed during the with patients with SH (Table 1). In the IH group (Table 2), the 49 MSLT and the 24 h sleep monitoring. The class II human leukocyte patients with long sleep time IH were younger and more fre- antigen (HLA) genotype DQB1⁄0602 was also obtained for all quently female than the 16 patients without long sleep time, but patients. they all had similar disease course and ESS. The characteristics of nocturnal sleep, MSLT, and long-term 2.5. Statistical analyses sleep monitoring did not differ between patients with IH and SH, except for a higher mean apnea–hypopnea index and longer total Descriptive statistics used the median and interquartile ranges sleep period in the SH versus IH group (Supplementary Table A). for quantitative variables, and numbers and percentages were used Patients with IH and long sleep time had longer sleep onset latency for qualitative variables. Group comparisons were tested using during nocturnal sleep, decreased apnea/hypopnea index, longer 2 unpaired Wilcoxon’s test for quantitative variables and v -tests, mean daytime sleep latency at the MSLT, and longer night-time and Fisher’s exact tests for qualitative variables. To determine duration and total sleep duration during 24 h monitoring com- whether the changes in ESS were significant (different from no pared to IH patients without long sleep time (Supplementary change or zero) within each group, a paired Wilcoxon’s test was Table B). performed. All tests were two-sided, and P < 0.05 was considered to be significant. Computations were performed using SAS V9 sta- 3.2. Dosage, combination and benefits of pitolisant tistical software program. The maximum daily dosage of pitolisant ranged from 5 to 3. Results 50 mg/day (once per day in the morning) over the course of five days to 37 months (Table 3). The drug was taken without any other 3.1. Clinical characteristics of the patient population stimulants in 67 (85.9%) patients and was combined with modafinil, methylphenidate and sodium oxybate in other cases. It was On the ATU list, 78 patients with a non-narcoleptic central hyper- also combined with an antidepressant in 24.3% of the patients. somnia received pitolisant (Table 1). Sixty-five patients had IH, The benefit of pitolisant on sleepiness, measured as the change including 49 with and 16 without long sleep time. In addition, 13 immediately before and at the time of the higher dosage of pitoli- patients had SH (with, n = 6, and without, n = 7, long sleep time) that sant, was À1.9 ± 2.6, a value that was significantly different from 0 could reasonably be attributed to stroke (n = 6, including one bilat- (no change), as shown by a paired t-test (P < 0.0001). There was no eral paramedian thalamic stroke, two frontal posterior strokes, one difference between the IH and the SH groups or between the IH thrombophlebitis, one cavernoma, and one ischemic caudate lesion), patients with and without a long sleep time (Table 4). The paired neurolupus (n = 1), neurosarcoidosis (n = 1), a brain tumor (n =1,an t-test indicated a significant improvement of the ESS when com- oligodendroglioma of the optic chiasma), a brain traumatic injury pared to 0 (no change) within each IH disease group (P < 0.002 (n = 1), a neuromyotonia with epilepsy suggestive of Isaac syndrome for IH without long sleep time, P < 0.0001 for IH with long sleep but with negative antibodies (n = 1), chronic fatigue syndrome time), but not for SH (P = 0.16). If responders were defined as 684 S. Leu-Semenescu et al. / Sleep Medicine 15 (2014) 681–687

Table 2 Demographic and clinical characteristics of the patients with idiopathic hypersomnia (IH) with versus without long sleep time.

Patients with IH With long sleep time Without long sleep time P N 49 16 Age (years) 32 [27–42] 48 [39–61] 0.001 Women (%) 81.63 43.75 0.005 Age at symptom onset (years) 19 [15–32] 34 [20–46] 0.02 Age at diagnosis (years) 28 [21–37] 45 [33–54] 0.003 Time between symptoms and diagnosis (years) 5 [2–10] 2 [1–5] 0.17 Disease course (years) 10 [16–18] 6 [4–14] 0.13 Epworth Sleepiness Scale Score at diagnosis 18 [15–19] 16 [15–18] 0.31 Score at baseline, before pitolisant 17 [14–18] 17 [16–20.5] 0.23 HLADQB1⁄06:02 positive (%) 10.2 0 0.43

Data are median [interquartile range] unless otherwise indicated.

Table 3 Dosage, combinations of drugs and beneﬁt of pitolisant in patients with idiopathic (IH) versus symptomatic (SH) hypersomnia.

Patients Idiopathic hypersomnia Symptomatic hypersomnia P N 65 13 Stimulants tried before pitolisant, n 2 [2–3] 2 [2–4] 0.29 (range) (1–4) (1–5) Maximum daily dosage (mg) 40 [25–40] 40 [40–40] 0.98 (range) (5–50) (10–40) Time on pitolisant, months 6 [2–13.5] 3 [2–13] 0.98 (range) (0.2–37) (2–33) Combination of pitolisant with other drugs, % (n) Antidepressants 24.6 (16) 7.7 (3) 0.45 Sodium oxybate 1.5 (1) 0 0.37 Modafinil 9.2 (6) 0 0.57 Methylphenidate 6.1 (4) 0 0.82 Epworth Sleepiness Scale Score at baseline 17 [15.5–18.5] 16 [13–18] 0.49 Score with pitolisant 14 [12–17] 14 [13–17] 0.34 Change in score, n with complete data À1.5 [À4to0]n =56 0[À4to0]n = 13 0.59 Responders, % (n)a 35.4 (23/65) 38.5 (5/13) 0.83 Treatment stopped, % (n) 63.0 (44) 76.9 (12) 0.34 Reasons for stopping pitolisant, % (n) Lack of efficacy 52.3 (23) 58.3 (7) 1 Adverse effects 18.2 (8) 25 (3) 0.90 Both 22.7 (10) 8.33 (1) 0.48 Loss of efficacy 4.5 (2) 8.33 (1) 1 Other (pregnancy, relocation) 2.3 (1) 0 0.48 Treatments given after pitolisant was stopped, % (n) Modafinil 6.8 (3) 0 0.84 Methylphenidate 29.5 (13) 8.3 (1) 0.02 Sodium oxybate 15.9 (7) 16.7 (2) 0.70 Mazindol 9.1 (4) 25 (3) 0.32 D-Amphetamine 15.9 (7) 16.7 (2) 0.70 No other treatment 22.7 (10) 33.3 (4) 0.71

Data are the median [interquartile range] unless otherwise indicated. a Responders are defined as patients with an Epworth Sleepiness Scale fall of P3, divided by the total of patients (intention to treat analysis). patients presenting a reduction in ESS of P3, 28/78 (36%) of 3.3. Safety profile of pitolisant patients were responders, including 37% of patients with IH with long sleep time, 31% of patients with IH without long sleep time, The side-effects of pitolisant are indicated in Fig. 1 and included and 38.5% of patients with symptomatic hypersomnia. The benefit symptoms referred to the gastrointestinal tract and the central of pitolisant on ESS was lower (À1 ± 2.2) in the patients with con- nervous system. Epigastralgia and abdominal pain were reported comitant depression or personality disorder (OCD, impulsivity, by 15.4% of patients, increased appetite was reported by 14.1%, psychopathy, attention disorder with hyperactivity) than in the and weight gain was reported by 14.1%. The weight gain was pro- other patients (À2.3 ± 2.7, P = 0.04). As many as 56 (72%) patients gressive over time and ranged from 2 to 15 kg after two years of stopped pitolisant, mostly because of a lack of efficacy and more treatment. All patients with weight gain on pitolisant had been rarely because of side-effects or other reasons. Some of these previously treated by methylphenidate, mazindol or D-amphet- patients were switched back to modafinil (n = 3), methylphenidate amine, and had reported anorexia under these treatments. How- (n = 8), sodium oxybate (n = 1), mazindol (n =4)orD-amphetamine ever, the weight gain exceeded the baseline weight before (n = 1); some were switched to no drugs (n = 14); and some were starting any anorectic treatment. Headaches were reported by switched to a drug that they had not taken before, including meth- 12.8% of the patients, insomnia was reported by 11.5%, and anxiety ylphenidate (n = 6), sodium oxybate (n = 8), mazindol (n = 3), and was reported by 9%. Two patients had emerging depressive idea- D-amphetamine (n = 8). tion under pitolisant: one with suicidal ideation after only one S. Leu-Semenescu et al. / Sleep Medicine 15 (2014) 681–687 685

Table 4 Dosage, combinations of drugs and beneﬁt of pitolisant in patients with idiopathic hypersomnia with versus without long sleep time.

Patients with idiopathic hypersomnia With long sleep time Without long sleep time P N 49 16 Stimulants tried before pitolisant, n 3 [2–3] 2 [2–3] 0.50 (range) (1–4) (1–4) Maximum daily dosage, mg 40 [30–40] 40 [25–40] 0.99 (range) (15–40) (5–50) Time on pitolisant, months 4 [2–14] 7 [2–12.5] 0.85 (range) (0.2–37) (1–34) Combined with other drugs, % (n) Antidepressants 22.4 (11) 31.2 (5) 0.70 Sodium oxybate 2.0 (1) 0 (0) 0.55 Modafinil 10.2 (5) 6.25 (1) 0.98 Methylphenidate 6.1 (3) 6.2 (1) 0.58 Epworth Sleepiness Scale Score at baseline 17 [14–18] 17 [16–20.5] 0.23 Score with pitolisant 14 [12–17] 16 [13–17] 0.34 Change in score, no. with complete data À1[À4to0] À1.5 [À4to0] 0.59 n =43 n =13 Responders, % (n)a 37 (18/49) 31 (5/16) 0.69 Treatment stopped, % (n) 67.3 (33) 68.7 (11) 0.84 Reasons for stopping pitolisant Lack of efficacy, % (n) 48.5 (16) 63.6 (7) 0.60 Adverse effects, % (n) 21.2 (7) 9.1 (1) 0.65 Both, % (n) 24.2 (8) 18.2 (2) 1 Loss of efficacy, % (n) 3.0 (1) 9.1 (1) 1 Other (pregnancy, relocation), % (n) 3.0 (1) 0 0.56 Treatments given after pitolisant was stopped, % (n) Modafinil 9.1 (3) 0 0.73 Methylphenidate 30.3 (10) 27.3 (3) 0.85 Sodium oxybate 18.2 (6) 9.1 (1) 0.81 Mazindol 6.1 (2) 18.2 (2) 0.54 D-Amphetamine 15.1 (5) 18.2 (2) 0.81 No other treatment 21.2 (7) 27.3 (3) 1

Data are median [interquartile range] unless otherwise indicated. a Responders are deﬁned as patients with an Epworth Sleepiness Scale fall of P3, divided by the total of patients (intention-to-treat analysis).

Fig. 1. Side-effects of pitolisant. week with 5 mg/day pitolisant and another with abrupt onset of symptom with modafinil). The persistent genital arousal syndrome sadness, low self-esteem, and pessimism after 18 months with is defined as spontaneous, persistent, uncontrollable genital arou- 40 mg/day pitolisant. In both cases, these symptoms stopped one sal unrelated to any sexual desire [17], with or without orgasm or to two weeks after pitolisant had been withdrawn. In contrast, genital engorgement (female mirror of priapism). The side-effects, there was no worsening of depression in the patients who were which were severe enough to stop the drug despite a clear benefit already depressed at the time of the study. Some rare unexpected on sleepiness, included major anxiety (n = 3), severe depression adverse events were reported, such as gynecomastia in a male (n = 2) and nightmares (n = 2). Notably, the five patients with patient, hot flashes, the resumption of menstrual periods in one comorbid OCD had no aggravation or improvement of those symp- postmenopausal woman, and persistent genital arousal in two toms under pitolisant. No cardiovascular effects (hypertension, female patients (one patient had already experienced this specific coronary heart disease, cardiac arrhythmia, and tachycardia) were 686 S. Leu-Semenescu et al. / Sleep Medicine 15 (2014) 681–687 observed. No habituation, addiction or abuse of pitolisant was from any stimulant with a long half-life. Pitolisant has a half-life of observed. Cessation of the drug in patients (72%) was not associ- 11 h, similar to that of modafinil, which can also cause insomnia. ated with any withdrawal syndrome. In contrast, irritability was extremely rare with pitolisant, whereas it was observed with previous drugs in the same patients. The absence of irritability was the most appreciable effect of pitolisant 4. Discussion in patients who previously felt aggressive while on modafinil and could constitute a reason for switching from modafinil to pitolisant. In this open, large series of patients with idiopathic and symp- Similarly, there were no increases in obsessive and compulsive tomatic central hypersomnia and excessive daytime sleepiness behaviors in patients with OCD, which is in keeping with the recent refractory to usual stimulants, pitolisant had a favorable benefit/ observation of a patient with Tourette syndrome and narcolepsy risk ratio in 23–38% of patients. The ESS was decreased from a who underwent a clear effect of pitolisant on sleepiness and no mean of À1.5 points in IH but not in SH and the benefit persisted worsening of tics; this is in sharp contrast with modafinil and methy- over the long term. The main side-effects were gastrointestinal phenidate [23]. There were also no problems of addiction or with- problems, weight gain, headache, insomnia and anxiety. drawal, which is in keeping with preclinical studies [24]. These In a previous series of 22 patients with narcolepsy, the benefit features (absence of irritability, compulsions, tics, addiction and of pitolisant was strong, with a mean reduction of ESS of 4.9 points withdrawal syndromes) differentiate the stimulant profile of this in an open single-blind short trial [14]. However, these patients H3R inverse agonist from that of other classes of wake-promoting were not resistant to previous stimulants, which could explain drugs such as amphetamines. their greater response rate to pitolisant. Similarly, an important This study carries the same limitations that are inherent to any (as high as À6 point) reduction in ESS has been recorded in open field study. It is an observational, retrospective, non-industry- trials of pitolisant in patients with residual sleepiness while being sponsored, open-labeled study including patients followed from a adequately treated for sleep apnea and in patients with sleepiness reference center for rare hypersomnias. However, it provides a associated with Parkinson’s disease [18]. Another new H3 receptor ‘real-life’ glance at these patients with no exclusions, e.g. due to antagonist, MK-0249, improved ESS (but not the mean latency at concomitant OCD or depression, and with no forbidden drug com- the maintenance of wakefulness test) by À3.99 points in patients bination (as is the case for Phase II and III trials). It is focused on the with sleep apnea syndrome and residual sleepiness despite the extreme spectrum of these rare diseases, i.e. the most resistant and adequate use of positive airway pressure [19]. In comparison, the complex patients. However, if the most resistant patients report a benefit of pitolisant in our patients with hypersomnia was mild; benefit (even mild) from pitolisant along with a good safety profile, however, the benefit on ESS was significant in IH (but not in SH). this supports formally testing the drug in patients with IH. From a If one defined responders as patients presenting an ESS fall of more general point of view, these results indicate that the hista- P3, 31–38.5% of patients were responders to pitolisant, with no mine neurons can be stimulated in patients with IH and SH. difference between idiopathic and symptomatic hypersomnia, and hypersomnia with versus without long sleep time. Note that Funding sources these patients suffered severe sleepiness refractory to major stimulants, and were not selected for absence of personality disorder or None. depression as in Phase II or III studies. They reflected more the real- life patients. However, one-third of the patients continued the Conflict of interest treatment, which suggests that there remains room for improvement in the use of histamine stimulation in ‘difficult’ patients Dr. Arnulf received honoraria and board for a speaking engage- resistant to modafinil, amphetamine, and amphetamine-like drugs. ment from UCB Pharma and Jazz pharmaceuticals. Dr. Leu-Sem- It remains necessary to evaluate the stimulating properties of enescu received honoraria for a speaking engagement from UCB pitolisant as a first- or second-line treatment for IH and SH and Pharma. In the present report, the pharmaceutical company Bio- to compare these properties with those of placebo and modafinil. projet Ltd, which manufactured pitolisant, had no role, as the drug The safety profile of the drug was good and parallels the side- was purchased by hospital pharmacists and paid for by the effects found when pooling all trials performed with this drug [18]. national security insurance under the individual authorization for The main side-effects were gastrointestinal problems (epigastralgia, use process organized by the French Drug Agency (Agence Natio- nausea, vomiting, abdominal pain, diarrhea, and constipation), but nale de Sécurité du Médicament et des produits de santé). Dr. they were not severe enough to stop the drug. Nausea was observed Arnulf and Dr. Leu-Semenescu have been investigators in several in four out of 22 patients in the narcolepsy trial [14]. Notably, hista- trials of pitolisant conducted by Bioprojet. mine is known to be a regulator of gastrointestinal functions such as The ICMJE Uniform Disclosure Form for Potential Conflicts of gastric acid production, intestinal motility, and mucosal ion secre- Interest associated with this article can be viewed by clicking on tion. Histamine H3 receptors have been identified in the rat gastric the following link: http://dx.doi.org/10.1016/j.sleep.2014.01.021. fundus at the surface of the enterochromaffin-like cells [20], and their blockade may stimulate gastric acid secretion and explain the frequent epigastralgia observed in our patients. However, H3 receptors are not expressed in other intestinal sites in humans, making a Appendix A. Supplementary data direct link between pitolisant and constipation or diarrhea more difficult to establish [21]. The observed weight gain is a surprising Supplementary data associated with this article can be found, in effect, as the other stimulants (methylphenidate, amphetamine the online version, at http://dx.doi.org/10.1016/j.sleep.2014.01.021. and, to a lesser degree, modafinil) are associated with decreased appetite and weight loss. Brain histamine plays a central and differ- References ential role in motivation by reinforcing the appetitive phase (including food anticipation in rats) while reducing the drive to consume in [1] American Academy of Sleep Medicine. The international classification of sleep motivated behaviors (review in [22]). It will be interesting to study disorders – revised. In: Hauri P, editor. Chicago, IL: American Academy of Sleep Medicine; 2005. more precisely whether patients using pitolisant report increases [2] Vernet C, Arnulf I. Idiopathic hypersomnia with and without long sleep time: a in appetite or food consumption. Insomnia is an expected side-effect controlled series of 75 patients. Sleep 2009;32:752–9. S. Leu-Semenescu et al. / Sleep Medicine 15 (2014) 681–687 687

[3] Vernet C, Leu-Semenescu S, Buzare M, et al. Subjective symptoms in idiopathic [14] Lin JS, Dauvilliers Y, Arnulf I, et al. An inverse agonist of the histamine H(3) hypersomnia: beyond excessive sleepiness. J Sleep Res 2010;19:525–34. receptor improves wakefulness in narcolepsy: studies in orexin–/– mice and [4] Rye DB, Bliwise DL, Parker K, et al. Modulation of vigilance in the primary patients. Neurobiol Dis 2008;30:74–83.

hypersomnias by endogenous enhancement of GABAA receptors. Sci Transl [15] Dauvilliers Y, Bassetti C, Lammers GJ, et al. Pitolisant versus placebo or Med 2012;4. 161ra51. modafinil in patients with narcolepsy: a double-blind, randomised trial. Lancet [5] Kanbayashi T, Kodama T, Kondo H, et al. CSF histamine contents in narcolepsy, Neurol 2013;12:1068–75. idiopathic hypersomnia and obstructive sleep apnea syndrome. Sleep [16] Inocente C, Arnulf I, Bastuji H, et al. Pitolisant, an inverse agonist of the 2009;32:181–7. histamine H3 receptor: an alternative stimulant for narcolepsy–cataplexy in [6] Bassetti CL, Baumann CR, Dauvilliers Y, et al. Cerebrospinal fluid histamine teenagers with refractory sleepiness. Clin Neuropharmacol 2012;35:55–60. levels are decreased in patients with narcolepsy and excessive daytime [17] American Psychiatric Association. Diagnostic and statistical manual of mental sleepiness of other origin. J Sleep Res 2010;19:620–3. disorders. 5th ed. Washington, DC: APA; 2013. [7] Dauvilliers Y, Delallee N, Jaussent I, et al. Normal cerebrospinal fluid histamine [18] Schwartz JC. The histamine H3 receptor: from discovery to clinical trials with and tele-methylhistamine levels in hypersomnia conditions. Sleep pitolisant. Br J Pharmacol 2011;163:713–21. 2012;35:1359–66. [19] Herring W, Liu K, Hutzelmann J, et al. Alerting and psychomotor performance [8] Morgenthaler T, Kapur V, Brown T, et al. Practice parameters for the treatment effects of the histamine H3 inverse agonist MK-0249 in obstructive sleep of narcolepsy and other hypersomnias of central origin. Sleep apnea patients on CPAP with excessive daytime sleepiness: a randomized 2007;30:1705–11. adaptive crossover study. Sleep Med 2013;14:955–63. [9] Lavault S, Dauvilliers Y, Drouot X, et al. Benefit and risk of modafinil in [20] Chazot P, Shenton F, Schunack W, et al. Influence of (R)-alpha- idiopathic hypersomnia versus. narcolepsy with cataplexy. Sleep Med methylhistamine on the histamine H3 receptor in the rat gastrointestinal 2011;12:550–6. tract. Inflamm Res 2007;56(Suppl 1):S19–20. [10] Ali M, Auger R, Slocumb N, et al. Idiopathic hypersomnia: clinical features and [21] Sander LE, Lorentz A, Sellge G, et al. Selective expression of histamine response to treatment. J Clin Sleep Med 2009;5:562–8. receptors H1R, H2R, and H4R, but not H3R, in the human intestinal tract. Gut [11] Anderson K, Pilsworth S, Sharples L, et al. Idiopathic hypersomnia: a study of 2006;55:498–504. 77 cases. Sleep 2007;30:1274–81. [22] Torrealba F, Riveros ME, Contreras M, et al. Histamine and motivation. Front [12] Ozaki A, Inoue Y, Hayashida K, et al. Quality of life in patients with narcolepsy Syst Neurosci 2012;6:51. with cataplexy, narcolepsy without cataplexy, and idiopathic hypersomnia [23] Hartmann A, Worbe Y, Arnulf I. Increasing histamine neurotransmission in without long sleep time: comparison between patients on psychostimulants, Gilles de la Tourette syndrome. J Neurol 2011;259:375–6. drug-naive patients and the general Japanese population. Sleep Med [24] Uguen M, Perrin D, Belliard S, et al. Preclinical evaluation of the abuse 2012;13:200–6. potential of pitolisant, a histamine H3 receptor inverse agonist/antagonist [13] Nittur N, Konofal E, Dauvilliers Y, et al. Mazindol in narcolepsy and idiopathic compared with modafinil. Br J Pharmacol 2013;169:632–44. and symptomatic hypersomnia refractory to stimulants: a long-term chart review. Sleep Med 2013;14:30–6.