Orexin 2 receptor-selective agonist danavorexton (TAK-925) promotes wakefulness in non-human primates and healthy individuals

Summary The orexin 2 receptor-selective agonist danavorexton (TAK-925) has been shown to produce wake-promoting effects in wild-type mice, narcolepsy-model mice, and individuals with narcolepsy type 1 and type 2. Here, we report wake-promoting effects of danavorexton in non-human primates and healthy men during their sleep phase. Electroencephalogram analyses revealed that subcutaneous administration of danavorexton significantly increased wakefulness in common marmosets ( p < 0.05 at 0.1 mg kg (cid:1) 1 , and p < 0.001 at 1 mg kg (cid:1) 1 and 10 mg kg (cid:1) 1 ) and cynomolgus monkeys ( p ≤ 0.05 at 1 mg kg (cid:1) 1 and 3 mg kg (cid:1) 1 ). In a phase 1b crossover, randomized, double-blind, placebo-controlled and active-controlled study in sleep-deprived healthy participants (ClinicalTrials.gov identifier

agonist increases wakefulness in non-human primates and healthy individuals during their sleep phase.

K E Y W O R D S
excessive daytime sleepiness, hypocretins, novel medicine, sleep regulation, sleep/wake transition

| INTRODUCTION
Orexin neurons are integral to the regulation of a variety of physiological mechanisms, including arousal, energy homeostasis, endocrine functions and reward processes (Tsujino & Sakurai, 2009). Although orexin neurons are located predominantly in the lateral hypothalamic area (LHA), they have widespread projections throughout the central nervous system, from the cerebral cortex to the spinal cord (Inutsuka & Yamanaka, 2013;Nambu et al., 1999;Peyron et al., 1998;Saper et al., 2005). Within the LHA, orexin neurons produce two types of orexin neuropeptides, orexin A and orexin B (also called hypocretin-1 and hypocretin-2), which result from the proteolytic cleavage of the single precursor molecule prepro-orexin Sakurai et al., 1998). Both orexin neuropeptides bind to the G-protein-coupled receptors orexin 1 receptor (OX1R) and orexin 2 receptor (OX2R). OX1R has a greater affinity for orexin A than for orexin B, whereas OX2R has a similar affinity for both orexin A and orexin B Sakurai et al., 1998;Scammell & Winrow, 2011).
Orexin plays a critical role in the regulation of sleep-wake behaviour, with increased orexin levels in cerebrospinal fluid (CSF) observed during active phases (Salomon et al., 2003;Yoshida et al., 2001;Zeitzer et al., 2003). Animal studies have shown that orexin neurons fire during wakefulness, and that firing is substantially reduced during sleep (Lee et al., 2005). Moreover, a highly selective and severe loss of orexin neurons, resulting in low-toabsent levels of orexin peptides in the brain and CSF, underlies narcolepsy type 1 (NT1), which is characterized by primary symptoms of excessive daytime sleepiness (EDS) and cataplexy (Nishino et al., 2000;Peyron et al., 2000;Thannickal et al., 2000). NT1 is also associated with comorbidities such as obesity and psychiatric conditions, including attention-deficit/hyperactivity disorder, depression, anxiety and schizophrenia (Morse & Sanjeev, 2018;Tsujino & Sakurai, 2009). OX2R-knockout mice show clear narcolepsy-like phenotypes, such as fragmented sleep/wakefulness and cataplexylike episodes, whereas OX1R-knockout mice have minor or no sleep abnormalities and no cataplexy (Tsujino & Sakurai, 2009). As such, activation of the orexin receptors, especially OX2R, is of interest as a therapeutic strategy for NT1 (Irukayama-Tomobe et al., 2017).
Small-molecule OX2R-selective agonists, such as danavorexton , provide a promising therapeutic approach for NT1, as orexin peptides are not suitable for clinical use because of little or no blood-brain barrier penetration (Fujiki et al., 2003). In preclinical studies, we previously found that danavorexton increased wakefulness during the sleep phase in wild-type mice, but not in OX2R-knockout mice (Yukitake et al., 2019). Danavorexton also increased wakefulness, reduced cataplexy-like episodes and suppressed abnormal weight gain in orexin/ataxin-3-transgenic mice, a mouse model of narcolepsy (Evans et al., 2022;Ishikawa et al., 2022). In line with these preclinical observations, intravenous infusion of danavorexton improved wakefulness in individuals with NT1, as measured by the
The pathophysiology of other hypersomnia disorders that involve only partial or no loss of orexin neurons, such as narcolepsy type 2 (NT2; narcolepsy without cataplexy) and idiopathic hypersomnia, is not well understood, and thus no reliable animal models are available.
In our previous study, danavorexton showed early signs of efficacy in a small number of participants with NT2 (Evans et al., 2022). If danavorexton enhanced wakefulness in healthy individuals with a normal orexin system, it could potentially provide therapy for EDS in a broader range of conditions. Given their polyphasic sleep-wake structure, mice may not be an optimal model for humans, who have a monophasic sleep-wake structure. Hence, we sought to characterize further the efficacy of danavorexton under the conditions of a normal orexin system in non-human primates, followed by a directly translational clinical study in healthy individuals.
Here, we report the effects of danavorexton on wakefulness during the sleep phase of common marmosets (Callithrix jacchus) and cynomolgus monkeys (Macaca fascicularis), both of which have a predominantly monophasic sleep-wake pattern, with wakefulness accounting for > 90% of the light period  The studies with common marmosets were conducted at Aptuit LLC (Verona, Italy). Common marmosets were housed in pairs in standard cages maintained at 25 ± 1 C and 55 ± 10% humidity under a 12-hr light/dark cycle (light on at 06:00 hours), and were fed twice a day with ad libitum access to water. The studies with cynomolgus monkeys were conducted internally (Takeda Pharmaceutical Company Limited, Fujisawa, Japan). Cynomolgus monkeys implanted with radiotelemetry transmitters (TL10M3-D70-EEE, Data Science International Inc., Saint Paul, MN, USA) were purchased from HAMRI Co. Ltd (Ibaraki, Japan). They were housed individually in stainless-steel cages maintained at 24 ± 3 C and 55 ± 15% humidity under a 12-hr light/ dark cycle (light on at 07:00 hours), and fed once daily with ad libitum access to water. All non-human primates were housed and handled in strict accordance with the guidelines for good animal practice under the supervision of veterinarians, and received environmental enrichment with monitoring for evidence of disease and changes in attitude, appetite or behaviour that are suggestive of illness.
2.1.4 | Electroencephalogram surgical procedure and data analysis in the common marmoset All experiments with common marmosets (three male and three female) were conducted by Aptuit LLC (Verona, Italy). The effect of danavorexton was assessed using telemetric recording of cortical electroencephalogram (EEG) signals and electrooculogram (EOG) signals.
A multichannel telemetric transmitter (TL11M2-F40-EET, Data Sciences International Inc., New Brighton, MN, USA) was implanted intraperitoneally using standard surgical techniques in anaesthetized marmosets (Crofts et al., 2001). Drug treatments were assigned in a random, blind, crossover design. A minimal 3-day washout period was allowed between test sessions. The wake-promoting effects of danavorexton at 0.1 mg kg À1 , 1 mg kg À1 and 10 mg kg À1 s.c. were evaluated in common marmosets. Each 10-s period of EEG was analysed using SleepSign™ software (Kissei Comtec America, Irvine, CA, USA) to define wakefulness, slow-wave sleep (SWS I, light phase), slowwave sleep II (SWS II, deep sleep) or rapid eye movement (REM) sleep.

| EEG surgical procedure and data analysis in cynomolgus monkeys
All experiments with cynomolgus monkeys were conducted by Takeda Pharmaceutical Company Limited (Fujisawa, Japan). Cynomolgus monkeys (four males) were implanted with radio-telemetry transmitters (TL10M3-D70-EEE, Data Sciences International Inc., Saint Paul, MN, USA) under anaesthesia, as previously described . Cortical EEG, EOG and electromyogram (EMG) signals were all recorded using Dataquest ART software (Data Sciences International). Drug treatments were assigned to four cynomolgus monkeys in a pre-post design. To explore minimum effective dosage in cynomolgus monkeys, the wake-promoting effects of danavorexton at 0.3 mg kg À1 , 1 mg kg À1 and 3 mg kg À1 s.c. were compared with that by corresponding vehicle. The signals in 20-s epochs were semiautomatically scored as wakefulness, non-REM sleep or REM sleep by a sleep scoring system (SleepSign™, Kissei Comtec, Nagano, Japan). This preliminary scoring was visually inspected and corrected if necessary.

| Diurnal and nocturnal experiment schedules in marmosets and monkeys
In common marmosets, vehicle or danavorexton (0.1, 1 and 10 mg kg À1 , s.c.) was administered at the start of the dark/sleep cycle at zeitgeber time 12 (ZT12), and EEG/EOG signals were recorded for 5 hr after administration ( Figure S1a in Appendix S1). In cynomolgus monkeys, vehicle or danavorexton (0.3, 1 and 3 mg kg À1 , s.c.) was administered at ZT12, and EEG, EOG and EMG signals were recorded for 8 hr after administration ( Figure S1b in Appendix S1).

| Statistical analysis
Statistical analysis was performed using EXSUS (CAC EXICARE, Tokyo, Japan). Wakefulness and sleep-phase data from common marmosets were analysed with one-way analysis of variance followed by two-tailed Dunnett's test. Wakefulness data from cynomolgus monkeys were analysed to detect pairwise differences between groups using a two-tailed paired t-test. A p-value of ≤ 0.05 was considered significant.

| Study populations
Healthy men (18-40 years old) were enrolled at a single study centre in the USA in 2018. Demographics and baseline characteristics were recorded at screening, including smoking status and caffeine consumption. Participants were excluded if they: had used tobacco-or nicotine-containing products in the 6 months before the start of the study; had excessive sleepiness defined by a selfreported Epworth Sleepiness Scale score at screening of greater than 10; had irregular work hours or routine night-shift work in the 1 month before randomization; were currently experiencing, or had a history of, any known/suspected sleep disorder or any disorder associated with EDS; had a caffeine consumption of more than 400 mg day À1 for 2 weeks before screening; or had other sleep disorders. In addition, participants were required to have a regular sleep-wake schedule that involved 6.5-8.0 hr of sleep per night, regularly falling asleep between 21:30 hours and 00:00 hours, and not oversleeping by more than 3 hr.

| Study designs and schedule of procedures
In this placebo-controlled, randomized, single-dose, four-period crossover study, eligible participants were equally randomized to four treatment sequence groups that defined the order of administration for danavorexton 44 mg, danavorexton 112 mg, modafinil 300 mg (used to demonstrate assay sensitivity) and placebo. On day 1 of treatment period 1, participants were assigned a randomization number in ascending numerical order at the clinical site, which linked to a randomization schedule generated by the sponsor before the study. Investigators and participants were blinded to treatment assignment.
Sample size justification was based on a similar study with a histamine subtype-3 receptor inverse agonist and modafinil 200 mg in healthy sleep-deprived males (Iannone et al., 2010). Sixteen completers were required to provide approximately 90% power for detecting an increase in sleep latency in the MWT of 9 min with danavorexton versus placebo, based on a two-sided test with 5% false positive rate. Approximately 20 participants were planned to be enrolled to account for dropouts.
Danavorexton 44 mg, danavorexton 112 mg and placebo (normal saline) were delivered as 9-hr overnight intravenous (i.v.) infusions starting at approximately 22:45 hours; modafinil 300 mg and matching placebo were given with 240 ml water for oral administration 1 hr before the start of the i.v. infusion (at approximately 21:45 hours). To maintain blindness, participants assigned to danavorexton additionally received oral placebo, participants assigned to modafinil additionally received i.v. placebo, and participants assigned to placebo received both i.v. and oral placebo. Participants were required to stay awake between assessments, and were allowed to sleep for approximately 6 hr after completion of the last study assessment. The interval between each treatment period was at least 7 days. Prior to randomization, 5 days of actigraphy were obtained to confirm regular sleep-wake habits. On the day prior to first dosing, participants underwent nocturnal polysomnography to ensure that no sleep disorders were present.
Doses of danavorexton evaluated in the current study were selected based on pharmacokinetic, safety and tolerability data from the first-in-human study in Japanese patients with NT1 and healthy participants (Evans et al., 2022), and efficacy data from non-clinical studies in cynomolgus monkeys shown in Figure 2.
Modafinil 300 mg was selected as a dose that would be efficacious in healthy adults, and that falls within the doses approved in the label for treatment of excessive sleepiness (up to 400 mg) based on available safety data ("PROVIGIL Prescribing Information", 2015).

| Secondary outcomes
Subjective sleepiness was assessed using the KSS, a 10-point selfrating scale, which was performed before dosing, and at approxi-

| Pharmacokinetic evaluation
Serial 3-ml blood samples were collected before dosing, and at 1, 2, 4, 6 and 9 hr after infusion start, then at 0.17, 0.50, 1, 2, 4, 6 and 9 hr after infusion end, and at discharge of each period. Plasma concentrations of danavorexton were measured by HPLC-MS/MS. The bioanalytical method information is summarized in supplement Table S1 in Appendix S1.
The plasma pharmacokinetic parameters of danavorexton were estimated from the concentration-time profiles for all evaluable participants using standard non-compartmental methods (Phoenix Win-Nonlin version 8.1, Certara USA, Princeton, NJ, USA).  F I G U R E 1 Outcomes from assessments of common marmosets. (a) Danavorexton plasma concentration (mean ± SD) in common marmosets following administration s.c. of danavorexton 0.1 mg kg À1 (n = 4) and 1 mg kg À1 (n = 3). (b) Time course of danavorexton effects on wakefulness measured by EEG/EOG monitoring (mean ± SD; n = 4 for 10 mg kg À1 , n = 6 for other groups). Effects of danavorexton (mean ± SD; n = 4 for 10 mg kg À1 , n = 6 for other groups) on (c)

| Study approval
The animal studies reported here were approved by the Institutional Ani-
To assess arousal effects, danavorexton at 0.1, 1 and 10 mg kg À1 , s.c. was administered to common marmosets at ZT12 (Figure S1a in Appendix S1). During the sleep phase, danavorexton significantly and dose-dependently enhanced wakefulness in common marmosets (p < 0.05 at 0.1 mg kg À1 , and p < 0.001 at 1 mg kg À1 and 10 mg kg À1 ; Figure 1b,c). After administration of doses of 1 mg kg À1 and 10 mg kg À1 , wakefulness was fully maintained for 3 hr and > 5 hr, respectively ( Figure 1b). In addition, danavorexton decreased SWS I, SWS II and REM sleep time (Figure 1d-f).

| Effects of danavorexton in cynomolgus monkeys
To measure plasma exposure, danavorexton was administered s.c. to cynomolgus monkeys at ZT12 ( Figure S1b in Appendix S1). Following doses of 1 mg kg À1 and 3 mg kg À1 , mean (SD) C max values were 73.5 (29.8) ng ml À1 and 102.1 (45.5) ng ml À1 , respectively 3) ng hr À1 ml À1 for danavorexton 1 mg kg À1 and 799.6 (375.0) ng hr À1 ml À1 for danavorexton 3 mg kg À1 (Figure 2a). The lack of PK linearity of danavorexton in cynomolgus monkeys may be explained by its poor physicochemical properties, especially low solubility.
To assess arousal effects, danavorexton at 0.3, 1 and 3 mg kg À1 , s.c. was administered to cynomolgus monkeys at ZT12 (Figure S1b in Appendix S1). During their sleep phase, danavorexton significantly and dose-dependently increased wakefulness in cynomolgus monkeys (p ≤ 0.05 at 1 mg kg À1 and 3 mg kg À1 ; Figure 2b,c). Danavorexton at 3 mg kg À1 showed > 75% arousal up to 7 hr after administration, which was associated with plasma concentration exceeding 32.5 ng ml À1 . These data were used to predict human systemic efficacious concentrations and guide dose selection in the clinic.

| Participant demographics and disposition
In total, 20 participants were enrolled and randomized, with five participants in each of the four treatment sequence groups (Figure 3). All participants were men, and 19 (95.0%) were white. Participants had a mean (SD) age of 26.9 (3.28) years, and mean (SD) height, weight and body mass index measurements of 177.3 (6.46) cm, 82.2 (9.58) kg and 26.1 (2.36) kg m À2 , respectively. The study was completed by 18 participants, with two participants withdrawing for personal reasons.

| Maintenance of Wakefulness Test
The mean (SD) sleep latency on the MWT for modafinil 300 mg, which was used to demonstrate assay sensitivity, was 29.2 (9.7) min;

| Pharmacokinetics
During a single 9-hr i.v. infusion, systemic exposure to danavorexton increased approximately dose-proportionally between the 44-mg and 112-mg dose groups in this healthy adult population. The mean plasma concentrations of danavorexton increased gradually, reaching more than 80% of the observed maximum concentration by 4 hr after the start of infusion (Figure 4c). The mean (SD) C max of danavorexton 44 mg was 106.4 (11.7) ng ml À1 , with a mean (SD) AUC from time 0 to infinity (AUC ∞ ) of 963.7 (134.7) ng hr À1 ml À1 , and the mean (SD) C max of danavorexton 112 mg was 268.4 (40.1) ng ml À1 , with a mean AUC ∞ of 2368.7 (185.0) ng hr À1 ml À1 (Table 2). After the end

| Treatment-emergent adverse events
There were no deaths or serious TEAEs, and no participants discontinued because of a TEAE ( commonly reported TEAEs with danavorexton 112 mg were pollakiuria (n = 3), and cough, dizziness and hypervigilance (all n = 2).
The most common TEAE with danavorexton 44 mg was infusionrelated reaction (n = 2; Table 4). Overall, 15 participants (75.0%) experienced a TEAE that was considered to be related to the study drug (Table 3).

| Other safety findings
No clinically significant changes from baseline in clinical laboratory evaluations, vital signs (including blood pressure), physical examinations or electrocardiograms were reported for any participant during the study.

| DISCUSSION
The present study is the first to demonstrate that an OX2R agonist can enhance wakefulness during the sleep phase in two non-human primate species and a population of healthy adults. These findings suggest that activation of OX2Rs represents a promising therapeutic approach for the treatment of EDS that is not associated with orexin deficiency.
In preclinical studies in non-human primates, danavorexton consistently and dose-dependently increased wakefulness during the night. These findings translated directly to humans, in whom both doses of danavorexton assessed were statistically significantly T A B L E 1 Sleep latency from the MWT after infusion start, and difference between average KSS score before and after infusion start  (Mileykovskiy et al., 2005;Ventzke et al., 2019;Zeitzer et al., 2003) and that sleep pressure increases over the course of sleep deprivation.
Although i.v. drug administration itself could have resulted in increased wakefulness, this risk was mitigated by the doublecrossover study design. No conclusions can be drawn about sustained efficacy following repeated chronic dosing, because only a single dose of danavorexton (or other randomized drug) was administered per period. Only healthy men were enrolled in the study to minimize any interindividual and intraindividual variability related to between-sex effects of the sleep cycle, as well as potential effects of the menstrual cycle on sleep. Therefore, generalizability of the results to other groups may be limited. The design of this study and its inclusion/ exclusion criteria are consistent with other studies of sleep deprivation.
In conclusion, these findings suggest that activation of OX2Rs during the sleep phase in non-human primates and healthy participants with a normal orexin system can promote wakefulness. As such, orexin agonists may be beneficial in disorders of EDS that have no reduction in orexin. Further studies are required to better evaluate the effects of danavorexton in these specific conditions.

AUTHOR CONTRIBUTIONS
All authors contributed to the study concept and interpretation of results, and participated in writing the manuscript. For the preclinical study, MN and HK conceived the study, and MN, TI, HY and MS con-