The most well-characterized function of the OX system is the regulation of sleep-wake transitions (14). Preclinical studies showed that experimental activation of OX neurons promotes and stabilizes wakefulness while their inhibition promotes non-rapid eye movement (non-REM) sleep (15–18) and that progressive loss of OX neurons in mouse models leads to fragmented wake (19). OX peptides exert their wake-stabilizing effects by interacting with the brain arousal network, which includes histaminergic, noradrenergic, and cholinergic neurons (14). Additionally, OX neurons support the wake state through promotion of muscle and motor tone (14, 20). Interestingly, the SAMP8 mouse model of accelerated aging show less non-REM and REM sleep and higher locomotor activity during the inactive phase of the light-dark cycle, which co-occur with higher cerebrospinal fluid (CSF) OX levels (21).

The orexin system contributes to the hypothalamic control of movement: brain injection of OX-A induces hyperlocomotion, stereotypies and grooming behavior in rats (22, 23); conversely, relative to wild-type mice, orexin-ablated mice show reduced exploratory activity under normal feeding conditions (24, 25). In addition, recent studies have shown that orexin cells are active during motion initiation (26). Their activation stimulates GAD65-expressing neurons in the lateral hypothalamus, which are then responsible for promoting normal locomotion and whose hyperactivity leads to hyperlocomotion (27).

Reactive (defensive) aggression in rodents is largely controlled by hypothalamic circuitry (28). Activity of the ventrolateral portion of the ventromedial hypothalamus (VMHvl) is particularly associated with an aggression-related arousal state and execution of aggressive actions (29). OX neurons receive projections from the VMH via the dorsomedial hypothalamus (DMH) (30) and recruit GAD2-expressing neurons in the lateral habenula, whose activation promotes reactive aggressive behavior; this suggests that OX cells integrate information regarding aggression-related arousal (OX neurons become activated following the expression of aggressive behavior, see Figure 1A.1, A.2) and influence the activity of areas involved in the expression of reactive aggression. Drugs that antagonize both OX1/2Rs, or selectively OX2Rs, reduce reactive aggression in rodents (31–33) (Figure 1B). As mentioned above, SAMP8 mice have higher inactive phase CSF OX levels (21), and this strain shows higher reactive aggression relative to control animals (34, 35).

The physiological and behavioral response to stressful environmental events allows animals to cope with situations relevant for survival. The hypothalamus is an integral part of the neurocircuitry coordinating the body's response to stress, and the contribution of the OX system has been largely investigated. Brain injection of OX peptides increases behavioral (e.g., grooming) and physiological (e.g., plasma levels of corticosterone) correlates of stress exposure (36, 37) and make mice agitated and hyper-responsive to sensory stimulation (38). Acute stressors recruit OX neurons (39, 40), and DORAs reduce stress-induced behavioral changes (41). In addition, brain infusion of OX peptides produces anxiogenic-like effects (42, 43), while OXR antagonists attenuate the expression of anxiety-like behaviors (44).

Rodent studies indicate that the orexin system plays a role in the regulation of reward processing (7, 45, 46). OX neurons project to and modulate the activity of brain areas involved in reward-related behaviors including those of the mesocorticolimbic pathway and the brain's “hedonic hotspots” (8, 47). For example, optogenetic stimulation of OX-neurons-to-VTA projections increase the preference for and seeking of food rewards (48), and OXR antagonists reduce seeking and motivation for drugs of abuse (49, 50). In addition, OX signaling has been proposed to promote the motivational activation necessary for interaction with the environment and adaptive behaviors (51).

The first drugs targeting orexin receptors demonstrating pharmacodynamic effects were two DORAs [namely, almorexant (ACT-078573) (11) and SB-649868 (52)]. Three DORAs are currently approved for the treatment of insomnia (i.e., suvorexant, lemborexant, daridorexant); furthermore, other psychiatric indications are also being investigated using DORAs or selective OX1R or OX2R antagonists (i.e., SO1RA or SO2RA) (13). OX2R agonists (SO2RAgs) with wake-promoting effects are in development for the treatment of narcolepsy (53, 54).

Different treatment regimens were tested in trials with OXR antagonists, depending on the targeted symptoms and the drug used. For example, in trials with DORAs/SO2RAs for insomnia and depression (55, 56) the drugs were administered at bedtime, and in a trial with SO1RA for binge eating disorder the drug was given during morning and evening meals (NCT04753164) (57). In the context of the possible use of OXR antagonists for the treatment of NPS in PwD, bedtime administration has the advantages of maximizing the drug's efficacy on nighttime problems and reducing the risk of daytime somnolence; however, little effect may be expected on disturbances that mostly manifest during the day [except those instances where amelioration of daytime symptoms is achieved through improved nighttime problems, as in the case of depression (56)]. Conversely, daytime dosing may have a greater impact on symptoms, which are expressed during the day, but (depending on the dose) it may be associated with a higher risk of somnolence. SO2RAgs, given in the morning to stabilize daytime wakefulness, may potentially also improve sleep at night by secondary means and thus normalize the disturbed wake-sleep rhythms observed in PwD.

PwD frequently present with difficulties in falling asleep (58), fragmented sleep (59), and excessive sleepiness during daytime (60). The sleep-wake disturbance in PwD leads to a progressive deterioration of circadian rhythm (61) and, in advanced stages of dementia, even to a reversal of the day-night sleep pattern (58). Sleep disturbance in PwD has been associated with the presence of other NPS such as anxiety and hyperactivity (62).

OXR antagonists administered at bedtime might have beneficial effects on sleep-related NPS in PwD. Indeed, two recent studies showed that sleep problems in AD can be ameliorated by DORAs (i.e., suvorexant:(63); lemborexant:(64)). Both trials showed improved sleep during the night. Actigraphy measurements in the lemborexant study demonstrated reduced sleep bouts during daytime, indicating a less disturbed sleep-wake rhythm as consequence of the nighttime treatment. Possibly due to the relatively long half-life of the drug, daytime somnolence was reported in the suvorexant trial, although this was not considered severe (65). Despite these beneficial effects on sleep parameters, effects of DORA on other nighttime NPS, including nighttime agitation, have not yet been assessed.

Given the wake-promoting effects of SO2RAgs, morning dosing may reduce the excessive daytime sleepiness observed in PwD and in addition help restore a normal sleep-wake rhythm.

Agitation is one of the most frequent and pervasive NPS in PwD. Agitated behaviors are accompanied by signs of emotional distress and may include excessive motor activity (e.g., pacing, rocking, restlessness), verbal aggression (e.g., yelling, screaming), and/or physical aggression (e.g., grabbing, scratching, slamming doors) (66). Agitated behaviors can occur both during the day and the night (67), with nighttime agitation posing a huge burden for caregivers (68). Most aggressive behaviors shown by PwD can be interpreted as being reactive to environmental circumstances (e.g., lack of understanding of the caregivers' intentions, leading to rejection of care) (69). Other hyperactive behaviors include irritability and aberrant motor behavior, which may present as wandering, purposeless activity (e.g., insistently repeating demands or questions) and inappropriate activities (e.g., hiding objects in inappropriate places) (70). Anxiety is also highly prevalent in PwD (71), and it is expressed with both psychological and somatic manifestations (e.g., worry, palpitations, shortness of breath) (72).

From a neuropsychological perspective, hyperactive NPS and agitation, in particular, have been proposed to stem from an aberrant regulation of emotional salience, which may lead to overestimation of threat, hypervigilance and lowered stress thresholds (2).

As described above, the OX system regulates several behaviors of relevance for hyperactive NPS and anxiety. Specifically, OX system activation increases the motor output and stress reactivity, and OXR antagonists reduce reactive aggression, stress-induced behavioral changes, and anxiety in animals. Initial evidence in acute challenge tests in humans also indicate anxiolytic or stress-reducing capacity of SO1RAs or suvorexant (73–75). These laboratory findings suggest that an OXR antagonist might reduce hyperactive NPS and anxiety. One behavior that could benefit the most from bedtime administration would be nighttime agitation, because of the direct impact of the drug on both sleep and emotional reactivity.

Interestingly, the first beneficial effects of DORAs on hyperactive behaviors have been recently demonstrated in the case of nocturnal delirium. Delirium is a common condition in elderly and critically ill patients (76), characterized by a disturbance of consciousness which can be accompanied by hyperactive behaviors (77). Administration of suvorexant has been shown to improve nocturnal delirium in PwD and critically ill patients (76) (studies conducted only in Japan for the moment). These findings suggest that hyperactive behaviors may benefit from administration of DORAs; however, properly designed, controlled studies are needed to assess whether DORAs/SORAs improve hyperactive NPS in PwD. A clinical trial (NCT05307692) currently investigates the effects of seltorexant (a SO2RA) specifically on agitation as a primary outcome in patients with AD (78). Nobody has yet explored the effect of DORAs/SORAs on hyperactive NPS in the context of neuropsychiatric conditions other than PwD.

To the best of our knowledge, no trial has yet studied the effect of OXR antagonists on hyperactivity, when given during the day. Such timing of administration might attenuate hyperactive daytime symptoms but, depending on the level on OX2R antagonism exerted by the drug, could simultaneously also increase the daytime sleepiness.

Apathy is another NPS prevalent in PwD, which often presents itself already in the pre-dementia phase. Apathy is characterized by a lack of motivation, reduced interest in rewarding situations, and akinesia and has major impact on patients' quality of life (2). As mentioned above, preclinical studies showed that activation of the OX system positively regulates reward processing and motivational activation, and OXR antagonism reduces seeking of and motivation for rewards.

Given the blunting of reward perception by antagonism of orexin signaling, one can hypothesize that OXR antagonists may worsen the apathetic phenotype shown by PwD, especially if the drugs are administered at daytime, when patients engage and interact with their environment. However, in several neurodegenerative disorders, apathy is positively correlated with sleep problems and with low levels of activity during the day (79–81), raising the question of whether restoring the sleep-wake rhythm [i.e., by reducing sleep fragmentation and increasing daytime activity as shown by the lemborexant trial (64)] may have beneficial effects on apathy. Given that orexins promote motivational activation, daytime treatment with SO2RAgs could also alleviate apathetic behaviors.