FormalPara Lemborexant (DAYVIGO™): Key Points

An orexin receptor antagonist was developed by Eisai Inc. for the treatment of adults with insomnia and is being investigated for the treatment of ISWRD associated with mild to moderate Alzheimer’s disease

Received its first approval (with interim final scheduling) on 20 December 2019 in the USA; approved in Japan on the 23 January 2020

Approved for use in adult patients for the treatment of insomnia, characterized by difficulties with sleep onset and/or sleep maintenance

1 Introduction

Insomnia, a chronic sleep disorder, imposes a heavy burden from a societal (e.g., lost productivity, missing work) and individual’s (e.g., daytime sleepiness, fatigue, irritability, memory impairment, increased potential for substance abuse) perspective, with women reporting insomnia approximately twice as often as men and its likelihood increasing with age [1]. Insomnia is associated with an increased risk of mortality (in the elderly, poor sleepers have twice the mortality rates as good sleepers) and morbidity (e.g., obesity, cardiometabolic disease, stroke, chronic pain, depression and anxiety) [1, 2]. First-line treatment of insomnia generally involves cognitive behavioural therapy, although most patients will decide on or be offered pharmacotherapy for its management [3]. Almost all pharmacotherapy for treating insomnia disorder involves the use of γ-aminobutyric acid agonists (e.g., zolpidem, benzodiazepines) or monoamine agonists (e.g., trazadone), many of which are associated with adverse effects such as unsteady gait and confusion [3, 4].

An improved understanding of the orexin peptides and their receptor signalling pathway provided a novel therapeutic target for treating insomnia via antagonism of orexin receptors (OXRs) [4, 5]. Orexin peptides (A and B peptide) are involved in various physiological processes, including playing a key role in regulating the sleep-awake cycle, with orexin neurones active during wakefulness and inactive during rapid eye movement (REM) and non-REM sleep. Orexin signalling is mediated via OXR1 and OXR2 [4, 5].

figure a

Key milestones in the development of lemborexant for the treatment of insomnia disorder. DEA Drug Enforcement Agency, MAA marketing authorization application, NDA new drug application, PDMA Pharmaceuticals and Medical Devices Agency, PDUFA priority review under Prescription Drug User Fee Act

Lemborexant (DAYVIGO™), an orally administered dual OXR1 and OXR2 antagonist, was discovered and developed by Eisai Inc., for the treatment of adults with insomnia disorder and is being investigated for the treatment of irregular sleep-wake rhythm disorder (ISWRD) associated with mild to moderate Alzheimer’s disease. On the 20 December 2019, the US FDA approved lemborexant (with final interim scheduling) for the treatment of adult patients with insomnia, characterized by difficulties with sleep onset and/or sleep maintenance [6]. Lemborexant was also approved in Japan in January 2020 for use in the treatment of insomnia [7]. It is also under regulatory review in Canada for use in this population [8]. In the USA, the recommended dosage of lemborexant is 5 mg taken no more than once per night, immediately before going to bed, with at least 7 h remaining before the planned time of awakening [9]. The dosage may be increased to 10 mg/night, based on clinical response and tolerability. The maximum recommended dosage is 10 mg/day. Lemborexant is contraindicated in patients with narcolepsy [9].

1.1 Company Agreements

In May 2019, Eisai Inc. re-acquired the worldwide rights for lemborexant from Purdue Pharma, with global development and commercialisation of lemborexant to be conducted solely by Eisai Inc. Lemborexant was discovered and developed by scientists at Eisai Inc. [10].

2 Scientific Summary

2.1 Pharmacodynamics

The mechanism of action of lemborexant is presumed to be via its dual OXR antagonism [9, 11]. In vitro, lemborexant acts as a competitive antagonist at OXR1 and OXR2, with respective 50% inhibitory concentrations (IC50) of 6.1 nmol/L and 2.6 nmol/L [9, 11]. A major metabolite of the lemborexant (M10) binds with comparable affinity as the parent drug to OXR1 and OXR2 (IC50 4.2 and 2.9 nmol/L) [9]. Competitive antagonism at OX1R and OX2R interferes with orexin neurotransmission to facilitate sleep onset and maintenance, without interfering with the ability to awaken to external stimuli [4, 5]. In vitro, lemborexant exhibited rapid binding and dissociation kinetics from OXRs, with high selectivity for OXR1 and OXR2 versus 88 other physiologically important receptors, transporters and ion channels [12].

figure b

Chemical structure of lemborexant

In mice, lemborexant (10 and 30 mg/kg) significantly (p < 0.001 vs vehicle) reduced wakefulness time during the first 3 h postdose, with a significant increase in both REM (p < 0.01 both doses) and non-REM sleep (only significant for 30 mg/kg dose; p < 0.001) [11]. In vivo in wild-type rat and mouse studies, lemborexant increased sleep time and enhanced REM and non-REM sleep equally (vs vehicle), with no change in the REM sleep ratio [13]. The dual mechanism of action of lemborexant on OXRs was supported by the lack of effect of lemborexant on sleep promotion in an orexin neuron-deficient mouse model. In wild-type and orexin neuron-deficient mice, lemborexant did not potentiate the sedative effects of ethanol or impair motor co-ordination [13].

In healthy volunteers, a supratherapeutic dose of lemborexant (5 × maximum recommended dose) did not prolong the Fridericia’s corrected QT (QTcF) interval to any clinically relevant extent, based on a concentration-QTcF analysis of two double-blind, placebo-controlled, ascending studies [9, 14].

2.2 Pharmacokinetics

Oral lemborexant is rapidly absorbed, with a median time to peak plasma concentration (tmax) of ≈ 1–3 h. The rate (tmax delayed by 2 h) and extent (decreased maximum plasma concentration by 23% and increased exposure by 18%) of absorption was altered when lemborexant was administered after a high-fat, high-calorie meal; sleep onset may be delayed if lemborexant is taken with or soon after a meal. In vitro, lemborexant is extensively bound to plasma protein (≈ 94%). The volume of distribution of lemborexant is 1970 L, with a blood to plasma ratio of 0.65 [9].

Lemborexant is primarily metabolized by CYP3A4 and, to a lesser extent, by CYP3A5. The major circulating metabolite is M10. Following oral administration, 57.4% of a lemborexant dose is eliminated in the faeces and 29.1% in the urine (< 1% as unchanged drug). The effective half-life of a lemborexant 5 mg dose is 17 h and for a 10 mg dose is 19 h [9].

There were no clinically relevant effects of age, sex, race/ethnicity and body mass index on the pharmacokinetics of lemborexant. No dosage adjustments are required in patients with mild, moderate or severe renal impairment; patients with severe renal impairment may experience an increased risk of somnolence. No pharmacokinetic studies have been conducted in paediatric patients [9]. No lemborexant dosage adjustment is required in patients with mild hepatic impairment (Child–Pugh class A) [9]; as exposure to the drug is increased in this population [15], patients may experience an increased risk of somnolence [9]. Dosage adjustment of lemborexant is recommended in patients with moderate hepatic impairment (Child–Pugh class B) [9], as exposure to and the terminal half-life of lemborexant is increased in this population [15]. In patients with moderate hepatic impairment, the initial and maximum dosage of lemborexant is 5 mg no more than once per night. Lemborexant has not been studied in patients with severe hepatic impairment (Child–Pugh class C); use in this population is not recommended [9].

In vitro, lemborexant and M10 have the potential to induce CYP3A4 and a weak potential to inhibit CYP3A4 and induce CYPB6. Lemborexant and M10 do not inhibit other CYP isoforms (CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2D6, CYP2A6, CYP2C19, CYP2E1) or transporters (P-gp, BCRP, BSEP, OAT1, OAT3, OATP1B1, OATP1B3, OCT1, OCT2, MATE1, MATE2-K). Lemborexant is a potential poor substrate for P-gp, whereas M10 is a substrate for this transporter. Lemborexant and M10 are not substrates for BCRP, OATP1B1 or OATP1B3 [9].

The concomitant use of weak CYP3A4 inhibitors increased lemborexant exposure by less than twofold, as predicted by physiologically-based pharmacokinetics modeling. Hence, clinically significant drug interactions are predicted between lemborexant and strong CYP3A4 inducers (e.g., rifampin, carbamazepine, St. John’s wort), moderate (fluconazole, verapamil) or strong (itraconazole, clarithromycin) CYP3A4 inhibitors, or CY2B6 substrates (bupropion, methadone). Avoid concomitant use of lemborexant and strong or moderate CYP3A4 inhibitors or inducers (moderate inducers include bosentan, efavirenz, etravirine, modafinil); the maximum recommended lemborexant dose with concomitant weak CYP3A4 inhibitors is 5 mg. There were no clinically relevant drug interactions between lemborexant and CYP3A4 substrates (midazolam) or oral contraceptives (ethinyl estradiol, norethindrone [9, 16]).

Table 2 Features and properties of lemborexant
Full size table

2.3 Therapeutic Trials

2.3.1 In Insomnia Disorder

Oral lemborexant 5 mg (n = 266) and 10 mg (n = 269) once nightly for 1 month was significantly (p < 0.0001) more effective than placebo (n = 208) at reducing objectively-assessed latency to persistent sleep (LPS; log-transformed, assessed using overnight polysomnography) in SUNRISE 1 (primary endpoint). SUNRISE 1 was a randomized double-blind, active comparator-controlled, multicentre phase III trial in adults (females aged ≥ 55 years and males aged ≥ 65 years) who met DSM-5 criteria for insomnia disorder, characterized by sleep maintenance difficulties. Patients could also have sleep onset difficulties, but this was not required. There were no significant differences between treatment groups for demographics and baseline characteristics. The overall mean age of patients was 64 years; 86.4% of patients were women [17].

In the lemborexant 5 mg, lemborexant 10 mg and placebo groups, geometric mean reductions in LPS at 1 month were 19.5 min, 21.5 min and 7.9 min, respectively, with least-squares geometric mean (LSGM) treatment ratios for the lemborexant 5 mg and 10 mg groups (both p < 0.001) of 0.77 (95% CI 0.67–0.89) and 0.72 (95% CI 0.63–0.83). Both dosages of lemborexant were also associated with significantly greater reductions in LPS at 1 month than zolpidem 6.25 mg once nightly (n = 263), with LSGM treatment ratios for the lemborexant 5 mg and 10 mg groups (both p < 0.001) of 0.63 (95% CI 0.56–0.72) and 0.59 (95% CI 0.52–0.68). Improvements in LPS were also significantly (LSGM treatment ratios all p ≤ 0.02) greater on the first two nights of treatment in the lemborexant 5 mg and 10 mg groups than in the placebo or zolpidem groups. Mean baseline LPS values were 28.3 min, 25.1 min, 31.9 min and 37.4 min in the lemborexant 5 mg, lemborexant 10 mg, zolpidem and placebo groups, respectively [17].

Sleep maintenance outcomes also favoured lemborexant treatment over placebo or zolpidem (key secondary endpoints; assessed using polysomnography), including sleep efficiency (SEF; i.e. the proportion of time spent asleep per time in bed) and wake-after-sleep-onset (WASO) at 1 month (all p ≤ 0.002) and on nights 1 and 2 (all p ≤ 0.02) [17]. In the lemborexant groups, improvements in WASO mainly reflected an improvement in the latter half of the sleep period, as determined by mean decrease in WASO in the second half of the night (i.e., minutes of wake time from 240 min after lights off until lights on; all p ≤ 0.004 vs placebo or zolpidem) [17]. Lemborexant treatment also significantly (both dosages p <0.0001) decreased (improved) the severity of insomnia symptoms compared with placebo at 1 month, based on Insomnia Severity Index total scores [18].

Oral lemborexant 5 mg (n = 316 [9]) and 10 mg (n = 315 [9]) once nightly were both superior to placebo (n = 318 [9]) for the mean change from baseline to 6 months for log-transformed subjective sleep onset latency (sSOL) [primary outcome; patient-assessed using a sleep diary] in SUNRISE 2 [9, 19, 20]. This randomized, multinational, phase III trial enrolled patients with insomnia disorder (based on DSM-5 criteria) characterized by difficulties with sleep onset and/or sleep maintenance. Participants had a median age of 55 years (28% were aged ≥ 65 years) and 68% were females [9]. The trial consisted of an initial placebo-controlled, double-blind, 6-month period, followed by a 6-month, blinded active treatment period. Patients were randomized to lemborexant 5 mg, lemborexant 10 mg or placebo once nightly for the first 6 months, after which, placebo recipients were re-randomized to lemborexant 5 mg or 10 mg for 6 months and lemborexant recipients continued with the same dosage of lemborexant [21].

At 6 months, LSGM reductions from baseline in sSOL in the lemborexant 5 mg and lemborexant 10 mg groups were 24.1 min and 23.0 min (respective mean baseline values 57.3 and 60.2 min) [no data reported for placebo group] [19]. The beneficial effects of both dosages of lemborexant on sSOL were evident from 1 month (LSGM change from baseline − 17.2 and − 18.6 min in the lemborexant 5 and 10 mg groups) and persisted at 12 months (− 25.8 and − 26.3 min; n = 444 and 437). Key secondary outcomes for sleep maintenance (patient-assessed using a sleep diary) of subjective SEF (sSEF) and subjective WASO (sWASO) also significantly favoured (all p < 0.05 vs placebo) both dosages of lemborexant over placebo at 1 month and persisted at 6 months (superiority shown for both outcomes for both lemborexant dosages vs placebo [9]) and 12 months. Respective LSM increases from baseline in sSEF in the lemborexant 5 mg and 10 mg groups at 1 month were 6.4 min and 7.3 min, at 6 months were 11.1 min and 11.1 min, and at 12 months were 12.6 min and 13.7 min. For sWASO, respective LSM reductions from baseline in the lemborexant 5 mg and 10 mg groups at 1 month were 17.3 min and 18.7 min, at 6 months were 36.1 min and 31.5 min, and at 12 months were 42.9 min and 43.8 min. Baseline mean sSEF values in the lemborexant 5 mg and 10 mg groups were 65.3 min and 64.7 min, with respective mean sWASO values of 124.8 min and 126.5 min [19]. There was no difference in the effectiveness of lemborexant treatment on these sleep outcomes based on age [9], race [9] and sex [9, 22].

Patients’ perception of study medication effects was assessed in SUNRISE 2 using subjective Patient Global Impression-Insomnia (PGI-I) scores [20]. At 6 months, a significantly (all p < 0.0001) higher proportion of patients in the lemborexant 5 mg and 10 mg groups than in the placebo group reported that their study medication helped sleep (67.3 and 68.8 vs 45.0%, respectively) and reduced the time to fall asleep (72.8 and 73.1 vs 46.1%) [20].

Based on a pooled analysis of SUNRISE 1 and 2, lemborexant 5 mg or 10 mg once nightly provided significantly (all p < 0.001) better efficacy than placebo for the first week of treatment, including for sSOL, sWASO and sSEF [23].

Results from a dose-finding, randomized, double-blind, multicentre phase II trial in adults and elderly patients (aged ≥ 65 years) with insomnia disorder indicated that lemborexant 2.5–10 mg once nightly for 15 days was effective for treating insomnia, whilst minimizing next-morning residual sleepiness (primary endpoint). On days 2 and 3, within 15 min and at 1 h and 2 h after wake-time, there were no significant differences in the mean change from baseline in Karolinska Sleepiness Scale (KSS) scores between lemborexant (1–10 mg) and placebo groups. At higher than recommended dosages of lemborexant (15 and 25 mg), there were significant (p < 0.05) differences between lemborexant and placebo groups for some timepoints for KSS scores on days 2 and 3 and on days 15 and 16 [24].

2.3.2 Irregular Sleep-Wake Rhythm Disorder Associated with Alzheimer’s Disease

In a 28-day, phase II, proof-of-concept study, lemborexant improved 24-h circadian rhythm and nocturnal sleep variables in patients with ISWRD associated with Alzheimer’s disease. This global randomized, double-blind trial enrolled patients (aged 60–90 years) who met DSM-5 criteria for ISWRD and mild to moderate Alzheimer’s disease (Mini-Mental State Exam score 10–26). After completion of a pre-randomization phase (14–28 days), patients were randomized to lemborexant 2.5–15 mg or placebo once nightly for 28 days, after which they were eligible to enter an ongoing open-label extension phase of up to 30 months’ duration [25].

Oral lemborexant 5 mg (n = 13) and 15 mg (n = 15) once nightly significantly (p < 0.5 vs placebo; n = 12) increased the relative amplitude of the rest activity rhythm and reduced nocturnal activity levels at 28 days (assessed using actigraph data and caregiver-reported sleep diary). Sleep Fragmentation Index scores were also numerically lower (higher scores mean more fragmented sleep) and total sleep time during the night numerically higher in these two lemborexant groups than in the placebo group at this timepoint. The mean duration of sleep bouts during the day was reduced with lemborexant treatment, as was the activity of the least active 5 h of the day (i.e., the average activity across the least active 5-h period of the 24-h rest-activity rhythm; higher scores mean more activity). Improvements in these parameters support the consolidation of sleep at night-time [25].

Table 3 Key clinical trials of lemborexant
Full size table

2.4 Adverse Events

Oral lemborexant 5 mg or 10 mg once nightly was generally well tolerated in adults with insomnia disorder during up to 12 months’ treatment, with a tolerability profile similar to that with placebo treatment [17, 22]. The tolerability profile of lemborexant in adults with ISWRD associated with mild to moderate Alzheimer’s disease was similar to that in adults with insomnia disorder [25].

Based on a pooled safety analysis of SUNRISE 1 (30 days) and SUNRISE 2 (6 months) in adults with insomnia disorder, relatively few patients in the lemborexant 5 mg (n = 580) and 10 mg (n = 582) groups discontinued treatment because of an adverse reaction (1.4 and 2.6 vs 1.5% of patients in the placebo group; n = 528 in placebo group) [9]. The most common adverse reactions leading to treatment discontinuation during the first 30 days in the lemborexant 5 mg, lemborexant 10 mg and placebo groups were somnolence (0.7, 1.0 and 0.4%, respectively) and nightmares (0.3, 0.3 and 0%). During the 6-month treatment period in SUNRISE 2, the most common adverse reactions leading to treatment discontinuation in the lemborexant 5 mg, lemborexant 10 mg and placebo groups were somnolence (1.0, 2.9 and 0.6%, respectively), nightmares (0.3, 1.3 and 0%) and palpitations (0, 0.6% and 0%). The most common adverse reaction reported in ≥ 5% of lemborexant recipients and at a frequency at least twice the rate of placebo during the first 30 days of SUNRISE 1 and SUNRISE 2 was somnolence (7, 10 and 1%, respectively). Adverse reactions reported in ≥ 2% of lemborexant recipients (5 and 10 mg groups) and at a higher frequency than in the placebo groups during this period were somnolence or fatigue (6.9, 9.6 and 1.3%, respectively), headache (5.9, 4.5 and 3.4%) and nightmare or abnormal dreams (0.9, 2.2 and 0.9%). Sleep paralysis was reported in 1.3% and 1.6% of patients in the lemborexant 5 mg and 10 mg groups (vs 0% in the placebo group), with hypnagogic hallucinations reported in 0.1%, 0.7% and 0% of patients, respectively. Two events of complex sleep behavior were reported in patients receiving lemborexant 10 mg once nightly. Treatment with lemborexant should be discontinued immediately if a complex sleep behavior occurs, including sleep-walking, sleep-driving and engaging in other activities while not fully awake) [9].

Lemborexant did not increase the frequency of apneic events or cause oxygen desaturation in a randomized, double-blind, crossover, multicentre study in patients with mild obstructive sleep apnea (OSA; apnea-hypopnea index < 15 events/h of sleep) [9, 26]. Lemborexant has not been studied in patients with chronic obstructive pulmonary disease or moderate to severe OSA; clinically meaningful respiratory effects in these populations cannot be excluded. In patients with compromised respiratory function, the effect of lemborexant treatment on respiratory function should be considered [9].

In healthy female (aged ≥ 55 years) or male (aged ≥ 65 years; median age 49 years) subjects receiving lemborexant 5 mg or 10 mg at night, there was no meaningful difference on the ability to awaken to sound 4 h postdose (vs placebo), as assessed using an audiometer that delivered 1000 Hz tones up to 105 dB, in a randomized, double-blind, placebo- and active comparator-controlled trial. Balance was impaired (measured by body sway) at 4 h postdose in lemborexant recipients compared with placebo recipients, with lemborexant treatment also associated with dose-dependent worsening (vs placebo) on measures of attention and memory at 4 h postdose. Patients should be cautioned about the potential for middle of the night postural instability, as well as attention and memory impairment [9]. There were no clinically meaningful differences between lemborexant (5 or 10 mg) and placebo on next-day postural stability or memory when study treatment was administered ≈ 8 h prior, based on results of two randomized, placebo- and active comparator-controlled trials in healthy adults (E2006-A001-108) and adults with insomnia disorder (SUNRISE 1) [9, 27].

In healthy elderly (aged ≥ 65 years) or younger adults (aged ≥ 18 years; median age 49 years) receiving lemborexant 5 mg or 10 mg, there was no statistically significant impairment in next-morning driving performance (vs placebo), as assessed using the change in standard deviation of lateral position in a randomized, double-blind placebo- and active comparator-controlled, four-period crossover trial (NCT02583451) [9, 28]. However, the ability to drive was impaired in some patients receiving lemborexant 10 mg; patients taking this dose should be cautioned about the potential for next-morning driving impairment because there is individual variation in sensitivity to lemborexant [9].

Sleep paralysis, an inability to move or speak for up to several minutes during sleep-wake transition and hypnagogic/hypnopompic hallucinations, including vivid and disturbing perceptions, can occur during lemborexant treatment. These events should be explained to patients by prescribers when prescribing lemborexant. Symptoms similar to mild cataplexy can occur with lemborexant treatment. These symptoms can include periods of leg weakness lasting from seconds to a few minutes, can occur at night or during the day and may not be associated with an identified triggering event (e.g., laughter or surprise) [9].

Lemborexant was not associated with rebound insomnia following treatment discontinuation, based on sleep diary-recorded sSOL and sWASO from the screening period to the 2 weeks following treatment discontinuation in SUNRISE 1 and 2 [9].

There was no evidence of withdrawal effects following lemborexant discontinuation at either dose (5 mg and 10 mg), as assessed using the Tyrer Benzodiazepine Withdrawal Symptom Questionnaire following treatment discontinuation in SUNRISE 1 and 2 [9].

In a study in recreational sedative abusers, lemborexant 10 mg, 20 mg (twice recommended dose) and 30 mg (three times the recommended dose) was associated with responses on positive subjective measures (e.g., drug liking, overall drug liking, take drug again, good drug effect) that were similar to those produced by the sedatives zolpidem 30 mg and suvorexant 40 mg, and statistically greater than with placebo. Since individuals with a history of abuse or addiction to alcohol or other drugs may be at increased risk for abuse and drug addiction to lemborexant, follow such patients carefully. Lemborexant has final interim scheduling; controlled substance schedule to be determined after review by the US Drug Enforcement Agency [9].

2.5 Ongoing Clinical Trials

A multicentre pilot study (NCT04009577; E2006-A001-312) is currently evaluating the effectiveness of transitioning from zolpidem (immediate- or extended-release), taken intermittently or frequently, to lemborexant 5 mg or 10 mg, as determined at the end of 2 weeks of lemborexant treatment in adults with insomnia disorder, with an anticipated completion date of July 2020. In addition, an ongoing open-label extension of a phase II study (NCT03001557; E006-G000-202; EudraCT2017-003306-40) is evaluating the long-term efficacy and safety of lemborexant in adults with ISWRD associated with mild to moderate Alzheimer’s disease, with an anticipated completion date of April 2020.

3 Current Status

On the 20 December 2019 [6], lemborexant received its first global approval (with final interim scheduling) in the USA for the treatment of adult patients with insomnia, characterized by difficulties with sleep onset and/or sleep maintenance [9]. Lemborexant was approved in Japan on the 23 January 2020 for use in the treatment of insomnia [7]. It is also under regulatory review in Canada for use in this population [8].