Insomnia is a highly prevalent disorder, characterized by either non-restorative sleep or difficulties initiating or maintaining sleep [1], which affects up to approximately 20% of adults in the United States (U.S.) [2]. Compromised sleep quality impacts on the daytime functioning of affected individuals, especially in social and occupational domains, leading to impaired overall quality of life and an increased risk of psychiatric comorbidities [3-5]. The prevalence of insomnia increases with age, with estimates ranging from 20–50% in the elderly population [6], which makes elderly adults the most common users of hypnotic drugs [7]. Insomnia in the elderly is associated with affective disorders, increased healthcare utilization [8], higher risk of accidents and falls [6,9], and cognitive impairment that can be misdiagnosed as dementia [10].

Currently, first-line pharmacotherapy for insomnia relies on the use of positive allosteric modulators of gamma-aminobutyric acid type A (GABA_A) receptors, which induce sedation through the potentiation of the inhibitory signaling of GABA_A receptors [11,12]. Although effective in promoting sleep, GABA_A receptor modulators elicit global inhibitory effects in the central nervous system that make them liable to producing a variety of side effects, including rebound insomnia, memory failure, and abuse potential [11,13]. Despite being the primary treatments for insomnia in elderly patients, GABA_A receptor modulators may not be an ideal treatment for this patient population, as they are associated with impairment of motor function and coordination in elderly patients, along with increased rates of falls and motor vehicle crashes [14,15].

Safety concerns associated with GABA_A modulators prompted the clinical development of sleep-promoting compounds that facilitate sleep via more specific and targeted mechanisms of action, and this was anticipated to translate into an improved side-effects profile. For example, orexin receptor antagonists (ORAs) have been developed to promote sleep by blocking physiological effects of the orexin compounds, which are key mediators in the maintenance of arousal and vigilance [16]. The ORA suvorexant has demonstrated sleep-promoting effects with minimal residual next-day somnolence or psychomotor impairment in clinical studies, including in elderly insomnia patients [7,17-20]. Other pharmacological treatments for insomnia include selective antagonists of histamine H1 and α1-adrenergic receptors, antidepressants and antipsychotic drugs, which are typically prescribed off-label, melatonin receptor agonists, and over-the-counter antihistamines [11]. Non-pharmacological therapies, such as cognitive behavioral therapy, have also shown efficacy as an alternative to pharmacotherapy [21].

Treatment with low doses of sedating antidepressants may be a promising approach for those patients who are more likely to experience sleep-maintenance insomnia, such as the elderly. Antidepressants have relatively discrete activity and act by modulating neurotransmitters, such as histamine, serotonin, and norepinephrine [22]. To date, low doses (3 and 6 mg) of the antidepressant doxepin (SilenorTM, Currax Pharmaceuticals LLC, Morristown, NJ, USA), a modulator of the histamine H1 receptor, have been approved in the U.S. for the treatment of insomnia characterized by difficulties with sleep maintenance [23]. At low doses, doxepin demonstrated efficacy in improving sleep maintenance and was associated with a favorable safety profile in both younger (aged 18–64 years) [24] and older (aged ≥65 years) adults [25]. Several other antidepressants have been prescribed off-label in the management of insomnia, but the scientific evidence supporting their use is weak [26].

Mirtazapine is an antidepressant with sleep-promoting properties [27,28] that acts via antagonism of histamine H1 receptors, 5-HT₂ and 5-HT₃ serotonergic receptors, and adrenergic α2autoreceptors [29]. It is a racemic mixture of S(+) and R(-) enantiomers, with the former characterized by a shorter half-life compared with the latter. The maleic salt of S(+) mirtazapine, referred to as esmirtazapine, was advanced to the clinical phase as an investigational therapy for insomnia based on the consideration that the shorter half-life of esmirtazapine (10 h) relative to the racemic mixture (range: 20 to 40 h [30]) would diminish risk of the emergence of associated side effects, in particular residual next-day effects. In clinical studies, acute (2-day) and 2- and 6-week treatment with esmirtazapine in non-elderly participants (aged 18 to <65 years) with insomnia demonstrated significant improvements in objective and subjective sleep measures at doses of 1.5 mg, 3.0 mg, and 4.5 mg relative to placebo, and was generally well tolerated, with minimal residual daytime effects [31-34] and no signs of rebound insomnia [31,32].

Given the burden of insomnia and the high prevalence of sleep-aid use in the geriatric population, the current study was conducted in elderly outpatients (aged ≥65 years) with chronic primary insomnia. The primary objective was to investigate the safety and tolerability of long-term (52-week) treatment with esmirtazapine at doses of 1.5 mg and 3.0 mg. The secondary objective was to collect exploratory efficacy data with respect to sleep improvement.

The current study examined the long-term safety of esmirtazapine in elderly patients with a diagnosis of chronic insomnia. The efficacy assessments were exploratory in nature, as no placebo comparator was included. The results indicate reasonable tolerability of both investigated doses of esmirtazapine (1.5 mg and 3.0 mg) administered before bedtime for the long-term treatment of elderly patients with insomnia. The exploratory efficacy data demonstrated a marked improvement from baseline in subjective sleep diary measures (i.e., TST, WASO, SL). Similarly, improvements from baseline were also noted in the ISI and IGR-S. These improvements from baseline were sustained over the 52-week treatment period and did not exhibit apparent differences between the two dose groups. The percentage of participants completing the current study (59.1%) was comparable with the completion rate (65.4%) of a 1-year, open-label, extension phase of two randomized trials of zaleplon [35].

The safety of 1.5-mg, 3.0-mg, and 4.5-mg doses of esmirtazapine was previously examined for a shorter duration of treatment and in a younger population of adult patients with insomnia (aged ≤65 years) [31,32,34]. In these studies, esmirtazapine was well tolerated with minimal residual effects. The incidence of AEs and SAEs was higher in the current study relative to previously published studies of esmirtazapine treatment in non-elderly patients with primary insomnia [31,32]. In a 2-week study in non-elderly patients, 25.5% and 32.8% of patients had AEs, and 1.5% and 0.8% had SAEs, with esmirtazapine 1.5 mg and 3.0 mg, respectively. In a 6-week study in non-elderly patients, 35.3% of patients had AEs following esmirtazapine 3.0 mg administration. No SAEs were reported with esmirtazapine 3.0-mg in the 6-week study. However, cross-study comparisons should be interpreted with caution due to the differences in study design and participant population. Consistent with previous findings from 2- and 6-week studies of esmirtazapine treatment in non-elderly patients, the current study results indicate that somnolence was among the most commonly reported AEs. This effect did not seem to interfere with overall daily functioning as assessed by the weekly diary, given the sustained improvements in scores of alertness at awakening, energy level, and the ability to work/function throughout the treatment period as compared with baseline. Increase in body weight was also a frequently recorded AE (8.6% and 15.3% in the 1.5-mg and 3.0-mg groups, respectively). This is not surprising, given that this side effect has been previously associated with mirtazapine treatment for depression [36,37] and was attributed to its mechanism of action, which is based on the blockade of H1, and 5-HT₂, and 5-HT₃ receptors [29].

The fall rate observed in this study was relatively low: three participants (2.3%) receiving a dose of 1.5 mg esmirtazapine reported a fall. While treatment of insomnia symptoms in elderly patients frequently leads to the prescription of GABAA receptor modulators [38], this population is more sensitive to the AEs of such therapeutic agents: it is reported that one in seven elderly patients treated with benzodiazepines will have an AE [39]. One study reported that 40% of elderly patients admitted for emergency care after a fall had been prescribed benzodiazepines and Z-drugs [40]. Furthermore, compared with other times, it is reported that the fall risk for nursing-home residents is greater (odds ratio 3.79; 95% confidence interval 1.10–13.00) within 24 h of benzodiazepine administration [41]. Increased caution when prescribing Z-drugs for long-term treatment of insomnia in the geriatric population is, therefore, recommended [42]. The relatively low fall rate observed for esmirtazapine in the current study is of interest; however, study limitations should be taken into consideration.

Discontinuation due to insufficient therapeutic effect appeared to occur more frequently in the 3.0-mg group than the 1.5-mg group. Although the small sample size limits the ability to draw conclusions regarding this finding, and a definitive explanation is therefore unclear, greater residual effects at the higher dose, which may have diminished participants’ interpretations of efficacy, may have contributed. This may be supported by the observation that approximately twice the number of participants in the 3.0-mg group reported somnolence and dizziness compared with the 1.5-mg group. Lack of compliance with taking medication per protocol could be another contributing factor; however, cumulative compliance in this study was 95.8% for the 3.0-mg group and 97.6% for the 1.5-mg group, and is unlikely to fully explain the difference.

The observation that both doses of esmirtazapine showed no apparent deleterious effects on measures of alertness at awakening, energy level, and ability to work/function (as captured with the sleep diary) is consistent with data from previous studies that have assessed the potential for next-day hangover effects from esmirtazapine in non-elderly adults with insomnia [31,32]. Following a 2-week treatment with esmirtazapine (1.5 mg, 3.0 mg, and 4.5 mg), there was no evidence of residual effects in the esmirtazapine groups [31]. A 6-week treatment with esmirtazapine also demonstrated minimal residual next-day effect, as evaluated by a digit symbol substitution test, VAS score for morning alertness, and the Bond-Lader rating scale [32]. In another study that evaluated effects of an acute, 2-day treatment of esmirtazapine in patients with insomnia, a decrease in energy level and ability to work/function was experienced after the first night compared with placebo; however, this effect was not reported after the second night [34]. In the current study, both treatment groups showed improvement from baseline in alertness at awakening and energy levels, as assessed by the weekly diary, and both esmirtazapine doses provided similar median changes from baseline for both parameters. Sedation has been reported for mirtazapine [43], although no evidence of next-day sedation was observed among patients with major depressive disorder treated with mirtazapine (45 mg) as compared with those treated with fluoxetine (A Winokur, personal communication). A study assessing the effects of esmirtazapine 1.5 mg and 4.5 mg on next-day driving performance showed that single and repeated administration of esmirtazapine 1.5 mg had no clinically meaningful effects on driving performance. With a single dose of esmirtazapine 4.5 mg, driving impairment was observed; however, this resolved following repeated dosing [33]. A lack of prominent morning sedation observed with esmirtazapine in this study may be of interest to psychiatrists.

Previous reports on the efficacy of esmirtazapine in non-elderly adult patients with primary insomnia have indicated sustained beneficial effects on objective and subjective patient-reported measures of sleep onset and maintenance over the study period [31,32,34]. However, the examined treatment durations did not exceed 6 weeks, which does not reflect the long-term course of therapy that might be required when treating patients with chronic insomnia. The results of the present study suggest that esmirtazapine has enduring sleep-promoting effects in elderly patients over a period of 52 weeks, and that these effects did not appear to be accompanied by the development of tolerance. This is supported by the marked improvements in TST, WASO, SL, ISI, and IGR-S values compared with baseline at each of the assessments conducted throughout the study.

The apparent sleep-promoting properties of esmirtazapine should be considered in the context of long-term efficacy data of other pharmacological treatments for insomnia. While benzodiazepines have established short-term benefits with regard to improvement in sleep initiation and maintenance, their longterm efficacy has not been established [26]. As such, they are recommended for short-term use, with continued use believed to be associated with tolerance [26]. Long-term use of non-benzodiazepines has been more rigorously tested in clinical trials. For example, in a randomized, double-blind, placebo-controlled study, nighttime use of eszopiclone in patients (aged 21–69 years) with chronic insomnia demonstrated efficacy over a 6-month period [44]. Similar to the sustained improvements in TST, WASO, and SL over the year-long period of the current study, the authors reported that the median SL and WASO decreased from baseline by 30 min and 39 min, respectively, and median TST increased from baseline by 83 min after 6 months [44]. Similarly, zolpidem extended-release (12.5 mg) was well tolerated in non-elderly adults (aged 18–64 years) and improved sleep onset and maintenance parameters compared with placebo in a 6-month randomized, double-blind, placebo-controlled, multicenter study [45]. In elderly patients (aged 59–95 years) with insomnia, nightly zaleplon (5 mg and 10 mg) was associated with sustained improvements in sleep onset, sleep duration, and number of awakenings in a year-long, open-label extension study [35]. However, it should be noted that relatively few studies have reported the systematic monitoring of TST, WASO, and SL parameters over a comparably long time period, and this, combined with differences in trial design, limits the ability to draw direct comparisons with other therapeutics. The current study highlights the need for systematic assessment of longterm treatment effects of drugs for insomnia.

Several limitations should be considered when interpreting the findings of this study. The study design did not include a non-active (placebo) treatment arm and participants were only blinded to the dose of treatment they received, which limits efficacy evaluation. The contribution of factors other than drug activity to the frequency of AEs and reported therapeutic effect cannot, therefore, be excluded. Also, the sample size of the study was not determined by power analysis for detecting a significant difference at a fixed, Type I error rate. As a result, there was no formal hypothesis formulation and testing in this study. The study was only designed to assess within-group statistical differences relative to baseline values—not differences between the dose groups. Although there were no apparent differences between the two dose groups with regard to exploratory efficacy outcomes, dose–response measurements were not collected for this study, so only a tentative statement regarding the lack of apparent dose–response effects of esmirtazapine in improving insomnia sleep parameters long-term can be made. Finally, co-occurrence of insomnia with anxiety and/or depression is prevalent in the elderly population, and patients with such comorbidities were excluded from this study. This limits the generalizability of the findings and warrants further research efforts in patients with comorbid insomnia to establish the safety and efficacy of esmirtazapine in these subpopulations.

Our findings suggest that esmirtazapine doses of 1.5 mg and 3.0 mg nightly are reasonably tolerated in elderly outpatients with insomnia as long-term treatment (up to 1 year) without evidence of significant residual effects. Despite the lack of placebo and the limited power of the analysis, the persistent improvements in sleep parameters throughout this year-long study in elderly patients are encouraging, especially when considering tolerance issues associated with the long-term use of many hypnotics.