Impaired sleep in Parkinson’s disease (PD) was first identified by James Parkinson in his 1817 monograph “An essay on the shaking palsy”.1 Since then, researchers and clinicians have determined that most patients with PD experience some manner of disrupted nighttime sleep and/or daytime drowsiness; impaired sleep and sleep-related disturbances occur in up to 98% of patients2 and can worsen as PD progresses. The sleep disturbances can be associated with increases of other nonmotor symptoms and often have a major impact on daily functioning and quality of life for patients and their caregivers,3,4 as it can often lead to nighttime awakenings with the associated risk for falls.5 In addition, sleep disturbance has been recognized as a risk factor for the emergence of Parkinson disease psychosis (PDP).6 Patients with PD with sleep dysfunction are 5 times more likely to experience psychotic symptoms,7 and up to 82% of patients with PD with psychotic symptoms (ie, hallucinations or delusions) experience some type of sleep disturbance.8
The origins of disordered sleep in PD are multifactorial and may include disease processes affecting dopaminergic neurons that have a role in sleep-wake regulation.9 Other factors contributing to sleep disruption include rapid eye movement behavior disorder (RBD), wearing off of medications resulting in difficulty turning in bed, resting tremor, painful dystonic spasms, and the side effects of PD medications such as vivid dreaming, nightmares, and other sleep disorders common in this age group, such as sleep apnea and restless legs syndrome.10
Pimavanserin is a novel compound developed for the treatment of psychosis associated with PD. The pharmacology of pimavanserin is selective for serotonergic neurotransmission: it is a serotonin (5-HT)2A receptor inverse agonist/antagonist with >10-fold selectivity over 5-HT2C receptors in binding and functional assays. Pimavanserin has little or no activity at adrenergic, dopaminergic, histaminergic, or muscarinic receptors.11 In distinction to pimavanserin, other antipsychotic drugs are neither approved for PDP nor ideal as treatment options because of their potential for adverse effects, such as exacerbation of parkinsonism and sedation.
Pimavanserin’s action on 5-HT neurotransmission is pertinent to the challenge of improving sleep in PD, because 5-HT plays a critical role in regulating the sleep/wake processes. 5-HT2A receptors modulate cortical GABAergic interneurons that influence mechanisms promoting sleepiness, inhibiting rapid eye movement sleep, and increasing sleep drive.12,13 Because of the role of 5-HT in sleep, a polysomnography study was previously carried out in healthy older adults. In this 2-week study, pimavanserin produced robust increases in the quantity of slow wave sleep.14 A similar effect for the patient with PD experiencing sleep fragmentation might be beneficial, because sleep disruption can be a contributing factor for psychotic symptoms at nighttime.15
In the current study, we determined the effects of pimavanserin on both nighttime sleep quality and daytime sleepiness. The sources of information for this investigation were 2 placebo-controlled clinical trials conducted with pimavanserin in patients with PDP. These trials were designed for pimavanserin as an antipsychotic treatment in PDP, but also collected exploratory information on nighttime sleep and daytime sleepiness.16 Although the effectiveness of pimavanserin for treating PDP has been reported previously,16 we now analyze results from the sleep/wake data.
Two randomized, placebo-controlled, double-blind, multicenter studies were conducted in patients with PDP [ACP-103-012 (unpublished data) and ACP-103-02016]. The primary objective was the evaluation of pimavanserin at reducing the frequency and severity of hallucinations and delusions.
The first of these trials (ACP-103-012) evaluated 2 doses (taken once daily) of pimavanserin, 8.5 and 34 mg (equivalent to 10 and 40 mg of pimavanserin tartrate, respectively) for 42 days. The subsequent study (ACP-103-020), which was used as the Food and Drug Administration registrational trial, evaluated the safety and efficacy of 34 mg for the treatment of PDP also over 42 days.
Participants in each study had a diagnosis of idiopathic PD lasting for at least 1 year. In addition, their psychotic symptoms developed after the diagnosis of PD and had been present for at least 1 month before a screening visit. For participation in the studies, severity and frequency of psychotic symptoms (hallucinations and/or delusions) warranted treatment, in the opinion of the investigator. Patients were also required to have a Neuropsychiatric Inventory (NPI)17 score of 4 or higher on either the “hallucinations” or “delusions” subscales, or a total combined NPI “hallucinations and delusions” (NPI-H + D) score of 6 or higher for ACP-103-020. Patients in ACP-103-012 were required to have a total combined NPI-H + D score of 4 or higher and a Scale for the Assessment of Positive Symptoms (SAPS)18 “hallucinations plus delusions” (SAPS-H + D) score of 5 or higher. Participants’ Mini-Mental State Examination19 score had to be 21 or higher, and in addition, participants were required to show orientation to time and place. There was no requirement for sleep impairment at screening or baseline. Medications for treating PD were required to be maintained at stable doses for at least 1 month before enrollment and throughout the study. Further details of study design, including methodology and other procedures, are described elsewhere.16
Both studies were designed and powered for the analysis of the effect of pimavanserin on PDP using a measure of frequency and severity of hallucinations and delusions (ie, mean change from baseline to week 6 in the PD-adapted SAPS20 for ACP-103-020, and the SAPS-H + D for ACP-103-012). In addition to the end point of psychosis ratings, as an exploratory end point, participants were also evaluated using the Scales for Outcomes in Parkinson’s Disease (SCOPA) sleep score, which quantifies nighttime sleep quality and daytime sleepiness.21 This scale has high internal consistency for the nighttime sleep and daytime sleepiness scales (0.88 and 0.91, respectively) and test-retest reliabilities (0.94 and 0.89, respectively).21 Scores on the SCOPA-Sleep scale show high correlations between the nighttime sleep scale and the Pittsburgh Sleep Quality Index22 (0.83), and between the daytime sleepiness scale and the Epworth Sleepiness Scale23 (0.81).21 The coefficient of variation of both the nighttime sleep (SCOPA-NS) and the daytime sleepiness scale (SCOPA-DS) is higher than that of the Pittsburgh Sleep Quality Index and the Epworth Sleepiness Scale, indicating a better ability to detect differences between individuals. The SCOPA-NS consists of 5 items [with each item scored between 0 (“not at all”) and 3 (“a lot”); total score range, 0–15]. The SCOPA-DS consists of 6 items, each with 4 scoring elements [rated as 0 (“never”) to 3 (“often”); total score range, 0–18], with higher scores representing worse sleep quality and sleepiness. The SCOPA-Sleep questionnaire was completed by site clinical research staff based on subject interview. Information was collected at 5 time points during ACP-103-012 and 4 time points during ACP-103-020. Data from the placebo and pimavanserin 34 mg groups of the 2 studies (which excludes the 8.5 mg arm in ACP-103-012) were pooled to provide further information on the effect of pimavanserin 34 mg on sleep. Additional analyses on the pooled studies were performed to determine the effect of pimavanserin on sleep in participants with significant nighttime sleep and daytime sleepiness at baseline, defined as SCOPA-NS ≥7 and SCOPA-DS ≥5, respectively. Treatment effect for this analysis was measured for the full analysis set population, consisting of participants who received 1 or more doses of study drug and had SAPS assessments at baseline and 1 or more postbaseline, using least squares mean (LSM) mixed model repeated measures.
ACP-103-012 [conducted at 73 sites in Europe (Bulgaria, France, Ukraine, and United Kingdom), Russia, India, and the United States] randomized 298 participants (1:1:1): placebo, 98 participants; pimavanserin 8.5 mg, 101 participants; and pimavanserin 34 mg, 99 participants. ACP-103-020 [conducted at 54 sites in North America (52 sites in the United States and 2 sites in Canada)] randomized 199 participants (1:1): placebo (94 participants) and pimavanserin 34 mg (105 participants). Demographic and clinical characteristics were comparable between groups in each study at baseline (Table 1). However, because of differences in the inclusion criteria, participants in ACP-103-020 tended to have more severe symptoms of psychosis.
The mean baseline SCOPA-NS score in the ACP-103-020 study was 5.8 for the pimavanserin group (n = 95) and 5.5 for the placebo group (n = 90). Treatment effects expressed as LSM reductions in SCOPA-NS at week 6 were −1.4 for pimavanserin and −0.5 for placebo. Pimavanserin showed numerical improvement in nighttime sleep over placebo based on the SCOPA-NS at each of the 3 time points after the baseline assessment (Fig. 1A), with statistical significance at weeks 4 (P < 0.001) and 6 (P = 0.045).
Nighttime sleep assessments during ACP-103-012 revealed marked improvements with the 34 mg pimavanserin dose, although not with 8.5 mg/d (Fig. 1B). Mean baseline SCOPA-NS scores were 5.4 for the 8.5 mg pimavanserin group (n = 98), 5.5 for the 34 mg pimavanserin group (n = 92), and 5.5 for the placebo group (n = 97). After 6 weeks, the mean changes in SCOPA-NS were −1.4, −2.0, and −1.2 for these groups, respectively. The change in the 34 mg pimavanserin group compared with placebo was not significant at week 6 (P = 0.058), although the separation of this dose from placebo was evident at both week 1 (P = 0.004) and week 2 (P = 0.030).
Pooled ACP-103-020 and ACP-103-012 Data
The results from analyzing the pooled data set were similar to the results observed in the individual trials (Fig. 1C). Mean baseline SCOPA-NS scores were 5.7 for the pimavanserin group (n = 186) and 5.5 for the placebo group (n = 187). Significant improvements in SCOPA-NS scores over placebo were evident at each assessment, especially in patients with observed improvements in psychosis (Table 2). The LSM reduction in SCOPA-NS at week 6 was −1.8 and −1.0 (P = 0.005) for the pimavanserin and placebo groups, respectively.
Pooled ACP-103-020 and ACP-103-012: Participants With Nighttime Sleep Impairment at Baseline
Participants in the pooled data set with impaired nighttime sleep at baseline, defined as SCOPA-NS ≥7, experienced greater improvement in nighttime sleep, as measured by SCOPA-NS, than the overall study population (Fig. 1D). Mean baseline SCOPA-NS scores were 9.8 and 9.6 for the pimavanserin 34 mg (n = 69) and placebo (n = 67) groups, respectively. The LSM reduction in SCOPA-NS at week 6 was −4.4 and −2.6 (P = 0.002) for the pimavanserin and placebo groups, respectively, with pimavanserin showing significant improvement over placebo at each assessment.
For ACP-103-020, mean baseline SCOPA-DS scores were 7.6 for the pimavanserin 34 mg group (n = 95) and 7.3 for placebo (n = 90). At week 6, the LSM change was −2.2 for pimavanserin versus −1.0 for placebo (P = 0.012), representing an improvement in daytime sleepiness when compared with placebo (Fig. 2A).
For ACP-103-012, the SCOPA-DS assessments throughout the 6-week treatment did not show a significant change between placebo and the 2 doses of pimavanserin (Fig. 2B). The SCOPA-DS baseline scores for 8.5 mg/d (n = 98), 34 mg/d (n = 92), and placebo (n = 97), were 6.5, 6.2, and 6.1, respectively. At week 6, decreases from the baseline scores were −1.4, −1.2, and −1.4, respectively.
Pooled ACP-103-020 and ACP-103-012 Data
The participants treated with pimavanserin 34 mg in ACP-103-020 and ACP-103-012 had similar results as in the individual trials (Fig. 2C). Participants in the pooled data set of the 2 trials had baseline SCOPA-DS scores of 6.9 and 6.7 for the pimavanserin 34 mg (n = 186) and placebo (n = 187) groups, respectively. Although the improvement in SCOPA-DS score at each assessment was numerically greater in the pimavanserin group, there was no significant difference throughout the 6 weeks. At week 6, the decrease from baseline for the pimavanserin 34 mg group was −1.7 and − 1.2 for the placebo group (P = 0.108). Improvements in daytime sleepiness scores were greater in patients with improved psychosis; these findings were not statistically significant (Table 2).
Pooled ACP-103-020 and ACP-103-012: Participants With Daytime Sleepiness at Baseline
At each assessment throughout the 6-week treatment period, participants with daytime sleepiness at baseline (defined as SCOPA-DS ≥5) treated with pimavanserin (n = 121) showed a similar change in SCOPA-DS score when compared with placebo (n = 119) (Fig. 2D). Mean baseline SCOPA-DS scores for participants with daytime sleepiness at baseline were 9.4 and 9.1 for the pimavanserin 34 mg and placebo groups, respectively. After 6 weeks of treatment, the pimavanserin group SCOPA-DS score change was −2.5, whereas the placebo group change was −1.7 (P = 0.594).
Pooled ACP-103-020 and ACP-103-012: Participants With Impaired Nighttime Sleep and Daytime Sleepiness
Evaluating the effect of pimavanserin on nighttime sleep in the participants who had both impaired nighttime sleep and increased daytime sleepiness at baseline (defined as SCOPA-NS ≥7 and SCOPA-DS ≥5) in the pooled data set showed greater improvements in nighttime sleep than in the overall study population (Figs. 3A, B). Mean baseline SCOPA-NS scores were 9.8 for the pimavanserin group (n = 47) and 9.5 for the placebo group (n = 51), whereas the mean SCOPA-DS scores were 10.5 for the pimavanserin group and 9.2 for the placebo group. The pimavanserin 34 mg group showed significant improvements over placebo in SCOPA-NS at each assessment point. After 6 weeks of therapy, the SCOPA-NS score change was −4.4 for the pimavanserin group and −2.3 for the placebo group (P = 0.002), whereas the SCOPA-DS change was −2.9 and −1.9 for the pimavanserin 34 mg and placebo groups, respectively (P = 0.120).
The data from the 2 randomized, placebo-controlled clinical trials suggest that subjective sleep reports of nighttime sleep improved with administration of pimavanserin, a novel 5-HT2A receptor inverse agonist/antagonist. Dose-dependent sleep improvements were noted, as no significant improvements arose from 8.5 mg as compared with placebo. In addition, stronger effects were observed in participants who exhibited higher severity of sleep impairment at baseline. Of note, these results for nighttime sleep were evident at the first study time point (at weeks 1 and 2 in ACP-103-012 and ACP-103-020, respectively), and effects were sustained through the 6-week trials. These findings are clinically significant, as there is a subjective improvement in nighttime sleep. Although subjective sleep improved in both groups, there was a greater improvement seen in patients who had improved psychosis. Study ACP-103-020 also suggested that 34 mg of pimavanserin improved daytime sleepiness when compared with placebo, but these findings were not observed in ACP-103-012.
Sleep disorders often stand out as a problem greatly impacting overall quality of life in PD, and sometimes developing soon after onset of motor symptoms. Current therapeutics do not always provide solutions, especially because drugs intended for nighttime sedation to treat insomnia often do not benefit fragmented sleep (awakenings during the course of the night). When longer-acting treatments are used, such as clonazepam, the carryover effect of sedation can have “hangover” effects the next day.24 Another compounding issue is that patients with PD often experience sedation from medications used to treat their motor disorder; hence, remedies for improving sleep quality without increasing daytime sleepiness is an unmet need in PD. The exploratory findings of the 2 randomized, placebo-controlled clinical trials with pimavanserin, as described above, suggest that pimavanserin may be an option to enhance nighttime sleep without increasing daytime sleepiness. Further studies are needed to determine if the mechanisms of improved of psychosis influences are directly linked to sleep.
Slow wave sleep has been associated with improved subjective sleep quality,25 and although these clinical trials did not evaluate sleep staging, significant improvement in subjective sleep quality with active treatment was evident in both investigations. It is important to consider sleep staging when considering nighttime medications, as some drugs, such as benzodiazepines, have been shown to increase stage 2 light sleep and decrease deeper slow wave sleep.26 Future investigations with pimavanserin may help establish its role in improving sleep and increasing slow wave sleep and RBD.
Several limitations to this analysis can be identified. These studies were carried out in a select population of nondemented patients with PD with psychosis; hence, the outcomes are not necessarily generalizable to all patients with PD. In addition, sleep was evaluated with subjective measures rather than objective measures. Similarly, sleep disorders such as obstructive sleep apnea and RBD in study participants were not screened for and so their possible presence might have confounded the findings we report. Specifically, screening for RBD is relevant for this population, as the risk of psychotic symptoms (and nighttime falls) is increased in PD with RBD. Finally, these results represent only short-term responses to the medication; future studies should evaluate long-term response and outcomes. Nonetheless, these trials are promising in future evaluations of pimavanserin for PDP as a treatment to improve both psychotic symptoms and sleep.
1. Parkinson J. An essay on the shaking palsy. 1817. J Neuropsychiatry Clin Neurosci
2002;14(2):223–236; discussion 222.
2. Lees AJ, Blackburn NA, Campbell VL. The nighttime problems of Parkinson's disease. Clin Neuropharmacol
3. Neikrug AB, Maglione JE, Liu L, et al. Effects of sleep disorders on the non-motor symptoms of Parkinson disease. J Clin Sleep Med
4. Karlsen KH, Tandberg E, Arsland D, et al. Health related quality of life in Parkinson's disease: a prospective longitudinal study. J Neurol Neurosurg Psychiatry
5. Brassington GS, King AC, Bliwise DL. Sleep problems as a risk factor for falls in a sample of community-dwelling adults aged 64–99 years. J Am Geriatr Soc
6. Bosboom JL, Wolters ECh. Psychotic symptoms in Parkinson's disease: pathophysiology and management. Expert Opin Drug Saf
7. Lee AH, Weintraub D. Psychosis in Parkinson's disease without dementia: common and comorbid with other non-motor symptoms. Mov Disord
8. Pappert EJ, Goetz CG, Niederman FG, et al. Hallucinations, sleep fragmentation, and altered dream phenomena in Parkinson's disease. Mov Disord
9. Bliwise DL, Watts RL, Watts N, et al. Disruptive nocturnal behavior in Parkinson's disease and Alzheimer's disease. J Geriatr Psychiatry Neurol
10. Albers JA, Chand P, Anch AM. Multifactorial sleep disturbance in Parkinson's disease. Sleep Med
11. Vanover KE, Weiner DM, Makhay M, et al. Pharmacological and behavioral profile of N
'-(4-(2-methylpropyloxy)phenylmethyl) carbamide (2R
)-dihydroxybutanedioate (2:1) (ACP-103), a novel 5-hydroxytryptamine(2A) receptor inverse agonist. J Pharmacol Exp Ther
12. Monti JM. Serotonin control of sleep-wake behavior. Sleep Med Rev
13. Vanover KE, Davis RE. Role of 5-HT2A receptor antagonists in the treatment of insomnia
. Nat Sci Sleep
14. Ancoli-Israel S, Vanover KE, Weiner DM, et al. Pimavanserin tartrate, a 5-HT(2A) receptor inverse agonist, increases slow wave sleep as measured by polysomnography in healthy adult volunteers. Sleep Med
15. Stutzmann JM, Eon B, Lucas M, et al. RP 62203, a 5-hydroxytryptamine2 antagonist, enhances deep NREM sleep in rats. Sleep
16. Cummings J, Isaacson S, Mills R, et al. Pimavanserin for patients with Parkinson's disease psychosis: a randomised, placebo-controlled phase 3 trial. Lancet
17. Cummings JL, Mega M, Gray K, et al. The neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology
18. Andreasen NC. Scale for the Assessment of Positive Symptoms (SAPS)
. Iowa City, Iowa: University of Iowa, College of Medicine; 1984.
19. Folstein MF, Folstein SE, McHugh PR. Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res
20. Voss T, Bahr D, Cummings J, et al. Performance of a shortened Scale for Assessment of Positive Symptoms for Parkinson's disease psychosis. Parkinsonism Relat Disord
21. Marinus J, Visser M, van Hilten JJ, et al. Assessment of sleep and sleepiness in Parkinson disease. Sleep
22. Buysse DJ, Reynolds CF 3rd, Monk TH, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res
23. Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep
24. Anderson KN, Shneerson JM. Drug treatment of REM sleep behavior disorder: the use of drug therapies other than clonazepam. J Clin Sleep Med
25. Akerstedt T, Kecklund G, Gillberg M. Sleep and sleepiness in relation to stress and displaced work hours. Physiol Behav
26. Poyares D, Guilleminault C, Ohayon MM, et al. Chronic benzodiazepine usage and withdrawal in insomnia
patients. J Psychiatr Res