Walsh, Christine M.

Abstract Biomarkers that aid in early detection of neurodegeneration are needed to enable early symptomatic treatment and enable identification of people who may benefit from neuroprotective interventions. Increasing evidence suggests that sleep biomarkers may be useful, given the bi-directional relationship between sleep and neurodegeneration and the prominence of sleep disturbances and altered sleep architectural characteristics in several neurodegenerative disorders. This study aimed to demonstrate that sleep can accurately characterize specific neurodegenerative disorders (NDD). A four-class machine-learning algorithm was trained using age and nine sleep biomarkers from patients with clinically-diagnosed manifest and prodromal NDDs, including Alzheimer’s disease dementia (AD = 27), Lewy body dementia (LBD = 18), and isolated REM sleep behavior disorder (iRBD = 15), as well as a control group (CG = 58). The algorithm was validated in a total of 381 recordings, which included the training data set plus an additional AD = 10, iRBD = 18, Parkinson disease without dementia (PD = 29), mild cognitive impairment (MCI = 78) and CG = 128. Test-retest consistency was then assessed in LBD = 10, AD = 9, and CG = 46. The agreement between the NDD profiles and their respective clinical diagnoses exceeded 75% for the AD, LBD, and CG, and improved when NDD participants classified Likely Normal with NDD indications consistent with their clinical diagnosis were considered. Profiles for iRBD, PD and MCI participants were consistent with the heterogeneity of disease severities, with the majority of overt disagreements explained by normal sleep characterization in 27% of iRBD, 21% of PD, and 26% of MCI participants. For test-retest assignments, the same or similar NDD profiles were obtained for 88% of LBD, 86% in AD, and 98% of CG participants. The potential utility for NDD subtyping based on sleep biomarkers demonstrates promise and requires further prospective development and validation in larger NDD cohorts.

Abstract Introduction Non-REM hypertonia (NRH) was recently reported to be independently associated with the synucleinopathy-mediated neurodegenerative disorders: dementia with Lewy Bodies/Parkinson Disease Dementia (DLB/PDD), Parkinson Disease (PD), and isolated REM sleep behavior disorder (iRBD). In this NRH investigation, we included progressive supranuclear palsy (PSP), a Parkinsonian-spectrum disorder caused by tau pathology. Methods In this multicenter study, patients broadly characterized as presumed Parkinsonian-spectrum disorders (PSD) included DLB/PDD (n=16), PD (n=16), iRBD (n=19), and PSP (n=13). Presumed non-PSD subjects included Alzheimer’s Disease dementia (AD=22), mild cognitive impairment (MCI=35), and normal cognition (NC=61). Sleep Profiler studies were acquired in all participants. NRH, auto-detected based on persistently elevated electromyographic (EMG) power relative to delta, theta, and sigma bands in the differential Af7-Af8 signal, was measured as a percentage of sleep time and then weight-averaged in the 75% of in-home studies with two-nights of data. A >5% threshold characterized abnormal-NRH. Twenty-nine NC were longitudinally retested after 364- to 563-days. Statistical analyses included inter-class correlations (ICC), Bland-Altman plots, multiple logistic regression, and receiver-operating-characteristic curves (ROC). Results In the PSD=41 and non-PSD=95 records with two-nights of data, NRH-severity demonstrated moderate consistency (ICC=0.78, bias=0.6+/-6.2%, P<0.0001). Across the two-nights, NRH was classified consistently as normal or abnormal in 59.6% and 27.2% of the records, vs. normal/abnormal=4.4% or abnormal/normal=8.8%. The test-retest reliability of NRH-severity was good (ICC=0.84, bias=0.06+/-3.8%, P<0.0001), with all retest comparisons repeating as normal (73%) or abnormal (27%). The frequency of abnormal-NRH in PSP=92% was significantly greater than MCI=26%, AD=14%, and NC=16% (all P<0.0001) and PD=56% (P<0.05), but not DLB/PDD=81% and iRBD=74%. Abnormal-NRH was significantly associated with the PSD group (P<0.0001) and it differentiated PSD versus non-PSD group with an area under the curve of 0.78 (95%CI: 0.72-0.85) based on a sensitivity of 0.75 (95%CI: 0.63-0.84) and a specificity of 0.81 (85%CI: 0.73-0.87). Conclusion NRH independently discriminated PSD patients from age-sex similar non-PSD subjects, suggesting that NRH is a common sleep motor signature across clinical PSD phenotypes. We speculate that NRH could be related to pathological changes within the key non-REM sleep motor modulating center in synucleinopathies and PSP. Support (If Any) NIA grants: R44AG050326, R44AG054256, P30AG62677 and R34AG56639.

Abstract Introduction Increased non-REM hypertonia (NRH) and decreased sleep spindle-durations were found to be independently associated with patients broadly characterized as presumed Parkinsonian-spectrum disorder [i.e., dementia with Lewy Bodies/Parkinson Disease Dementia (DLB/PDD), Parkinson Disease (PD), progressive supranuclear palsy (PSP), or isolated REM sleep behavior disorder (iRBD)] when compared to subjects with Alzheimer Disease dementia (AD), mild cognitive impairment (MCI) and normal cognition (NC). In this investigation, the NRH and spindle-duration features were combined to determine whether these biomarkers could distinguish neurodegenerative disorder subtypes. Methods This multicenter investigation included analysis of several neurodegenerative disorder patient subtypes including: DLB/PDD (n=16), PD (n=16), iRBD (n=19), PSP (n=13), AD (n=22), MCI (n=35), and NC (n=61). Sleep Profiler (SP) studies were conducted in all participants. NRH was auto-detected based on persistently elevated electromyographic (EMG) power relative to delta, theta, and sigma bands, while spindles were recognized by temporal excursions in alpha and sigma power > 250-milliseconds. With NRH >5% of sleep-time and spindle-duration <1-minute considered abnormal, tallies were compared across neurodegenerative disorder subtypes with Fisher Exact tests. Results Combined SP features of Normal-NRH/normal-SpD were greater in the NC (56%), AD (46%), and MCI (43%) subtypes versus PSP (8%) and DLB/PDD (6%), and when iRBD (21%) was compared to NC (all P<0.05). Abnormal-NRH/abnormal-SpD were more frequent in PSP (85%) and DLB/PDD (75%) subtypes versus iRBD (26%), PD (25%), MCI (11%), AD (9%), and NC (8%)(all P<0.02). Abnormal-NRH/normal-SpD were greater in iRBD (47%) compared to MCI (14%), NC (8%), PSP (8%), DLB/PDD (6%), and AD (5%) subtypes, and in PD (31%) versus NC (all P<0.05). Normal-NRH/abnormal-SpD occurred more often in AD (41%) than iRBD (5%) and PSP (0%)(both P<0.05). Conclusion The combination of NRH and spindle-duration abnormality occurred more frequently in DLB/PDD and PSP subtypes. Abnormal NRH with normal sleep-duration was more frequent in iRBD and PD subtypes. Normal NRH and spindle-duration occurred most often in NC and MCI types. AD exhibited normal NRH with isolated cases of either normal or abnormal spindle-duration. Our preliminary findings suggest that auto-detected sleep biomarkers may aid in the characterization of neurodegenerative disorder subtypes. Larger prospective cohort studies are needed. Support (If Any) NIA grants: R44AG050326, R44AG054256, P30AG62677 and R34AG56639.

Abstract Introduction Decreased sleep spindle oscillations were previously associated with cognitive decline in older adults, increased tau levels, and phenoconversion to dementia in patients with Parkinson disease (PD). We analyzed quantitative sleep spindle measures to determine if this biomarker was associated with particular neurodegenerative disorder syndromes. Methods Sleep spindle oscillations ascertained in patients broadly characterized as presumed Parkinsonian-spectrum disorders (PSD), which included the subtypes dementia with Lewy Bodies/Parkinson Disease Dementia (DLB/PDD, n=16), PD (n=16), isolated REM sleep behavior disorder (iRBD, n=19), and progressive supranuclear palsy (PSP, n=13), were compared with non-PSD subtypes Alzheimer Disease dementia (AD, n=22), mild cognitive impairment (MCI, n=35), and normal cognition (NC, n=61). Sleep Profiler studies were conducted in all participants. The automated spindle detection algorithms recognized temporal excursions in the alpha (8-12 Hz) and sigma (12-16 Hz) power of 250 milliseconds or greater, with spindle duration being the sum of all spindle elapsed times. Night-to-night variability was assessed in PSP=13, PD=16, DLB/PDD=12, AD=17, MCI=25, and NC=53. Statistical analyses included intraclass correlations (ICC) and Bland-Altman plots for two-night data, and Mann-Whitney U-tests and multiple logistic regression applied to sleep-time weight-averaged spindle-durations. Results The night-to-night spindle-duration ICC was 0.95 (P<0.0001), with a Bland-Altman bias of 0.05+/-2.83 minutes. Spindle-duration was independently associated with PSD versus non-PSD groups (P=0.017, OR 1.08, 95%-CI 1.01-1.15), but not significantly associated with age (P=0.12, OR 1.03, 95%-CI 0.99-1.07) or sex (P=0.54). When stratified by subtype, age was associated with spindle-duration when NC were compared to AD and MCI (P=0.0003, OR 1.10, 95%-CI 1.04-1.16) and when iRBD were compared to DLB/PDD, PD and PSP (P<0.05, OR 1.00, 95%-CI 0.89-1.13)Spindle-durations were reduced in PSP (0.9+/-2.1) and DLB/PDD (2.0+/-5.1) when individually compared to AD (3.2+/-7.1), iRBD (3.3+/-3.4), PD (5.3+/-6.6), MCI (5.3+/-9.7), and NC (8.0+/-11.1) subtypes (all P<0.05). AD patients also exhibited lower spindle-durations than NC (P=0.03). Conclusion Auto-detected sleep spindle-durations exhibited excellent night-to-night reliability in both NC and patients with neurologic disorders. Decreased sleep spindle-duration was independently associated with PSP and DLB/PDD, and in AD. Reduced sleep spindle duration may be a distinct sleep biomarker for those disorders likely indicating thalamocortical dysfunction. Support (If Any) NIA grants: R44AG050326, R44AG054256, P30AG62677 and R34AG56639.

Abstract Introduction The severity of REM sleep without atonia, a prodromal biomarker for synucleinopathy-related neurodegenerative disorders, is influenced by selective serotonin reuptake inhibitor (SSRI) or serotonin-norepinephrine retake inhibitor (SNRI) use. This study investigates whether SSRI/SNRIs similarly impacts the severity of Non-REM hypertonia (NRH), a biomarker independently associated with patients broadly characterized with presumed Parkinsonian-spectrum disorders (PSD). Methods In this multi-center study, the relationship between NRH and SSRI/SNRIs was evaluated in PSD patients [Lewy Body/Parkinson Disease Dementia (DLB/PDD=16), Parkinson Disease (PD=16), isolated REM sleep behavior disorder (iRBD=19), and progressive supranuclear palsy (PSP=12)], and non-PSD subjects [Alzheimer Disease (AD=22), mild cognitive impairment (MCI=35), and normal cognition (NC=61)]. Studies were conducted with the Sleep Profiler in all participants. NRH was auto-detected based on persistently elevated electromyographic (EMG) power relative to delta, theta, and sigma bands. Abnormal-NRH was based on a threshold of >5% of sleep time, and weight-averaged by sleep time, in the 75% of studies with two-nights of data. Statistical analyses included multiple logistic regression, Fisher Exact, and Mann-Whitney U tests. Results Across all 181 records, 38% had abnormal-NRH and 24% used SSRI/SNRIs (P<0.005). No differences were observed in the distributions of NRH for those with NRH-only vs. those with combined NRH and SSRI/SNRI use (n=45, 14.5+/-7.7% versus n=24, 18.6+/-11.6%, respectively; P=0.19). Abnormal-NRH was associated with the PSD versus non-PSD groups (P<0.0001, odds-ratio=14.0) and SSRI/SNRI use (P<0.05, odds-ratio=2.4), but not age and sex. Within the PSD and non-PSD groups, the frequency of abnormal-NRH was 75% vs. 19% (P<0.0001), and SSRI/SNRI use was 32% vs. 19% (P=0.07), respectively. The distributions of abnormal-NRH in the PSD and non-PSD subgroups ranged from PSP=92%, DLB/PDD=81%, PD=56% ,and RBD=74% versus MCI=26%, and NC=16%, AD=14%, respectively. By comparison, distributions of SSRI/SNRI use were DLB/PDD=50%, RBD=42%, PSP=33%, and PD=0% in the PSD subgroups, compared to MCI=34%, AD=23%, and NC=10% in the non-PSD subgroups. Conclusion Abnormal non-REM hypertonia exhibited a very strong association with Parkinsonian-spectrum disorder diagnoses (odds-ratio=14.0) and a relatively weaker association with selective serotonin reuptake inhibitor or serotonin-norepinephrine retake inhibitor use (odds-ratio=2.4), suggesting a finding that reflects an interactive, rather than causal inference. Support (If Any) NIA grants R44AG050326, R44AG054256, P30AG62677 and R34AG56639.