Welcome to the AD/PD™ 2024 Interactive Program
The conference will officially run on Western European Standard Time (Lisbon, UTC+0)
To convert the conference times to your local time Click Here
REGIONAL ASSOCIATIONS OF SLEEP ARCHITECTURE AND ALZHEIMER'S DISEASE PATHOLOGY
Abstract
Aims
Recent evidence suggests that disturbances of sleep architecture are linked to Alzheimer’s disease (AD) pathology. Here, we assessed the association between sleep architecture and regional amyloid and tau pathology employing a sleep-monitoring device in addition to PET imaging.
Methods
Data of sixteen individuals with confirmed AD-biomarker positivity were included in the current analysis (M(Age)= 65.81 (7.23), Sex(M/F)=8/8, M(MMSE)= 27.7 (2.21)). All subjects underwent amyloid ([11C]-PiB) and tau ([18F]-AV1451) PET imaging. PET images were normalized and intensity standardized to the whole cerebellum ([11C]-PiB) or the inferior cerebellum ([18F]-AV1451). All subjects were provided with a portable sleep-monitoring headband by Beacon Biosignal (former Dreem), which has been shown to reliably detect the different sleep phases similar to polysomnography. Participants were asked to wear the device for at least three consecutive nights within six months of the PET acquisition. Total duration of sleep phases per minutes (i.e. REM, N1, N2, N3) were extracted for the recordings. Nightly measurements were then averaged after confirmation of their stability across the three nights. Next, whole-brain voxel-wise correlation analyses were performed between the sleep measures (i.e. mean duration in respective sleep phase) and amyloid and tau load as assessed by PET imaging, respectively. Analyses were corrected for age. Significance threshold was set at a p-value of p<.001 (uncorrected).
Results
Amyloid burden in the insula, prefrontal cortex and precuneus was linked to changes in nightly N2 and N3 duration, whereas tau burden in the medio-temporal, superior parietal and precentral gyrus was predominantly associated with nightly N3 and REM duration.
Conclusions
Local changes in sleep architecture may arise from regionally-specific accumulation patterns of AD pathologies. Yet, it remains unknown whether disruptions in sleep architecture are the cause or the consequence of pathology build-up.
POOR SLEEP EFFICIENCY IS LINKED TO ELEVATED BLOOD-BASED BIOMARKERS OF ALZHEIMER’S DISEASE AND NEURODEGENERATION IN OLDER ADULTS LIVING WITH SUBJECTIVE COGNITIVE DECLINE
Abstract
Aims
Research indicates that there is a powerful bi-directional relationship between poor sleep and neurodegeneration, particularly in Alzheimer’s disease. Recent breakthroughs in the use of blood-based biomarkers (BBBs), demonstrating an extremely high sensitivity and specificity to neurodegenerative pathology, may serve as the conduit combining research efforts into sleep medicine and dementia. In this study, we examined the cross-sectional inter-relationship between blood-plasma levels of amyloid-beta 42 and 40 (Aβ42, Aβ40), glial acidic fibrillary protein (GFAP) and neurofilament light chain (NfL) as well as sleep efficiency from polysomnography in older adults ‘at risk’ of dementia.
Methods
157 older adults (mean age=69.2 years) presenting with cognitive concerns were assessed at the Healthy Brain Ageing Clinic, meeting criteria for mild cognitive impairment (MCI, 43%), subjective cognitive decline (SCD, 41%), as well as dementia (6%) and no cognitive decline (10%). All participants provided a same day fasting blood sample and underwent overnight polysomnography within two weeks, from which sleep efficiency was derived. Blood-plasma Aβ42, Aβ40, NfL and GFAP levels were measured with the Quanterix Simoa Human N4PE assay.
Results
Sleep efficiency ranged from 16.5% to 96.6% (group mean=72.7%). In the SCD subgroup reduced sleep efficiency was significantly and moderately associated (Spearman’s) with increased plasma levels of Aβ42 (ρ =-.30, p=0.02), Aβ40 (ρ =-.31, p=0.01), GFAP (ρ =-.32, p=0.01) and NfL (ρ =-.37, p=0.003), findings not present in the MCI subgroup.
Conclusions
Our preliminary results support the notion that poorer sleep, even as assessed by sleep efficiency, is associated with BBBs of neurodegeneration in a subsample of older adults considered ‘at risk’ of dementia. Further research is now warranted to examine the inter-relationships of BBBs and various sleep microarchitecture measures within and between ‘at risk’ subgroups, both cross-sectionally and longitudinally.
CORTICAL RHYTHMS REFLECT DIFFERENTIAL DEGENERATION OF BRAINSTEM AND BASAL FOREBRAIN NUCLEI IN PARKINSON’S DISEASE
Abstract
Aims
Patients with Parkinson's disease (PD) exhibit changes in cortical rhythms across several frequency bands, which scale with motor and cognitive impairments. Whether and how these neurophysiological-clinical relationships can be attributed to the hallmark degenerations of the noradrenergic locus coeruleus (LC), dopaminergic substantia nigra (SN), and acetylcholinergic nucleus basalis of Meynert (nbM) is unknown.
Methods
We combined magnetoencephalographic neuroimaging (N = 79), neuromelanin-sensitive MRI of the SN and LC (N = 58), and diffusion weighted imaging of the nbM (N = 68) in patients with idiopathic PD. We tested for relationships between frequency-defined cortical rhythms and integrity of the SN, LC, and nbM using nonparametric general linear modeling across the whole cortex. We investigated the neurochemical specificity of the resulting relationships using neurochemical colocalization analysis with autocorrelation-preserving null models via the neuromaps multi-atlas.
Results
We show that alpha-band deviations in PD are related to degeneration of the LC, while beta-band deviations are related to reduced SN and nbM integrity. Additionally, slowing of the arrhythmic component of cortical activity in somato-motor regions is related to nbM degeneration. Intuitively, the alpha-LC and beta-nBM relationships are stronger in cortical regions with high densities of norepinephrine and vesicular acetylcholine transporters, respectively. These effects are also relevant for clinical features: the alpha-LC relationship for attentional impairments, the beta-SN for axial motor symptoms, and the beta-nbM for visuospatial impairments.
Conclusions
Together, these findings suggest specific neurochemical origins of commonly-reported neurophysiological changes in PD, with implications for clinical outcomes.
PROGRESSIVE ALTERATION IN SLEEP-ACTIVITY RHYTHMS AND CLOCK GENE EXPRESSION ACROSS IN REM SLEEP BEHAVIOR DISORDER, PARKINSON'S DISEASE AND DEMENTIA WITH LEWY BODIES
Abstract
Aims
Parkinson's disease (PD) and Dementia with Lewy bodies (DLB) are strongly associated with sleep disturbances, including Rapid Eye Movement Sleep Behaviour Disorder (RBD) which can precede diagnosis by several years. The aim of the present study was to investigate sleep/wake cycles and circadian rhythms in PD and DLB patients in the early stages of their disease, along with ‘at risk’ subjects diagnosed with iRBD.
Methods
15 healthy controls, 20 iRBD, 16 PD and 17 DLB patients with a disease duration of less than 5 years underwent clinical assessment. Sleep/wake cycles were evaluated using actigraphy. Salivary and oral mucosa samples were collected every 3 hours for 24 hours to measure melatonin levels and Bmal1 clock gene expression.
Results
DLB patients were characterised by greater daytime somnolence, disrupted night-time sleep, less activity during the daytime but more activity overnight. Amplitude and mesor of the actigraphic rest-activity profiles declined significantly with progression from HC to iRBD, PD and DLB patients(p<0.001) and correlated inversely with motor parkinsonism (r=-0.38, p=0.024; r=-0.46,p=0.006 respectively). Severe disruption in melatonin rhythms were found in the DLB group with no significant change in relation to parameters of secretion across the remaining groups. There was a significant pattern of decreasing median Bmal1 amplitude from controls, to iRBD, to PD and then to DLB p = 0.037). No trend was seen for the Bmal1 mesor or acrophase across groups.
Conclusions
This work highlights a differential gradient of objective disruption in the daily circadian rhythms from prodromal cohorts to established PD and DLB and is the first to demonstrate disruption of clock gene expression in DLB thus support sleep/wake disruption as a potential biomarker.
POORER OBJECTIVE SLEEP QUALITY METRICS ARE ASSOCIATED WITH LOWER CEREBRAL STRUCTURAL INTEGRITY INDEPENDENT OF AD PATHOLOGY BIOMARKERS
Abstract
Aims
Sleep disturbances are prevalent in Alzheimer’s disease (AD), with sleep quality being already impaired in the preclinical stages. Specifically, poor sleep quality has been linked to reductions in cerebral volume and cortical thickness (CTh). Less is known about how sleep is related to brain structural integrity while considering AD pathology. We aimed to investigate the associations between actigraphy-measured sleep and regional CTh in cognitively unimpaired (CU) adults at risk of AD.
Methods
We included 174 CU adults from the ALFASleep study (Table 1). Sleep quality was measured using actigraphy (Actiwatch2®, Philips Respironics) for two weeks. The CTh of the “AD signature” composite region of interest (ROI) was calculated using FreeSurfer v6.0. We performed separate general linear models where the CTh was the dependent variable, while the following actigraphy-derived sleep parameters were the predictors of interest: total sleep time, sleep efficiency (SE), sleep latency, wake after sleep onset (WASO) and sleep fragmentation index (SFI). All models were corrected for age, sex, APOE-ε4 status and the time difference between magnetic resonance imaging and actigraphy acquisition. Final models were corrected by CSF AD biomarkers (A +: Aβ42/40<0.071, T+: p-tau-181>24 pg/mL).
Results
Lower SE, higher WASO and SFI (indicating poorer sleep) were associated with lower AD signature thickness (p=0.040; p=0.013; p=0.008), respectively. These associations remained significant after adjustment for CSF amyloid and tau status.
Conclusions
In CU older adults, higher sleep fragmentation metrics and lower sleep efficiency are associated with reduced CTh in areas vulnerable to early AD-related neurodegeneration. Importantly, these effects were independent of core AD pathology biomarkers, suggesting that disrupted sleep may lead to decreased cortical integrity via amyloid and tau-independent pathways.
DECODING BRAIN AGE: PREDICTING COGNITIVE DECLINE WITH BRAIN NETWORK ANALYTICS (BNA)
Abstract
Aims
This study investigates the utility of the Brain Network Analytics (BNA) tool, an automatic analysis tool for EEG data, in detecting cognitive decline. Based on a a new analysis of BNA’s ERP and EEG normative database we seek to validate a BNA-based brain age prediction model and explore whether disparities between biological and chronological age are indicative of heightened risks for future neurological conditions, including cognitive decline, mild cognitive impairment (MCI), or Alzheimer's disease (AD).
Methods
We developed a robust machine learning model utilizing a longitudinal, normative EEG dataset sourced from The Villages Health centers (FL, USA), This dataset included 871 recordings collected from 277 men aged between 55 and 85. Our model was designed to predict age based on BNA scores. BNA involves cloud-based EEG data analysis aimed at characterizing and comprehending the activity patterns within the brain's networks. This is achieved by comparing an individual's brain patterns with those of an age-matched normative database. We evaluated the model's performance using k-fold cross-validation and a separate test set, using R² and Root Mean Square Error (RMSE) metrics. To assess the potential for early prediction of cognitive decline, we included EEG data from individuals with Mild Cognitive Impairment (MCI; N=25) and early-stage Alzheimer's disease (EAD; N=13) who were monitored over a year. We’ve compared the differences of deviations relative to chronological age between the differenc clinical group.
Results
The model successfully predicted age with a RMSE of 6.9 years on 8-fold cross-validation test sets. A strong correlation was found between actual and predicted age in a held-out validation dataset consisting of 225 recordings from 70 helathy subject (r=0.6, phealthy p=0.001, EAD>MCI p=0.005). Subjects with MCI showed significantly larger brain age difference relative to healthy (mean difference: + 3.7 years, p=0.001). Additionally, brain age differences were negatively correlated with the years of education (r=-0.3, p<0.0001).
Conclusions
The results underscore the efficacy of a BNA-based model in predicting brain age and its association woth cognitive decline. The BNA-based brain age prediction model successfully identified disparities between biological and chronological age, showing promise in identifying individuals at heightened risk for neurological conditions such as mild cognitive impairment (MCI) or Alzheimer's disease (AD). These findings may also have implications for monitoring disease progression, guiding treatment interventions, and pinpointing patients at high risk of deterioration for inclusion in clinical trials.
NEW TECHNOLOGIES BASED ON VIRTUAL REALITY AND TELEMEDICINE FOR COGNITIVE REHABILITATION IN ALZHEIMER’S DISEASE
Abstract
Aims
The objective of the present study was to evaluate the "usability" of two remote cognitive rehabilitation training methods, based on virtual reality, carried out with tablet devices or Apps, in patients with Alzheimer's Disease (AD).
Methods
AD patients with mild to severe impairment (MMSE 13-24) and their caregiver were recruited from the 14 participating IRCCS Centers. Each patient was assigned a remote cognitive rehabilitation treatment group, in tablet or app mode. Treatments lasted 45 minutes per day, 5 days a week, for two months. Patients were assessed pre- and post-treatment with: MMSE, Quality of Life in Alzheimer's Disease (QoL-AD), Geriatric Depression Scale (GDS). Beck Depression Inventory-II (BDI) and System Usability Scale (SUS) were administered to caregivers.
Results
48 patients (mean age 72.8±29.1y; 21 Female; mean education 12±5.4y, mean pre-treatment MMSE: 20.9±10.1) and one of their caregivers were enrolled. 24 patients had medium-severe AD (13≤MMSE≤17) and 24 mild-moderate AD (18≤MMSE≤24). 21 patients were assigned a tablet, 27 patients the App. Post telerehabilitation, MMSE values were stable in each group, without significant differences between groups (p>0.05). Caregivers rated both technologies satisfactorily and similarly (average SUS tablet score: 67.3±25.6; average SUS App score: 65.6±31.9).
Conclusions
These preliminary analyzes show an equivalent level of "usability" between tablet and App modes in cognitive telerehabilitation in patients suffering from mild to severe AD. Such results would open new possibilities in telerehabilitation, providing patients with accessible tools to undergo home-based rehabilitation.
EFFECTS OF A VIRTUAL REALITY-BASED INTERVENTION ON MENTAL WELLBEING IN INFORMAL CAREGIVERS OF PEOPLE WITH DEMENTIA: PRELIMINARY RESULTS
Abstract
Aims
Psychoeducational interventions are recognised to reduce informal caregiver (iCGs) distress. This study aims to investigate the feasibility of integrating an eHealth psychoeducation program with a virtual reality (VR)-based training of the cognitive empathy, and to evaluate its effect on iCGs’ mental health.
Methods
In an ongoing randomised clinical trial, 19 iCGs (mean age = 55; female: 79%; children: 68%) of mild-to-moderate patients with Alzheimer’s disease participated in the 6-week experimental intervention. Each online 2-hour session consisted of a psychoeducational lecture and exposure to a 7-minute VR video that allowed first-hand experience of AD symptoms. Remote VR experiences were made possible using cardboards and smartphones. Following each VR session, participants completed the Extended Reality Presence Scale (XRPS). Before and after the intervention, they completed the Zarit Burden Interview (ZBI), State-Trait Anxiety Inventory (STAI-Y1-2), and Interpersonal Reactivity Index (IRI). Descriptive statistics were presented for levels of presence in immersive videos (XRPS), and pre-post differences were tested with paired t-tests.
Results
Caregivers demonstrated good levels of presence during the VR experience (M= 2.6; SD= 1.08), with high score on videos related to the AD disclosure (M= 2.99; SD=0.73). The intervention decreased levels of state anxiety (STAI-Y: t(18) = 3.003, p = 0.008, Cohen’s d = 0.689) and the tendency to experience distress and discomfort in response to extreme distress in others (IRI-PS: t(18) = 2.469, p = 0.024, Cohen’s d = 0.566). Data showed a trend for a decrease in caregiver burden (ZBI: t(18) = 2.059, p = 0.054, Cohen’s d = 0.472).
Conclusions
These preliminary results support the feasibility and effectiveness of the integrated intervention in improving cognitive empathy and decreasing distress in iCGs. These results should be confirmed in larger samples.