Search

INTRODUCTION: Metabotropic glutamate receptor 5 (mGluR5) is involved in regulating integrative brain function and synaptic transmission. Aberrant mGluR5 signaling and relevant synaptic failure play a key role in the pathophysiological mechanism of Alzheimer's disease (AD). METHODS: Ten cognitively impaired (CI) individuals and 10 healthy controls (HCs) underwent [18F]SynVesT‐1 and [18F]PSS232 positron emission tomography (PET)/magnetic resonance to assess synaptic density and mGluR5 availability. The associations between mGluR5 availability and synaptic density were examined. A mediation analysis was performed to investigate the possible mediating effects of mGluR5 availability and synaptic loss on the relationship between amyloid deposition and cognition. RESULTS: CI patients exhibited lower mGluR5 availability and synaptic density in the medial temporal lobe than HCs. Regional synaptic density was closely associated with regional mGluR5 availability. mGluR5 availability and synaptic loss partially mediated the relationship between amyloid deposition and cognition. CONCLUSIONS: Reductions in mGluR5 availability and synaptic density exhibit similar spatial patterns in AD and are closely linked. Highlights: Cognitively impaired patients exhibited lower mGluR5 availability and synaptic density in the medial temporal lobe than HCs.Reductions in mGluR5 availability and synaptic density exhibit similar spatial patterns in AD.Regional synaptic density was closely associated with regional mGluR5 availability.mGluR5 availability and synaptic loss partially mediated the relationship between amyloid deposition and global cognition.With further research, modulating mGluR5 availability might be a potential therapeutic strategy for improving synaptic function in AD. [ABSTRACT FROM AUTHOR]

Background: Prognostic and monitoring capabilities of blood‐based biomarkers for early cognitive decline, amyloid‐β (Aβ) accumulation and grey matter (GM) loss in the asymptomatic phase of Alzheimer disease require further investigation over extended time periods. This longitudinal study investigated serum glial fibrillary acidic protein (GFAP), neurofilament light (NfL) and Aβ1‐42/Aβ1‐40 as potential prognostic and monitoring biomarkers in cognitively unimpaired (CU) elderly. Method: We included 184 CU older adults (Table 1) who underwent baseline serum sampling and 2‐yearly neuropsychological assessment for up to 10 years (median: 6 years, Figure 1). A subset (N = 109) underwent serial blood sampling, Aβ‐PET and structural MRI at 2 (N = 73), 3 (N = 34), or 4 (N = 2) time points (time interval: 2‐10 years, median: 6 years). Cognition was evaluated in memory, language and executive functioning domains through conversion of neuropsychological tests scores to standardised z‐scores and subsequent averaging per domain. Aβ‐accumulation and GM loss were analysed in a voxelwise manner using Voxelstats with cluster‐level pRFT<.05 and voxel‐level puncorrected<.001 significance thresholds.[1] Aβ1‐42/Aβ1‐40 was inverted to facilitate biomarker comparison. Linear‐mixed effects models were used to (i) identify the predictive value of serum biomarkers for longitudinal cognitive decline, Aβ‐accumulation and GM loss, (ii) calculate annual serum biomarker rates of change and (iii) investigate associations between longitudinal biomarker changes and longitudinal changes in cognition, Aβ‐PET and GM volume. Result: High baseline GFAP and NfL predicted a longitudinal decline in memory (βGFAP*time = ‐0.020, pFDR = 0.03; βNfL*time = ‐0.031, pFDR = 0.002) and language (βGFAP*time = ‐0.025, pFDR<0.001, βNfL*time = ‐0.017, pFDR = 0.04), and longitudinal GM loss within the hippocampus. Low Aβ1‐42/Aβ1‐40 predicted memory decline (βAβ1‐42/Aβ1‐40*time = ‐0.022, pFDR = 0.03) as well as Aβ accumulation within Aβ‐vulnerable regions (Figure 1). Serum GFAP (βtime = 0.022, pFDR<.001) and NfL (βtime = 0.040, pFDR<.001) increased over time. Longitudinal serum NfL increases were associated with a concomitant decline in memory (β∆NfL*time = ‐0.031, pFDR=.004), language (β∆NfL*time = ‐0.022, pFDR=.01) and GM volume (Figure 2). Serum GFAP increases were associated with precuneal Aβ accumulation. Aβ1‐42/Aβ1‐40 decreases were stronger in Aβ‐ subjects (pFDR<.001) and did not associate with Aβ accumulation, nor any other longitudinal marker. Conclusion: Serum GFAP, NfL and Aβ1‐42/Aβ1‐40 are prognostic and/or monitoring biomarkers providing complementary information by reflecting different Aβ‐dependent or ‐independent pathophysiological processes. REFERENCE [1] Mathotaarachchi et al. Front Neuroinform 2016. [ABSTRACT FROM AUTHOR]

Background: The muscarinic acetylcholine receptor 4 (M4) has attracted attention as a target for treating neuropsychiatric symptoms, such as agitation and aggression, associated with Alzheimer's disease. We have developed novel selective, potent, and in vivo suitable M4 positive allosteric modulators (PAMs) and demonstrated robust pharmacology in preclinical models predictive of antipsychotic and pro‐cognitive effects. Importantly, we have also shown that these selective M4 PAMs produce greater therapeutic margins for cholinergic adverse effects compared to less selective muscarinic receptor agonists. Methods: To enable translation in the clinic, we set out to discover an M4 PAM PET tracer suitable for measuring central receptor occupancy (RO). Result: [11C]MK‐6884 binds with high affinity and selectivity to the allosteric site on the M4 receptor that our other M4 PAMs bind. The affinity of [11C]MK‐6884 was found to be influenced by the concentration of orthosteric agonist in vitro and in vivo, consistent with the high cooperativity of the molecule in vitro. PET studies successfully determined the target engagement of PAMs to enable modeling the relationship between drug plasma exposure, RO, and pharmacodynamic activity. Donepezil increased the tracer binding potential of [11C]MK‐6884 in rhesus and healthy human subjects, confirming in vivo the influence of cholinergic tone on [11C]MK‐6884 binding that was observed in vitro. Consistent with a presumed cortical deficit in synaptic acetylcholine, [11C]MK‐6884 PET studies in AD patients showed lower cortical binding potential compared to healthy elderly participants, as well as notable cortical heterogeneity and bilateral asymmetry of M4 receptor binding in AD patients. Striatal binding potential was unchanged in these different populations. Conclusion: Our findings demonstrate that the M4 PAM PET tracer [11C]MK‐6884 is a useful tool for determining target engagement of M4 PAMs, as well serving as a unique probe to study changes in cholinergic neurotransmission in neurodegenerative disease. The Enigma Biomedical Group has an exclusive license agreement with Merck & Co., Inc., for the global development and commercialization of MK‐6884. This allows Enigma and its partner, Cerveau Technologies Inc. to provide biomarkers to industry and academia to accelerate research in the field of neurodegenerative and neuropsychiatric diseases. [ABSTRACT FROM AUTHOR]

Background: In a previous study, we showed the sensitivity of the Buschke Selective Reminding test Total Retention (BSRT TR) for detecting and predicting 4‐year cognitive change in cognitively intact older adults. Here, we examined how BSRT TR related to tau load. Method: The study sample consisted of 29 cognitively intact participants of the Flemish Prevent AD cohort KU Leuven (CDR global=0, MMSE=28‐30, mean age=68.5 year) and eight memory‐clinic recruited patients with prodromal AD (CDR global=0.5, MMSE=21‐30, mean age=71 year). All participants underwent neuropsychological testing, MRI and 18F‐MK6240 PET. Structural T1‐weighted MRI images were processed using CAT12. Tau load was quantified as 18F‐MK6240 DVR with the inferior cerebellum used as reference region. The primary analysis was based on voxel‐wise regression analysis between 18F‐MK6240 DVR scans and BSRT TR mean scores, with age as covariate. The second primary endpoint was the correlation between BSRT TR mean scores and tau DVR in an entorhinal and perirhinal volume of interest (VOI) based on the Brainnetome atlas. The secondary analysis was a VOI analysis based on an early metaVOI also containing hippocampus, parahippocampus, fusiform gyrus and inferior temporal cortex. Results were compared to those obtained for other episodic memory scores (BSRT delayed recall (DR), Auditory Verbal Learning Test (AVLT) total learning (TL) and AVLT DR), and an MRI volumetric analysis. Result: Voxel‐wise regression analysis showed a negative correlation between BSRT TR mean and DVR in ento‐ and perirhinal cortex bilaterally (x=22, y=‐4, z=‐32, extent=612 voxels, cluster level pcorr<0.001) (x=‐30, y=‐22, z=‐20, extent=2910 voxels, cluster level pcorr<0.001) (Figure A). This was confirmed by the analysis of the entorhinal/perirhinal VOI (rho=‐0.44, p=0.0063; Figure B). Significant correlations with ento‐ and perirhinal DVR were also present for BSRT DR (rho=‐0.39, p=0.017) and AVLT TL (rho=‐0.41, p=0.012), but not for AVLT DR (rho=‐0.29, p=0.084). No significant correlation was found with tau in the early metaVOI for any of the verbal episodic memory measures. Voxel‐wise regression analysis between gray matter volume and BSRT TR mean scores did not show significant results. Conclusion: The earliest measurable tau aggregate levels in ento‐ and perirhinal cortex correlate with classical episodic memory parameters. [ABSTRACT FROM AUTHOR]

Background: Plasma phosphorylated tau‐181 (p‐tau181) is one of the most reliable blood‐based biomarkers for Alzheimer's Disease (AD), yet has been consistently quantified with assays employing an antibody cross‐reacting with p‐tau175. We investigated the ability of a novel phospho‐specific plasma p‐tau181 assay to predict longitudinal amyloid accumulation in cognitively unimpaired (CU) elderly. Methods: Plasma p‐tau181 was quantified with a novel Simoa‐based assay using a p‐tau181 specific antibody (ADx252) in a prospective longitudinal cohort of 78 CU elderly (baseline age[mean] = 70 years, 47% female, 51% APOE‐ε4 carriers). Plasma sampling, amyloid‐β (Aβ)‐PET ([18F]flutemetamol or [11C]PiB) and episodic memory tests (average Buschke Selective Reminding total retention /12) were performed at baseline and Aβ‐PET was repeated on average 5 years later (SD: 2, range 1‐10 years). Longitudinal amyloid accumulation was modelled through calculation of Aβ‐PET rate of change by dividing the difference between baseline and follow‐up tracer uptake (in Centiloids) by the time interval between scans (in years). A Spearman correlation matrix between longitudinal amyloid accumulation, plasma p‐tau181, baseline amyloid load, age and baseline episodic memory was calculated. Z‐scores of all variables were entered into a hierarchical regression analysis in a stepwise manner based on correlation strength and finally sex and APOE were added. Results: Baseline plasma p‐tau181 correlated with longitudinal amyloid accumulation (ρ = 0.28, Figure 1A) as well as age (ρ = 0.31, Figure 1B). A negative correlation was found between age and baseline episodic memory scores (ρ = ‐0.33), and between p‐tau181 and baseline episodic memory scores (ρ = ‐0.19), although the latter did not reach significance. Apart from plasma p‐tau181, none of the other baseline variables correlated significantly with longitudinal amyloid accumulation (Figure 2). Hence, plasma p‐tau181 was entered into the base regression model. This base model (F(1,76) = 7.78, p = 0.007) explained 27% of the variance (R2) in longitudinal amyloid accumulation. Including additional predictors like demographics, clinical data or baseline amyloid load, did not further improve the model. Conclusion: Plasma p‐tau181 measured with the novel phospho‐specific Simoa assay best reflects longitudinal amyloid accumulation in asymptomatic Alzheimer's Disease with no additional predictive value of the other baseline factors included. [ABSTRACT FROM AUTHOR]

Background: Alzheimer's disease (AD) blood‐based biomarkers are becoming widely researched. Their requirement in the asymptomatic phase of AD is high to determine individuals who are in the early stages of disease. We investigate gene network expression to better understand molecular changes in this asymptomatic phase. Method: RNA was extracted and sequenced for 63 cognitively healthy Flemish Prevent AD Cohort KU Leuven (F‐PACK) participants (70 (56‐80) years) at baseline and follow‐up (interval: 4.6 (3.4‐8.59) years, 30 APOE4 carriers). Participants received amyloid‐PET and structural MRI at both timepoints. We performed weighted gene co‐expression network analysis (WGCNA) to identify networks of co‐expressed genes at baseline. These modules were correlated with traits of interest (APOE4 (present/absent) and amyloid rate of change ([follow‐up – baseline amyloid load], divided by time interval (years)) using Pearson's correlation coefficient. Significantly correlated modules were functionally annotated using Gene Ontology (GO) and Kyoto Encyclopaedia of Genes and Genomes (KEGG) in R. Potential hub genes were identified if the module contained genes with eigen‐based connectivity >0.7. These were visualised using STRING to create protein‐protein interaction networks. The top four hub genes (with highest Maximal Clique Centrality values) were then selected. Result: A gene co‐expression network was built on 9,149 genes that passed the read count threshold (genes with >10 read counts in >90% of samples). From this co‐expression network, 22 co‐expression modules were constructed (Figure 1A) and correlated with traits of interest. For the tan module (including 188 genes) APOE4 carriership was significantly negatively correlated (p=0.03, r=‐0.22), whereas amyloid rate of change was significantly positively correlated (p=0.03, r=0.21). Top GO processes for this module included T cell activation and differentiation, and top KEGG pathways included T cell receptor signalling, and Th1 and Th2 cell differentiation. The top four hub genes were GZMA, NKG7, TBX21 and PRF1 (Figure 1B). Conclusion: Baseline gene expression networks differ between APOE4 carrier status, and with respect to amyloid rate of change. This suggests there are network alterations in the asymptomatic phase of AD that can be detected in blood. These results may aid in development of targets for biomarker development or druggable targets for AD. [ABSTRACT FROM AUTHOR]

Background: Language dysfunction is common in Alzheimer disease (AD). There is increasing interest in the preclinical phase of AD. Here we tested longitudinally how activity in the associative‐semantic language network, measured with task‐based fMRI, changes in relation to increases in amyloid load, measured with 18F‐flutemetamol and 11C‐PIB. The main hypothesis was that an increase in amyloid would be associated with changes in activity in left or right posterior superior temporal sulcus (STS) and/or medial temporal gyrus (GTm). Method: A cohort of 35 cognitively intact older adults (mean age=65.4; range: 52‐78 years, 54.3% APOE‐ε4) participated who all received a task‐based fMRI and amyloid PET in 2009‐2011 and a repeat scan for both modalities in 2016‐2017. As primary analysis, we examined how longitudinal change in amyloid, measured using Centiloids (CL), relates to longitudinal activation change in the posterior temporal nodes of the associative‐semantic language network. The left and right posterior STS and GTm were defined as regions of prior interest using the Fan Brainnetome atlas, based on prior studies in amnestic mild cognitive impairment (MCI) and clinically probable AD using the same functional paradigm (Nelissen et al., 2007; Vandenbulcke et al., 2007; Adamczuk et al., 2016). Within this VOI, significance was determined based on SPM12 small volume correction. As a secondary analysis, we also performed a voxelwise whole‐brain analysis with a FWE whole‐brain corrected threshold <0.05. We also examined how the change in amyloid and the change in fMRI activation related to changes in offline language measures. Result: In the right posterior STS, the change in response amplitude during the associative‐semantic versus the visuoperceptual task showed a positive correlation with the change in amyloid load (x=51 mm, y=‐31 mm, z=5 mm; Z=3.47, Pcorr=0.010; Fig. 1). The change in response amplitude also correlated with longer offline confrontation naming reaction times (RT; r=0.42, P=0.035; Fig. 2). In addition, in the whole‐brain analysis we observed a decrease in response in the right inferior frontal sulcus with increasing amyloid load (x=3 mm, y=20 mm, z=56 mm; Z=4.64, Pcorr=0.021). Conclusion: This study confirms the central role of the posterior STS as area of predilection in the earliest stages of AD. [ABSTRACT FROM AUTHOR]

Background: We investigated whether an Alzheimer's disease (AD) polygenic risk score (PRS) could predict longitudinal brain amyloid accumulation in cognitively intact older adults participating in the Flemish Prevent AD Cohort KU Leuven (F‐PACK). Previously published data have highlighted associations between AD susceptibility loci and AD‐related brain changes. Method: Sixty‐one cognitively healthy F‐PACK participants (median age: 68 (56‐79), 29 females, 29 APOE ε4 carriers) received an [18F]Flutemetamol‐PET scan and structural MRI at baseline and follow‐up (average time interval 4.6 years (3.4‐8.6)). Amyloid rate of change was defined as absolute change divided by time interval. Participants were genotyped using the Illumina GSA and imputed using the Michigan server and HRC reference panel. Data were quality controlled using PLINK(v1.9, http://pngu.mgh.harvard.edu/purcell/plink/, Purcell et al., 2007) (minor allele frequency >0.01). PRS calculation was performed using PRSice(Choi & O'Reilly, 2019) with the Stage 1 summary statistics (containing SNP p‐values and corresponding effect sizes) from the GWAS performed by Kunkle et al.,(2019). European individuals from the 1000 Genomes dataset were used for clumping (N=503). We calculated PRSs at three thresholds for SNP inclusion: pT=0.5(Escott‐Price et al., 2015); 1x10‐5; 5x108. A Mann‐Whitney U‐test was used to determine amyloid rate of change differences between APOE ε4 carriers and non‐carriers. Linear models were used to assess if the PRS could predict changes in amyloid, with age, sex and the first three principal components from the PRS calculation as covariates, at each of the p‐value thresholds for SNP inclusion. Result: APOE ε4 carriers showed a trend towards a significantly higher rate of change than non‐carriers (U=459.5, p=0.056). When predicting amyloid rate of change, the linear model outputs were similar when the thresholds for SNP inclusion were pT=1x10‐5 and 5x10‐8. Although models did not reach significance, there was a significant effect of PRS on amyloid rate of change: pT=1x10‐5: β=0.0056 (CI: 0.0000785‐0.011), p<0.05; pT=5x10‐8: β=0.0054 (CI: 0.000042‐0.011), p<0.05. PRS was not a significant predictor for amyloid accumulation when the threshold for SNP inclusion was pT=0.5: β=0.0013 (‐0.004‐0.007), p>0.05. Conclusion: PRSs calculated with a SNP inclusion threshold of pT=1x10‐5 or 5x10‐8 explain a small but significant amount of variance in amyloid accumulation in cognitively intact older adults. [ABSTRACT FROM AUTHOR]

Background: We use the Flemish Prevent AD Cohort KU Leuven (F‐PACK) to determine changes in amyloid load over a longitudinal period, and investigate the influence of APOE ε4 on these changes, comparing three reference regions. Method: 64 cognitively intact healthy F‐PACK participants (baseline age: 68 (55‐80), 30 carriers of APOE ε4) received a static [18F]Flutemetamol‐PET scan at a baseline visit (2010‐2015, acquisition window: 90‐120 minutes), as well as extensive neuropsychological testing (scores within published norms) and structural MRI scan. A follow‐up PET scan has been taken 3.4‐8.6 years after baseline in order to calculate changes in amyloid SUVRcomp in a neocortical composite volume of interest. Either subject‐specific cerebellar grey matter (CGM), or eroded subcortical white matter (ESWM), or pons was used as a reference region. Spearman's correlation was used to determine the influence of baseline SUVRcomp on follow‐up SUVRcomp. APOE ε4 effects were determined using Wilcoxon matched pairs signed rank tests or Mann‐Whitney U‐tests, for the effects on absolute SUVRcomp and rate of change respectively. Rate of change was defined as absolute change divided by time interval. Result: There was a significant correlation between baseline and follow‐up SUVRcomp for each of the reference regions (CGM: R = 0.69, p = 3.25e‐10; ESWM: R = 0.75, p = 1.14e‐12; pons: R = 0.75, p = 1.29e‐12). There was a significant increase in SUVRcomp at follow‐up, compared to baseline, in APOE ε4 carriers when using only the CGM or pons (CGM W = 118.5, p = 0.02; pons: W = 79, p = 0.003; ESWM: W = 208.5, p = 0.63). No significant differences were found between baseline and follow‐up in non‐carriers, nor in baseline values between carriers and non‐carriers. APOE ε4 carriers had a significantly higher follow‐up amyloid SUVRcomp compared to non‐carriers, only when using the CGM or pons (GCM: U = 658.5, p = 0.05; pons: U = 706.5, p = 0.008; ESWM: U = 586, p = 0.31). Rate of change was also significantly elevated in APOE ε4 carriers versus non‐carriers when using the CGM and pons (GCM: U = 680, p = 0.02; pons: U = 717.5, p = 0.005; ESWM: U = 560, p = 0.51). Conclusion: Amyloid deposition and rate of change are enhanced by APOE ε4. Cerebellar grey matter and pons are the more suited reference regions to capture early differential effects of the APOE ε4 polymorphism on amyloid accumulation. [ABSTRACT FROM AUTHOR]

Background: A key component of cognitive aging is episodic memory decline. Here we examined to which degree structural integrity of (para)limbic cortex and amyloid load contribute to normal episodic memory variation in subjects between 50 and 80 years old using mediation analysis. The main hypothesis was that the effect of age on episodic memory would be mediated through hippocampal volume. Method: 180 community‐recruited cognitively intact older adults belonging to the Flemish Prevent Alzheimer's Disease Cohort KU Leuven (F‐PACK) cohort had cognitive test scores within norms as inclusion criterion, genotyping (APOE ε4 and BDNF), MRI and [18F]flutemetamol‐PET. Factor analysis was applied to determine the latent structure in the neuropsychological dataset. The primary factor (variance explained: 23%, Eigenvalue: 2.26) mainly contained episodic memory test scores: Buschke selective reminding (loadings: learning: 0.83, delayed‐recall: 0.96) and Rey Auditory Verbal Learning Test (0.60). Structural T1‐weigthed MRI were processed using voxel‐based morphometry. Hippocampal, entorhinal, perirhinal and parahippocampal volumes were derived by intersecting the modulated gray matter with Brainnetome parcellations. Amyloid load was quantified in a composite neocortical volume on [18F]flutemetamol SUVR with cerebellar grey matter reference. Linear regressions or Mann‐Whitney U tests investigated associations with episodic memory, which were further analyzed with bootstrapped mediation (R‐package 'lavaan'). Result: Higher age was negatively related to lower episodic memory scores (β=‐0.066, p<0.001), to lower hippocampal volume (β=‐0.005, p<0.001) and to lower entorhinal cortex volume (β=‐0.003, p<0.001). Lower hippocampal volume (β=3.80, p=0.0004) as well as entorhinal cortex volume (β=3.67, p<0.001) were related to lower episodic memory. There was no significant correlation of amyloid nor a genetic effect regarding age or episodic memory (p>0.12). Hence, we did not include amyloid or genotypes in further analyses. In the mediation analysis, hippocampal volume did not mediate the relation between age and episodic memory (Figure 1A). By contrast, entorhinal cortex volume did mediate the relation between age and episodic memory, however the direct path coefficient was also significant (Figure 1B). No significant indirect effects were found in the path analysis for the perirhinal cortex or parahippocampus (p>0.14). Conclusion: The effect of age on episodic memory is partly mediated by entorhinal rather than hippocampal volume in cognitive aging. [ABSTRACT FROM AUTHOR]

Background: Depression is associated with a risk of developing dementia and is hypothesized to enhance (pathological) brain aging. Previously we reported that amyloid burden in late‐life depression (LLD) did not differ from healthy controls in cortical regions typically associated with Alzheimer's disease (AD) [1]. Here we extend our analysis beyond hippocampal volume and AD‐specific volumes‐of‐interest, to the whole brain. Method: Forty‐eight individuals with LLD (age 74.1±7.5 years, 33 female) and 52 healthy controls (72.4±6.4 years, 37 female) underwent static amyloid PET imaging with [18]F‐flutemetamol and 3D T1‐weighted structural MRI. Amyloid deposition was quantified using the standardized uptake value ratio (SUVR) relative to the cerebellar grey matter (GM). Whole‐brain voxel‐wise comparisons of amyloid PET SUVR and GM volume (GMV) between LLD and controls were performed using SPM12, with age as covariate. Total intracranial volume was included as an additional covariate in the GMV analysis. The significance threshold was set to a family‐wise error (FWE) corrected p <0.05 at cluster‐level with an individual voxel height threshold p = 0.001. Result: There were no statistically significant group differences in amyloid deposition. In contrast, compared to controls, individuals with LLD showed widespread significant GMV reductions, predominantly in the left temporo‐parietal and the right occipital regions (Figure). Conclusion: Late‐life depression is associated with lower GMV in regions also implicated in AD, but is not associated with increased amyloid burden. Future analysis investigating correlations with neuropsychological measures of mood and cognition, and accounting for the potential effects of regional atrophy and white matter pathology on amyloid quantification, may further extend our knowledge about differences in the underlying neurobiology of LLD and AD. [ABSTRACT FROM AUTHOR]