AbstractBackgroundKnowledge of the temporal evolution among plasma Alzheimer’s biomarkers and PET will inform disease pathophysiology and appropriate implementation of plasma biomarkers (Therneau, NeuroImage 2021; Betthauser, Brain 2022). This study investigates the relative timing of progression of plasma p‐tau181, p‐tau217, GFAP, and amyloid PET on individual and population levels.MethodWe included 957 participants from the Mayo Clinic Study of Aging with clinical diagnoses of cognitively unimpaired (n = 732), MCI (n = 205), and Alzheimer’s clinical syndrome dementia (n = 20). An accelerated failure time (AFT) model was fit with amyloid (PiB) PET, plasma p‐tau217 (Meso Scale Discovery, Lilly), plasma p‐tau181, and GFAP (Simoa, Quanterix HD‐X) as quad‐variate endpoints, and sex, APOE genotype status, and education as covariates. The model output included a per subject adjustment or time‐shift for each biomarker indicating how much earlier or later it was estimated to progress relative to the average participant, the correlation (R) between adjustments for each pair of biomarkers, and the impact of covariates on each biomarker’s timing. Secondly, we estimated progression at a given age for selected cutpoints.ResultTemporal trajectories for each biomarker are shown in Figure 1. Individual‐level adjustments between plasma p‐tau217 and amyloid PET were highly correlated, R = 0.69(0.64‐0.74); how early or late an individual progressed on plasma p‐tau217 explained 48% of the variation in relative timing of amyloid PET progression. Associations of plasma p‐tau181 and GFAP with amyloid PET individual adjustments were moderate and low, respectively (Figure 2). Using the chosen cutpoints, the estimated order of biomarker progression was GFAP, amyloid PET, plasma p‐tau181, and plasma p‐tau217 (Figure 3).ConclusionOn the individual level, the strong association of the relative timing of progression of plasma p‐tau217 and amyloid PET supports that these processes are mechanistically related. The relatively lower associations of the timing of p‐tau181 and GFAP with amyloid PET likely reflect a lower biological association and greater measurement noise for p‐tau181. The individual adjustments, the primary output of this model, are independent of cutpoints. On the population level, relative timing of biomarker change is highly dependent on the cutpoints applied. This true for any analysis that employs cutpoints, regardless of the approach.
Comparing approaches based on the global functional organization of the brain versus local connectivity to predict tau‐PET across the Alzheimer’s disease phenotypic spectrum
AbstractBackgroundPredictive models of tau‐PET accumulation have largely relied on the functional connectome to estimate its spatial patterns in Alzheimer’s disease (AD). Patterns of global functional organization represent a promising avenue as they encapsulate many biological properties relevant to neurodevelopment, cognition and degeneration. We compared approaches based on local connectivity (from task‐free fMRI) and global organization (from FDG‐PET) to predict regional tau‐PET across AD phenotypes.MethodWe included 430 cognitively unimpaired (CU) individuals (214 A+T‐, 216 A+T+) and 62 amnestic mild cognitive impairment (aMCI), 64 amnestic AD, 21 dysexecutive AD (dAD), 41 logopenic progressive aphasia (LPA) and 49 posterior cortical atrophy (PCA) patients (all A+T+) from the Alzheimer’s Disease Research Center, Mayo Clinic Study of Aging, and Neurodegenerative Research Groups (referred to as “Mayo cohort”). The connectivity approach consisted of assessing the relationship between a region’s tau level and its functional connectivity to the region of highest tau (epicenter). The global organization approach consisted of 1) performing a principal component analysis on 3000 FDG‐PET images that did not overlap with the Mayo cohort, 2) projecting FDG‐PET images of the Mayo cohort onto the low‐dimensional space generated in Step 1 and extracting corresponding weighted values across the first 100 components, 3) using these weighted values as predictors of regional tau‐PET in a linear model and determining optimal parameters based on predicted R 2 values, and 4) using the parameters from the model built in Step 3 to predict tau‐PET in the ADNI cohort (n = 86) (Fig.2).ResultThe connectivity‐based approach yielded adjusted R 2 values ranging from 0.19‐0.45 across diagnostic groups (Fig.1). The global organization approach yielded averaged adjusted R 2 and predicted R 2 values of 0.71 and 0.63, respectively, in the Mayo cohort. Out‐of‐sample median R 2 in the ADNI cohort was 0.41 (range 0.10‐0.60) (Fig.3).ConclusionAn unbiased approach based on the global functional organization of the brain and using an imaging modality widely used in clinical settings outperformed a connectome‐based one to predict regional tau‐PET across the AD phenotypic spectrum. Patterns of tau accumulation may better relate to the large‐scale physiology of the brain underlying specific mental functions rather than local connectivity factors.
White matter changes with cerebrovascular disease and Alzheimer’s disease pathologies
AbstractBackgroundWhite matter (WM) plays an important role in cognitive function and undergoes substantial changes due to aging and increase in AD and AD related dementia (AD/ADRD) pathologies. While recent work proposed measurement of global WM changes from diffusion MRI as biomarkers for Cerebrovascular disease (CVD) such as PSMD and free water, there needs to be a better understanding of WM changes due to AD/ADRD pathologies. The goal of our recent work was to understand white matter changes with CVD and AD – common pathologies in AD/ADRD ‐ utilizing data from Mayo Clinic Study of Aging and Mayo ADRC.MethodUsing diffusion MRI (dMRI) in 718 participants from Mayo Clinic Study of Aging (mean age = 71.1 years, 56% male), we investigated relationships between commonly used dMRI measures and vascular risk as well as global amyloid PET. In a subset of 233 participants who were amyloid‐PET positive and also competed Tau‐PET (flortaucipir) from Mayo Clinic Study of Aging and Mayo ADRC, we investigated associations between regional flortaucipir SUVRs in Braak stages and regional WM changes.ResultsWe found greater frontal lobe WM damage with increasing vascular risk across all dMRI measures. We previously validated the strong relationship between the genu of the corpus callosum (anterior interhemispheric connections) dMRI measurements made on antemortem scans and cerebrovascular pathology observed on postmortem tissue (Nguyen et. al. Acta Neuropathologica 2022) supporting the utility of genu of the corpus callosum dMRI measures as a potential CVD biomarker. Also, we found greater posterior WM damage with increasing amyloid burden. Among the 233 amyloid positive participants, we found spatially dependent WM degeneration associated with regional flortaucipir SUVRs in Braak stages with greater extent of posterior WM damage. Further, dMRI measurements of the WM damage provided complementary information about disease staging and progression of AD in addition to flortaucipir SUVR measurements.ConclusionsAnterior WM measurements may be more specific to CVD and posterior WM changes are seen with increasing AD pathological burden. Our results suggest that WM changes observed with CVD and typical AD may be different and have clinical utility for differential diagnosis and disease staging.
Comparison of sleep and wake EEG biomarkers in mild cognitive impairment and Alzheimer’s disease
AbstractBackgroundEEG slowing, a biomarker of cognitive decline in Alzheimer’s disease (AD), has been reported in wake or sleep EEG [1,2,3]. However, the relation between wake and sleep EEG and their comparative sensitivity to early stage of cognitive decline is less understood. In this study we compared sleep and wake EEG biomarkers using a common set of endpoints.MethodsParticipants included individuals with mild cognitive impairment (MCI, n = 30, ages 53‐88), Alzheimer’s disease dementia (ADem, n = 15, ages 60‐84), and controls with normal cognition (HC, n = 55, ages 50‐84). Twenty‐channel resting‐state EEG with eyes‐closed was collected during 5‐minutes of wakefulness (STAT X24). Sleep EEG recordings from AF7‐AF8 were collected in‐home and staged using Sleep ProfilerTM. Power spectral densities (PSD) and the derived Theta‐Alpha ratios were computed for wake EEG sites (counterpart bipolar montage), with the wake FP1‐FP2 compared by stage to the sleep EEG at AF7‐AF8. Significant differences in Theta‐Alpha ratios between MCI or ADem vs. HC groups were identified using independent t‐test (p<0.05) with the normalized effect size measured using Hedges‐g (E.S.).ResultsCorrelation coefficients between wake Alpha, Theta power and non‐REM sleep were: r = 0.83, 0.75 for stage N1, r = 0.71, 0.66 for stage N2, and r = 0.55, 0.50 for stage N3. Based on the frontal Theta‐Alpha ratios, the ADem group showed significant EEG slowing during both wake (E.S. = 0.73) and sleep (E.S. = 1.65, 1.31, 0.61, 1.71 for N1, N2, N3, and REM) as compared to the HC, while MCI group only exhibited EEG slowing during sleep (E.S. = 0.56, 0.77, 0.51 for N1, N2 and REM). Temporal (T5‐T6) wake EEG manifested significant slowing in both ADem (E.S. = 1.21) and MCI (E.S. = 0.56) groups. Compared to the HC, ADem patients exhibited significantly less sleep time (E.S. = 0.91, p = 0.002) and less stage REM (E.S. = 1.0, p = 0.001).ConclusionsCognitive decline in the AD spectrum is associated with common signatures in both sleep and wake EEG. However, the magnitude of these effects and their spatial distribution may be different in sleep versus wakefulness. These findings support a multi modal approach to neurophysiological assessment using both sleep and wakefulness data.
Cerebrospinal fluid biomarkers at presentation associate with etiologic diagnosis and treatment responsiveness in patients with rapidly progressive dementia
AbstractBackgroundThe causes of rapidly progressive dementia are heterogeneous, including neurodegenerative, vascular, and autoimmune/inflammatory diseases. Overlap in presenting symptoms and signs, results of cognitive testing, and findings on brain imaging and other common diagnostic tests confounds the etiologic diagnosis, contributing to diagnostic delays, missed opportunities for treatment, and poorer outcomes in patients with RPD. New tools and approaches are needed to improve diagnostic accuracy, with particular emphasis on improving early recognition of patients with potentially treatable causes of RPD.MethodsEstablished and emerging biomarkers of Alzheimer disease neuropathology (Aβ42/40, p‐tau181, p‐tau231), neuroaxonal and neuronal injury (neurofilament light [NfL], VILIP‐1, total tau), neuroinflammation (YKL‐40, sTREM2, GFAP, MCP‐1), and synaptic dysfunction (SNAP‐25, neurogranin) were measured in cerebrospinal fluid obtained at presentation from 78 prospectively accrued patients with RPD due to neurodegenerative (n = 35), vascular (n = 9), and autoimmune/inflammatory (n = 34) diseases, and 72 age‐ and sex‐similar cognitively normal individuals (controls). Causes of RPD were further classified as potentially treatment‐responsive or nonresponsive, referencing the extant literature. Biomarker levels were compared across etiologic diagnoses and by treatment responsiveness, referencing control values. Analyses were adjusted for the effects of age (ANCOVA).ResultsAlzheimer disease biomarkers (low Aβ42/40 and elevated p‐tau181 and p‐tau231) were associated with neurodegenerative causes of RPD (ANCOVA, p = 0.02 p = 0.30, p = 0.15), while high NfL (p = 0.61), low VILIP‐1 (p = 0.009), and elevated markers of inflammation (sTREM2, p = 0.009; YKL‐40, p<0.001) identified patients with autoimmune/inflammatory diseases (Figure). MCP‐1 levels were highest in patients with vascular causes of RPD (p = 0.08). 44 patients (56.4%) had potentially treatment‐responsive causes of RPD. Treatment responsiveness was associated with higher levels of CSF Aβ42/40 (p = 0.003), NfL (p<0.001), sTREM2 (p<0.001) and YKL‐40 (p<0.001); and lower levels of p‐tau‐181 and p‐tau231 (p<0.001, p<0.001) and VILIP‐1 (p<0.001).ConclusionSelected biomarkers measured in CSF at presentation associate with etiologic diagnosis and treatment responsiveness in patients with neurodegenerative, vascular, and autoimmune/inflammatory causes of RPD. Biomarker panels may be adapted to improve recognition of patients with treatment‐responsive causes of RPD early in the symptomatic course when treatments may be most effective.
Leveraging Large‐scale CSF Proteomics to identify Disease Progression Biomarkers in Autosomal Dominant Frontotemporal Lobar Degeneration
AbstractBackgroundThe pathophysiological mechanisms driving frontotemporal lobar degeneration (FTLD) disease progression and corresponding biomarkers that capture these changes are not fully understood. We leveraged high‐throughput proteomics to identify networks of co‐expressed CSF proteins and candidate biomarkers for prediction of disease progression in adults with autosomal dominant FTLD mutations.Method113 adults carrying autosomal dominant FTLD mutations (C9orf72/GRN/MAPT) and 38 noncarrier controls from the ALLFTD consortium completed baseline lumbar puncture and longitudinal neuropsychological testing. CSF was assayed for SOMAmer proteomics on SOMAscan® v3.0 (4,140 proteins). Network analysis identified weighted protein co‐expression modules, which were annotated via gene ontology and cell‐type enrichment analysis. Disease age estimates that capture proximity to symptom onset were derived from a Bayesian disease progression model incorporating genotype‐specific clinical (cognition, CDR®+NACC FTLD) and biomarker (brain volume, plasma NfL) information. At baseline, local‐weighted regression examined module z‐scores (based on mean/SD of module levels [first principal component] in noncarriers) in relation to disease age. Longitudinal linear mixed‐effects models examined global cognitive trajectories in relation to baseline module z‐scores, as well as individual proteins within prognostic modules.ResultNetwork analysis identified 32 CSF protein co‐expression modules (37 to 223 proteins/module). 16 modules showed mutation‐related deviations from noncarriers. Deviations in ion transport, neurodevelopmental, metabolic, and transcriptional modules occurred several decades before estimated symptom onset (disease age: ‐30 to ‐20 years) and persisted across disease age. Deviations in neuroinflammatory and ubiquitination/autophagy modules occurred ∼10 years before estimated symptom onset. Sharp deviations in RNA binding, protein phosphorylation, neurogenesis, synaptic function, and lysosomal modules occurred near symptom onset (disease age: ‐1 to 4 years). Baseline synaptic and neurogenesis modules were the strongest prognosticators of cognitive decline (module x time: bs>.10, ps<.05). Within these modules, individual proteins with stronger intramodular connectivity exhibited stronger associations with cognitive decline (e.g., synaptic/transmembrane module: r = .40, p = 2.85e‐6; e.g., EPHA5, LRRTM2).ConclusionLarge‐scale analysis of the CSF proteome reveals a biologically‐diverse landscape of protein networks associated with presymptomatic and symptomatic stages of autosomal dominant FTLD. Synaptic and neuron‐enriched protein networks exhibited prodromal deviations that forecasted incipient cognitive decline. ‘Hub’ proteins derived from these affected networks warrant further validation as candidate biomarkers for early FTLD disease progression.
AbstractBackgroundImaging markers are central to diagnose and monitor disease progression in dementia with Lewy bodies (DLB). This presentation provides an up‐to‐date overview of the contributions of positron emission tomography (PET) and magnetic resonance imaging (MRI) to the field of DLB.MethodsExpert opinion and road‐map for the future based on the current state‐of‐the‐art.ResultsPET for Lewy pathology is an active area of research but a reliable tracer is still lacking. Tracers for glucose metabolism and dopaminergic deficits have merited inclusion in diagnostic criteria for DLB. Alzheimer’s disease (AD) co‐pathology is common in DLB. Amyloid PET has been used in DLB for a long time and most studies show an elevated frequency of a positive amyloid scan in DLB. Tau PET has recently attracted attention, with studies suggesting an atypical pattern of tau PET uptake in DLB, different from the pattern seen in AD. Newest advances include PET tracers for synaptic activity, microglial activation, and neuromelanin. MRI findings in DLB traditionally include preservation of hippocampal volume, as compared with AD. More recent findings encourage the inclusion of a pattern of cortical neurodegeneration, mostly affecting posterior cortical areas, with sparing of the hippocampus. This finding is consistent across MRI markers (volume, thickness, functional MRI, diffusion, perfusion, and connectivity). Advanced techniques for cholinergic system have shown vulnerability of cholinergic basal forebrain and its white matter projections.ConclusionsNeuroimaging is central for DLB diagnosis and for advancing mechanistic knowledge about the disease. Biomarkers of Lewy pathology are emerging and will revolutionize the field, but in the absence of topographical markers, the reviewed techniques will have an important role in precision medicine, for example, helping to identify subgroups with different disease trajectories or treatment response.
Regulation of meningeal lymphatic function by APOE
AbstractBackgroundThe meningeal layers form a unique interface between the central nervous and peripheral immune systems. The lymphatic vasculature that is present in the meningeal dura and drains cerebrospinal fluid (CSF) becomes dysregulated during aging and in models of neurodegenerative disorders such as Alzheimer’s disease (AD). Apolipoprotein E (APOE), and the isoform APOE4 in particular, increases the risk for late‐onset sporadic AD. However, it is still poorly understood whether changes in meningeal lymphatic morphology and drainage are mediating the deleterious effects of APOE4 expression. In the present study, we investigate the effects of human APOE3 or APOE4 on meningeal lymphatic function and immunity.MethodWe evaluated the morphology of lymphatic vessels in dural whole mounts of age‐matched wild type, Apoe‐KO, APOE3, and APOE4 mice at 2‐3, and 11‐12 months. This assessment was performed in both male and female mice. The drainage function of meningeal lymphatics is being assessed in the same mice by in vivo imaging of CSF‐derived fluorescent tracers into the deep cervical lymph nodes. We have also performed RNAscope and protein measurements using murine meningeal tissue to investigate the main cellular source(s) of Apoe, and single‐cell RNA sequencing to explore the cell‐specific transcriptomic signatures linked to the changes in meningeal lymphatic function.ResultLymphatic vessel length is increased in 11‐12‐month‐old APOE4 male mice, when compared to their APOE3 counterparts. Interestingly, the same was not detected in female mice. Dural myeloid cells, expressing Itgam (encoding for CD11b), are the main source of Apoe mRNA transcripts, and one of the main sources of apoE protein in the brain dura mater. APOE4 expression and age markedly alter the transcriptional profile of meningeal myeloid cells, which show overexpression of lysosomal, antigen presentation, and complement genes.ConclusionAPOE4 expression leads to a sex‐dependent meningeal lymphatic vascular remodeling, which is associated with an exacerbated activation of meningeal macrophages. In future studies, we will test whether the changes in meningeal lymphatics observed in mice expressing APOE4 are mediated by myeloid cell derived APOE.
Single‐nuclei transcriptomic identifies type‐specific neuronal cell vulnerability and non‐neuronal molecular changes in Amnestic and Logopenic Variant Primary Progressive Aphasia Alzheimer’s disease
AbstractBackgroundIndividuals with neuropathological criteria for Alzheimer’s disease (AD) may manifest with atypical clinical syndromes. Past work showed that the neurobiological basis for these differences is related to specific neuronal vulnerabilities for tau pathology. For instance, amnestic cases have a higher burden of neurofibrillary changes in CA1. In contrast, logopenic variant primary progressive aphasia (lvPPA) cases have a higher tau burden in the superior temporal gyrus (STG). Single‐cell technology enables investigations on the molecular basis of differential neuronal vulnerability in AD. Therefore, we investigated factors underlying selective vulnerability using brain samples from individuals with pure AD with different clinical manifestations.MethodsnRNA Sequencing using the Chromium Single Cell 3′ (10X Genomics, USA) on nuclei cells extracted from the CA1 sector and posterior STG of postmortem brain tissue of 24 individuals either meeting pathological criteria for AD (A3B3C3; 12 amnestic and eight lvPPA) and healthy controls (A≤1B≤1C≤1; n = 4) (Table 1, Fig. 1A/B). Bioinformatics analyses were conducted using Cell Ranger and R software. Comparisons between cell subpopulations were conducted with Wald statistical test, and p‐values < .05 were considered significant.ResultAfter quality control, we recovered 98,180 nuclei with a mean of 2,130 genes per nuclei. Upon cross‐sample alignment and t‐stochastic neighborhood embedding clustering (Fig. 1C), we found 21 excitatory neuronal subpopulations (Exc‐sub) in CA1 and 26 in STG, and 22 and 27 inhibitory neuronal subpopulations (Inh‐sub) in CA1 and STG, respectively; 16 astrocytes subpopulations in both areas and 20 microglia subpopulations in CA1 and 17 in STG (Fig. 2). One STG Exc‐sub, expressing CUX2 and LAMP5 genes showed vulnerability in lvPPA patients. Also, one STG Inh‐sub, expressing the ADARB2 gene, showed vulnerability for all AD patients. Furthermore, two astrocytes subpopulations (one in both areas and one in STG) and three STG microglia subpopulations showed proportional composition changes comparing AD patients and HC.ConclusionOur preliminary study identified a vulnerable population of excitatory neurons related to lvPPA. We are conducting validation studies using quantitative pathology to confirm these results. Furthermore, analysis of a higher number of cases is ongoing and will continue to inform on factors associated with neuronal vulnerability.
Alzheimer’s disease‐associated post‐translational modifications of tau influence tau propagation
AbstractBackgroundIn Alzheimer’s disease (AD), tau accumulates with a stereotyped progressive involvement of brain regions. The cellular mechanism for this is the neuron‐to‐neuron propagation of tau, but the factors that influence tau propagation are not fully understood. We hypothesize that AD‐associated post‐translational modifications (PTMs) of tau reduce its propagation since lysines are critical for LRP1‐mediated tau uptake.MethodA construct was designed to model tau propagation in Chinese hamster ovary (CHO) cells and allow identification of donor and recipient cells by immunocytochemistry and flow cytometry. In addition to the wildtype (WT) tau sequence, constructs were made with pseudophosphorylation and pseudoacetylation mutations corresponding to the phosphorylations and acetylations found on low molecular weight (LMW) and high molecular weight (HMW) tau species in AD brains. Additionally, individual PTM mimic mutations corresponding to PTMs of the LMW tau species were made. The constructs were transfected into CHO cells and tau propagation was assessed by flow cytometry after 48 hours. Statistical analysis was performed by ANOVA with adjustment for multiple comparisons.ResultTau propagation using this cell culture model was demonstrated by immunocytochemistry and flow cytometry. Comparison of the WT, LMW‐mimic (8D2Q mutant), and HMW‐mimic (20D3Q mutant) tau propagation constructs revealed that the 8D2Q and 20D3Q mutants displayed reduced tau propagation compared to WT by 23.7% (p = 0.0186) and 26.5% (p = 0.0087), respectively. Comparing the ten single mutant constructs to WT, T181D and T231D showed increased tau propagation by 20.3% (p = 0.0002) and 15.9% (p = 0.0094), respectively. Compared to WT, T217D, K281Q, K353Q, and S404D showed reduced tau propagation by 17.3% (p = 0.0030), 15.3% (p = 0.0123), 23.7% (p <0.0001), and 16.6% (p = 0.0046), respectively.ConclusionThe mimics of AD‐associated tau PTMs found on LMW and HMW tau reduce tau propagation, consistent with the initial hypothesis. The impact of individual PTM mimics on tau propagation is variable, but notably both pseudoacetylation mimics reduce propagation, possibly due to lower affinity with LRP1. Understanding how AD‐associated tau PTMs impact tau propagation may suggest novel targets to slow the progression of AD pathology.