Citations

AbstractBackgroundThere have been no standard definitions of successful cognitive aging (SA). Our goal was to apply different SA definitions to the 80+ population in the Mayo Clinic Study of Aging (MCSA) on the basis of (i) cross‐sectional vs longitudinal cognition (“Exceptionality” vs “Maintenance”) and (ii) memory vs global cognitive scores and explore the participant characteristics for each SA definition on the basis of baseline imaging markers and mid/late‐life health.MethodWe included 2157 participants from the MCSA, all ≥80 years at baseline and classified them into two SA definitions using memory z‐scores or global z‐scores: “Exceptionality” when high cognitive performance was observed (z‐scores ≥ 1 SD), “Maintenance” when cognitive function was preserved over time (Slope ≥ ‐0.1/year using mixed‐effect models). Our imaging markers were: PET‐based amyloid positivity (A+), MRI‐based neurodegeneration positivity (N+), WMH, FDG‐PET SUVR markers of resilience (anterior cingulate, ACC) and AD‐related areas (posterior cingulate). We also examined APOE4 status and other health characteristics. Group comparison was performed within each definition (e.g. exceptional vs non‐exceptional memory) using ANCOVA and binary logistic regressions with adjustments for age, sex, and education.ResultA lower number of participants met the “Exceptionality” criteria (10%: memory and 4%: global). A greater number met the “Maintenance” criteria (53%: memory and 32%: global). Less than 7% showed both “Exceptional” cognition and “Maintenance” over time. Participants across SA definitions were in general younger, more often women, with a higher education/occupation and less often APOE4 carriers. All SA groups showed higher ACC metabolism and lower N+ compared to the corresponding non‐SA group. More than 50% of the participants in the “Exceptional” groups were A+ (57%: memory, 50%: global). A+ frequency was significantly lower in the “Maintainers” (<42%). Better mid/late‐life health indicators were observed in the SA groups, including lower cardio‐metabolic conditions, higher mid‐life cognitive activities, lower depressive symptomatology, lower hypersomnolence and better self‐perception of health.ConclusionDifferent SA definitions identified mainly non‐overlapping 80+ groups that showed several common features including higher education/occupation, percent of women, better lifestyles, lower N+ frequency, and higher ACC metabolism. Lower A+ frequency was a key distinct feature of “Maintenance”.

Abstract The Worldwide Alzheimer's Disease Neuroimaging Initiative (WW‐ADNI) is a collaborative effort to investigate imaging and biofluid markers that can inform Alzheimer's disease treatment trials. It is a public‐private partnership that spans North America, Argentina, Australia, Canada, China, Japan, Korea, Mexico, and Taiwan. In 2004, ADNI researchers began a naturalistic, longitudinal study that continues today around the globe. Through several successive phases (ADNI‐1, ADNI‐GO, ADNI‐2, and ADNI‐3), the study has fueled amyloid and tau phenotyping and refined neuroimaging methodologies. WW‐ADNI researchers have successfully standardized analyses and openly share data without embargo, providing a rich data set for other investigators. On August 26, 2020, the Alzheimer's Association convened WW‐ADNI researchers who shared updates from ADNI‐3 and their vision for ADNI‐4.

AbstractBackgroundThe Mayo Clinic Study of Aging and Alzheimer’s Disease Research Center used GE scanners exclusively for MRI from 2005‐2017, employing consistent imaging parameters throughout. In 2017 the MRI protocol was updated and migrated to Siemens Prisma scanners; 113 participants were scanned with each configuration. The GE protocol included a 2D FLAIR. The Siemens protocol includes 3D FLAIR and T2‐weighted images. Both vendor protocols included a 3D T1. Experience with multi‐center trials suggests such changes may introduce systematic variations.MethodsGE scanning was carried out on 3T scanners with 12 channel phased array head coils. Siemens Prisma scanning was carried out with 64 channel head/neck coils. Imaging sequences and parameters are given in Table 1. We used the MCALT template and priors to perform standard six‐class segmentation on T1‐weighted (T1w) images. SPM12 unified segmentation supports the simultaneous use of multiple input contrasts; segmentation of multi‐contrast inputs was done using eight tissue classes including deep grey and white matter hyperintensity (WMH) priors. For a given MRI study (either GE or Siemens) all images were registered and resampled to the study’s T1w image. Two‐channel (T1+FLAIR) segmentation was performed on GE scans. Two and three‐channel segmentations were performed on Siemens scans. Comparisons included total intracranial volume, total tissue volume and total grey matter volumes, as well as hippocampal volumes and grey matter volume in a temporal lobe meta‐ROI.ResultsFour participants were excluded for poor image quality: one poor GE T1w quality, one poor Siemens T1, T2 and FLAIR quality, and two poor Siemens T2 quality. Comparisons were made among 109 participants aged 34–97 (Table 2). By visual assessment, adding T2‐weighted images always improved intracranial volume correctness. Using the combination of single‐channel GE and three‐channel Siemens generally improved cross‐vendor agreement on grey matter volumes with the exceptions of hippocampal and intracranial volume estimates.ConclusionRequiring multiple input channels increases the likelihood that the result of a study being excluded. Combining T1w, FLAIR and T2w images allows the delineation of WMH and deep grey matter probability outputs but may introduce differences in GM volumes.

AbstractBackgroundAmyloid‐related imaging abnormalities (ARIA), edema (E) or hemorrhagic (H) type, have been reported in trials of anti‐β‐amyloid passive immunotherapies in sporadic and dominantly inherited Alzheimer Disease (DIAD). However, beyond APOE‐ɛ4 the risk factors and clinical implications of ARIA are not well understood, especially in DIAD populations. Here we characterize ARIA, focusing on ARIA‐E in the DIAN‐TU‐001 trial evaluating gantenerumab and solanezumab in DIAD.MethodThe DIAN‐TU‐001 trial enrolled 194 participants, including 144 DIAD mutation‐carriers receiving gantenerumab (n=52), solanezumab (n=52), or a placebo (n=40). Clinical assessments included the Clinical Dementia Rating (CDR). Imaging assessments included PiB‐PET and safety MR, including T2‐FLAIR and T2*‐GRE. Treatment dosage and APOE‐ɛ4 status were also reported.ResultEleven participants developed a total of 14 ARIA‐E episodes (discovered on scheduled safety MR) and three had associated mild symptoms reported in retrospect (e.g. headache, imbalance disorder). Though one ARIA‐E case was observed in the solanezumab arm, the following results focus on the gantenerumab arm to prevent unblinding. Recipients of gantenerumab were more likely to develop ARIA‐E compared to placebo (odds ratio (OR)=9.29, 95% confidence interval (CI)=[1.1,75.9], p‐value<0.05). ARIA‐E participants were PiB‐PET+ and 60% were CDR>0. APOE‐ɛ4 tends to be associated with risk for developing ARIA‐E (OR=5.0, 95%CI=[1.0,30.4], p‐value=0.055). ARIA‐E were not observed at initial 225mg dose and were first observed after a titration step (within 4‐24 weeks, after 1‐6 injections, Table 1). Time‐to‐resolution of ARIA‐E was 10.4±6.2weeks. Seven of ten ARIA‐E patients paused/reduced dose escalation and three discontinued treatment. Overall, developing ARIA‐E did not increase the odds of trial discontinuation (OR=1.57, 95%CI=[0.34,7.33], p‐value=0.57). ARIA‐E occurred primarily in the occipital lobe (71%) with associated incident ARIA‐H (microhemorrhages or superficial siderosis) in 60% of ARIA‐E participants. ARIA‐E size was associated with microhemorrhage count (Spearman’s rho=0.72, p‐value<0.05), rate of change (rho=0.78, p‐value<0.01), and baseline PiB‐PET (rho=0.68, p‐value<0.05). Normalizing by baseline, PiB‐PET decreased similarly in ARIA‐E+ (‐0.21±0.20) and ARIA‐E‐ (‐0.16±0.20) participants.ConclusionIn DIAD, gantenerumab dose over 225mg increased ARIA‐E risk, possibly more for APOE‐ɛ4 carriers. ARIA‐E was reversible, generally asymptomatic, and without increased odds of trial discontinuation. These aspects of drug response give insights for managing ARIA‐E occurrence in future trials.

AbstractBackgroundEarly changes in white matter (WM) measured using diffusion tensor imaging (DTI) are sensitive markers of WM degeneration and predict future cognitive decline in mild cognitive impairment (MCI). We sought to examine whether changes in WM integrity differed among empirically derived subtypes of incident MCI.MethodParticipants included 66 individuals with incident MCI enrolled in the population‐based Mayo Clinic Study of Aging (MCSA) previously identified via cluster analysis as having one of four MCI subtypes: subtle cognitive impairment (SCI; n = 13, mean age 78; 15% women), amnestic (n = 24; mean age 79; 54% women), dysnomic (n = 8; mean age 79; 38% women), and dysexecutive (n = 21; mean age 82; 33% women). All participants completed 3T MRI at the time of the MCI diagnosis. We compared the MCI subtypes to 100 cognitively unimpaired (CU) participants (mean age 77; 49% women). We used tract‐based spatial statistics to examine voxel‐level differences in fractional anisotropy (FA). We also fit a linear regression model to regional FA data from the corpus callosum (CC), the major interhemispheric WM connection, to quantify the relative differences in WM damage.ResultAll MCI subtypes except the subtle cognitive impairment group had widespread decreased FA. Relative to CU, the amnestic cluster showed decreased FA in the CC (genu), cingulum, fornix and uncinate fasciculus (UF); the dysnomic cluster showed decreased FA in the CC (genu, splenium), fornix, UF, anterior thalamic radiation (ATR), and corticospinal tract (CST); and the dysexecutive group showed decreased FA in the CC (genu, body, splenium), cingulum, UF, ATR, CST, and superior longitudinal fasciculus. The dysexecutive cluster had the most severe bilateral involvement among all MCI subtypes. The regional analysis showed that the amnestic, dysnomic, and dysexecutive groups (but not the subtle cognitive impairment group) had significantly lower FA of the CC compared to CU (p<0.001; Figure 1).ConclusionWe found widespread WM degeneration in the commissural and long association fibers in amnestic, dysnomic, and dysexecutive incident MCI subtypes. The extent of CC damage seen in Figure 1. confirmed the voxel‐level results and suggests that interhemispheric WM disconnection is relevant to the development of MCI.

AbstractBackgroundOptimizing software to measure Flortaucipir (AV‐1451) tau PET SUVR change‐over‐time is crucial for longitudinal observational studies and potential trials of anti‐tau agents. Ideal measurements should have good repeatability (low noise), and should clearly separate clinical cognitive groups.MethodWe compared 456 software methods varying in their use of: two major packages, each with cross‐sectional and longitudinal variants (FreeSurfer and an in‐house pipeline using SPM12, MCALT, and ANTs); three temporal‐lobe target regions (each with/without adjacent WM); seven reference regions; five PVC methods; and two statistics (mean vs. median). We used linear mixed‐effects models to compare methods using scans from 88 Mayo Clinic participants (46 cognitively impaired, 42 impaired, excluding those with symptoms inconsistent with the AD pathology imaged by Flortaucipir), each with three visits with PET and MRI. From these models, we calculated two criteria for each pipeline: relative measurement error (residual from the linear mixed model), and the t‐statistic for testing differences in group‐wise rates of change (which has a Cohen’s d interpretation of effect size for longitudinal data). We plotted both criteria orthogonally, then ranked methods overall according to the average of their ranks on both individual criteria.ResultMethods using the entorhinal cortex target region performed poorly in both criteria (Figure 1). The top 10 methods by the combined criteria (top‐left in Figure 2) all used the Mayo SPM12‐based pipeline with an eroded supratentorial WM (eSWM) or composite (eSWM + pons + cerebellum) reference region. Among them, 9/10 used the longitudinal variant, 8/10 used 2‐compartment PVC, and 7/10 used median. Altogether, our results favored SUVR methods with a longitudinal design (simultaneously calculating all time‐points), a temporal‐lobe meta‐ROI including adjacent WM, a composite reference region, 2‐class voxel‐based (Meltzer) PVC, and median statistics. Methods using GTM‐Tau PVC (Baker et al., Data in Brief, 2017) had the highest group separation but relatively large measurement error, giving them ranks 197 and 199 out of 456 with the joint (average rank) criteria.ConclusionCompared with popular FreeSurfer‐based methods (ranks 176, 258; Figure 2), our top‐ranked approaches approximately halved error in SUVR measurements while maintaining or improving group separation.

AbstractBackgroundPrior evidence has suggested associations between hearing loss and smaller brain volumes particularly in the temporal lobe where the primary auditory cortex is located, probably due to changes in cortical organization associated with reduced auditory stimulation. Evidence regarding associations of central auditory processing with brain volumes is relatively limited. The study aims to investigate cross‐sectional associations of central auditory processing with brain volumes in older adults.MethodWe studied 587 cognitively normal (Mini‐Mental State Exam ≥23 if high school degree or less and ≥25 if above high school degree) older adults (mean age: 77±4 years, 56% female, 17% black), including 426 with hearing loss from the Aging and Cognitive Health Evaluation in Elders (ACHIEVE) baseline (2018‐19), which is an ancillary randomized controlled trial of the Atherosclerosis Risk in Communities (ARIC) Study and 161 with normal hearing, comparable age and cognitive status from the ARIC Visit 6/7 (2016‐17/2018‐19). Brain volumes were measured by magnetic resonance imaging (MRI) scans and were transformed to z‐scores. Central auditory processing was measured by the Quick Speech‐in‐Noise (QuickSIN) test. Total scores range from 0 to 30 with lower scores indicating worse speech‐in‐noise performance. Multivariable‐adjusted linear regression was used to estimate cross‐sectional differences in standardized brain volumes associated with differences in QuickSIN score. Models adjusted for age, sex, race, education, cohort, body mass index, smoking, hypertension, diabetes, stroke, intracranial volume, and pure‐tone average.ResultEvery 5‐point decrease in QuickSIN score (worse performance) was associated with smaller volumes of the temporal lobe (‐0.07SD, 95% CI: ‐0.13, ‐0.01). QuickSIN score was also associated with subregions of the temporal lobe related to auditory processing, memory and emotion, including amygdala (‐0.13SD, 95% CI: ‐0.21, ‐0.04), middle (‐0.08SD, 95% CI: ‐0.15, ‐0.00) and superior temporal gyri (‐0.08SD, 95% CI: ‐0.15, ‐0.01).ConclusionPoorer central auditory processing was associated with smaller temporal lobe volumes, including but not limited to subregions important for auditory processing. The ongoing collection of ACHIEVE data with longitudinal MRI data will facilitate understanding of association between hearing loss and brain atrophy as well as the potential promising effects of hearing treatment in mitigating brain aging.

Abstract We investigated the cross-sectional associations of speech-in-noise performance with magnetic resonance imaging brain volumes among 588 cognitively normal participants (77±4 years, 56% female) from the Aging and Cognitive Health Evaluation in Elders Study (randomized trial embedded in the Atherosclerosis Risk in Communities (ARIC) Study) baseline in 2018-19 (N=427, with hearing loss) and ARIC (N=161, normal hearing) Visit 6/7 in 2016-17/2018-19. Central auditory processing was measured by Quick Speech-in-Noise (QuickSIN) test; range: 0 to 30, lower scores=worse performance. In models adjusted for demographic and disease covariates, every 5-point decrease in QuickSIN score was associated with smaller volumes of the temporal lobe overall (-0.07SD, 95% CI:-0.13,-0.01) as well as subregions including but not limited to those important for auditory processing (amygdala:-0.13SD, 95% CI:-0.21,-0.04; middle temporal gyrus:-0.08SD, 95% CI:-0.15,-0.00; superior temporal gyrus:-0.08SD, 95% CI:-0.15,-0.01). Further research is needed to understand the mechanisms underlying these observed associations.

AbstractAimWe investigate if periodontal disease is prospectively associated with cerebrovascular and neurodegenerative markers of dementia and Alzheimer's pathology.Materials and MethodsN = 1306 participants (Visit 5 mean age = 76.5 [standard deviation = 5.4] years) in the Atherosclerosis Risk in Communities study with completed dental exams at Visit 4 underwent brain magnetic resonance imaging scans at Visit 5 while N = 248 underwent positron emission tomography scans. Participants were classified as edentulous or, among the dentate, by the modified Periodontal Profile Class. Brain volumes were regressed on periodontal status in linear regressions. Cerebrovascular measures and β‐amyloid positivity were regressed on periodontal status in logistic regressions.ResultsPeriodontal disease was not associated with brain volumes, microhaemorrhages, or elevated β‐amyloid. Compared with periodontally healthy individuals, odds ratios [95% confidence interval] for all‐type infarcts were 0.37 [0.20, 0.65] for severe tooth loss and 0.56 [0.31, 0.99] for edentulous participants.ConclusionsWithin the limitations of this study, periodontal disease was not associated with altered brain volumes, microhaemorrhages, or β‐amyloid positivity. Tooth loss was associated with lower odds of cerebral infarcts.

Abstract Introduction We investigated the longitudinal relationship between cortical amyloid deposition, anxiety, and depression and the risk of incident mild cognitive impairment (MCI). Methods We followed 1440 community‐dwelling, cognitively unimpaired individuals aged ≥ 50 years for a median of 5.5 years. Clinical anxiety and depression were assessed using Beck Anxiety and Depression Inventories (BAI, BDI‐II). Cortical amyloid beta (Aβ) was measured by Pittsburgh compound B positron emission tomography (PiB‐PET) and elevated deposition (PiB+) was defined as standardized uptake value ratio ≥ 1.48. We calculated Cox proportional hazards models with age as the time scale, adjusted for sex, education, and medical comorbidity. Results Cortical Aβ deposition (PiB+) independent of anxiety (BAI ≥ 10) or depression (BDI‐II ≥ 13) increased the risk of MCI. There was a significant additive interaction between PiB+ and anxiety (joint effect hazard ratio 6.77; 95% confidence interval 3.58–12.79; P = .031) that is, being PiB+ and having anxiety further amplified the risk of MCI. Discussion Anxiety modified the association between PiB+ and incident MCI.