Francis E. Cambronero

Cerebrospinal fluid β-amyloid 42 and neurofilament light relate to white matter hyperintensities

White matter hyperintensities (WMHs) are associated with poorer brain health, but their pathophysiological substrates remain elusive. To better understand the mechanistic underpinnings of WMHs among older adults, this study examined inxa0vivo cerebrospinal fluid biomarkers of β-amyloid42 deposition (Aβ42), hyperphosphorylated tau pathology, neurodegeneration (total tau), and axonal injury (neurofilament light [NFL]) in relation to log-transformed WMHs volume. Participants free of clinical stroke and dementia were drawn from the Vanderbilt Memory & Aging Project (nxa0= 148, 72 ± 6xa0years). Linear regression models adjusted for age, sex, race/ethnicity, education, intracranial volume, modified Framingham Stroke Risk Profile (excluding points assigned for age), cognitive diagnosis, and APOE-ε4 carrier status. Aβ42 (βxa0=xa0-0.001, pxa0= 0.007) and NFL (βxa0= 0.0003, pxa0= 0.01) concentrations related to WMHs but neither hyperphosphorylated tau nor total tau associations with WMHs reached statistical significance (p-values > 0.21). In a combined model, NFL accounted for 3.2% of unique variance in WMHs and Aβ42 accounted for an additional 4.3% beyond NFL, providing novel evidence of the co-occurrence of at least 2 distinct pathways for WMHs among older adults, including amyloid deposition and axonal injury.

Subclinical Compromise in Cardiac Strain Relates to Lower Cognitive Performances in Older Adults

Background Global longitudinal strain (GLS), reflecting total shortening of the myocardium during the cardiac cycle, has emerged as a more precise myocardial function measure than left ventricular ejection fraction (LVEF). Longitudinal strain may be selectively affected in subclinical heart disease, even in the presence of normal LVEF. This study examines subclinical cardiac dysfunction, assessed by GLS and LVEF, and cognition among older adults. Methods and Results Vanderbilt Memory and Aging Project participants who were free of clinical dementia, stroke, and heart failure (n=318, 73±7 years, 58% male) completed neuropsychological assessment and cardiac magnetic resonance to quantify GLS and LVEF. Linear regression models related GLS and LVEF to neuropsychological performances, adjusting for age, sex, race/ethnicity, education, Framingham Stroke Risk Profile, cognitive diagnosis, and APOE*ε4 status. Models were repeated with a cardiac×cognitive diagnosis interaction term. Compromised GLS (reflected by higher values) related to worse naming (β=−0.07, P=0.04), visuospatial immediate recall (β=−0.83, P=0.03), visuospatial delayed recall (β=−0.22, P=0.03), and verbal delayed recall (β=−0.11, P=0.007). LVEF did not relate to worse performance on any measure (P>0.18). No diagnostic interactions were observed. Conclusions Our study results are among the first to suggest that compromised GLS relates to worse episodic memory and language performance among older adults who are free of clinical dementia, stroke, and heart failure. Subclinical cardiac dysfunction may correlate with cognitive health in late life, even when LVEF remains normal. The results add to growing evidence that GLS may be a more sensitive and preferred method for quantifying subclinical changes in cardiac function.

APOE genotype modifies the association between central arterial stiffening and cognition in older adults

Arterial stiffening is associated with cognitive impairment and prodromal Alzheimers disease. This studyxa0tested the interaction between arterial stiffening and an Alzheimers disease genetic risk factor (apolipoprotein E [APOE] genotype) on cognition among older adults. Vanderbilt Memory & Aging Project participants with normal cognition (nxa0= 162, 72 ± 7xa0years, 29% APOE-ε4 carrier) and mild cognitive impairment (nxa0= 121, 73 ± 8xa0years, 42% APOE-ε4 carrier) completed neuropsychological assessment and cardiac MRI to assess aortic stiffening using pulse wave velocity (PWV, m/s). Linear regression models stratified by cognitive diagnosis related aortic PWVxa0× APOE-ε4 status to neuropsychological performances, adjusting for demographic and vascular risk factors. PWVxa0× APOE-ε4 related to poorer performance on measures of lexical retrieval (βxa0=xa0-0.29, pxa0= 0.01), executive function (βxa0=xa0-0.44, pxa0= 0.02), and episodic memory (βxa0=xa0-3.07, pxa0= 0.02). Among participants with higher aortic PWV, APOE-ε4 modified the association between central arterial stiffening and cognition, such that carriers had worse performances than noncarriers. Findings add to a growing body of evidence for APOE-vascular interactions on cognition in older adults and warrant further research into less heart-healthy cohorts where the association between PWV and cognition among older adults might be stronger.

POSTERIOR COMMUNICATING ARTERY VARIATION IN THE CIRCLE OF WILLIS IS ASSOCIATED WITH COMPROMISED TEMPORAL LOBE HEMODYNAMICS IN OLDER ADULTS

which mean diffusivity(MD), fractional anisotropy(FA) and white matter hyperintensity(WMH) were calculated. Cerebrospinal fluid was obtained for 63 subjects and Ab, tau and phosphorylated-tau levels were measured. Subjects were classified as low-binders when rs6971 homozygous and high-binders when rs6971 homozygous. Chi-square and t-tests were used for demographic description whereas linear models adjusted for covariates were used to assess the effect of rs6971 on the available phenotypes. Results: The rs6971 minor-allele frequency in our study population is 0.30 (48,8% CC,41.7% CT,9.4% TT) (Figure 1), which is in accordancewith frequencies described for Caucasian populations in public databases. Differences in gender, diagnosis and APOE-e4 status were not observed across rs6971 genotypes. The rs6971 was not associated with any cognitive score as well as with none of the CSF or imaging biomarkers tested here. Conclusions:The study population under investigation is genetically representative of the Caucasian population –for the rs6971–and the SNP showed no effect in any dementia-related phenotypes analyzed here. Results confirm that the cohort is appropriate for neuroimaging studies linking dementia and neuroinflammation.

AORTIC WALL STRUCTURE IS DISORGANIZED AND FRAGMENTED IN APP/PSEN1 COMPARED TO AGE-MATCHED WILDTYPE MICE

than the number of p-Tau+ neurons does (Table 1). The frequencies of PTMs, the percent overlap between PTMs, and significance of difference between early and late Braak stages are depicted in Table 2. Conclusions:Caspase-6 cleavage at Asp401 facilitates the formation of paired helical filaments (PHF). The PTMs that trigger this caspase activity are unclear. These data suggest that acetylation of tau at Lys274 correlates better with cleavage by caspase-6 than phosphorylation of tau at Ser202/Thr205 does. While there is colocalization between p-Tau and c-Tau, the majority of neurons positive for both p-Tau and c-Tau are also positive for a-Tau, suggesting that caspase-mediated cleavage of tau in these neuronsmay be primarily induced by acetylation, rather than phosphorylation. These preliminary data highlight that the acetylation of tau represents a key step in AD pathobiology and warrants further study.

ANATOMICAL VARIATION IN THE CIRCLE OF WILLIS IS ASSOCIATED WITH WHITE MATTER HYPERINTENSITIES IN OLDER ADULTS

Entorhinal -0.26 (.17) 0.137 -0.61, 0.09 -0.17 (.23) 0.461 -0.63, 0.29 Fusiform -0.21 (.09) 0.020 -0.38, -0.04 -0.16 (.12) 0.182 -0.39, 0.08 Parahippocampal -0.20 (.14) 0.157 -0.48, 0.08 0.01 (.18) 0.946 -0.38, 0.35 Cingulate Isthmus -0.25 (.11) 0.026 -0.47, -0.03 -0.10 (.14) 0.477 -0.39, 0.18 Posterior Cingulate -0.21 (.10) 0.042 -0.42, -0.01 -0.17 (.14) 0.229 -0.44, 0.11 Precuneus -0.48 (.07) 0.000 -0.63, -0.34 -0.35 (.09) 0.000 -0.53, -.016

CARDIOVASCULAR FUNCTION IS COMPROMISED IN APP/PSEN1 MICE COMPARED TO AGE-MATCHED WILDTYPE MICE

Body weight (BW) (g) 452 6 10 715 ± 22 ## 754 ± 24 ## 329 ± 7 ** 465 ± 7 ** ## 470 ± 8 ** ## Aortic wall collagen (%) 8.0 6 1.2 8.5 6 1.3 9.6 6 0.7 12.1 6 2.3 17.7 ± 0.7 ** # 20.8 ± 1.7 ** MWM, time spent to find a platform (sec) 17 6 4 18 6 4 19 6 2 16 6 2 24 6 3 31 ± 1 * ## OFT, total distance traveled day 1 first 5 mins (m) 6.7 6 0.4 3.9 ± 0.4 ## 6.9 6 0.3 5.2 ± 0.3 ## OFT, vertical activity day 1 first 5 mins (n) 329 6 25 106 ± 13 ## 345 6 15 229 ± 29 ** ##

APOE GENOTYPE INFLUENCES HOW CEREBRAL BLOOD FLOW AND VASOREACTIVITY PREDICT NEUROPSYCHOLOGICAL DECLINE OVER AN 18-MONTH FOLLOW-UP: THE VANDERBILT MEMORY AND AGING STUDY

other groups, including sAD.Conclusions:Our results suggest a high frequency of microhemorrhages in DSAD cases in the occtx, more so than sAD. This along with the increased amount of CAA suggests that cerebrovascular pathology may be an under-recognized, yet significant contributor to aging and the development of dementia in peoplewith DS.We are currently in the process of quantifying microbleeds using MRI in the occtx in our aging DS cohort to link this pathology with cognitive decline. Funding from NIH/NICHD R01HD064993.

INTRACRANIAL ARTERY LUMEN DIAMETER RELATES TO CEREBRAL BLOOD FLOW AND CEREBROVASCULAR REACTIVITY IN MCI

independent steps. First, for each scan session (subject), the normalized signal intensity for each voxel was assessed for fit to the same session’s CSF signal across the 4 sequences (T, T, FLAIR, PD) (Fig. 2). Second, clusters ( 4 contiguous voxels) with positive vCSF regression coefficients were submitted to the morphological assessment component of the algorithm (Fig 3). For clusters surviving linearity and volume thresholds, morphological features including linearity, width, and volume of each cluster (ePVS) were extracted (Fig 3). Results:A significant correlation was found between visual counts and 1) clusters detected in the same slice (r(26)1⁄4 0.65, p<0.001; r(26)1⁄4 0.69, p<0.001; r(26)1⁄40.54, p<0.01) and 2) with total burden volume (r(26)1⁄4 0.58, p<0.01) and total ePVS number detected by mMAPS (r(26)1⁄4 0.76, p<0.01, Figure 4C). Average ePVS was 24.6 (range 3 to 71) and total burden volume (SD) 303.0 (267.74) mm per dataset. Average width of ePVS was 3.12 mm (min 1.7, max 13.5). Whole brain counts were 7.1 times that of a single slice, a nearly 100-fold increase in the range of the measured proxy variable for burden. In this limited dataset, no significant relationships were observed between

ABNORMAL CARDIAC STRUCTURE AND FUNCTION MEASURES ARE ASSOCIATED WITH INCREASED PERIVASCULAR SPACES IN OLDER ADULTS

Background:Three different 18F-labeled radiotracers for b-amyloid PET are available, which demonstrated similar performance as 11C-PiB. The leaflets of the three tracers indicate 3 different visual criteria for PET reporting based on controlled trials, but it’s not yet verified in clinical practice if these criteria are equivalent in term of diagnostic accuracy or if there is a reading criteria better than an other. The aim of this study is to determine the inter-rater variabilityof the visual interpretation of 18F-Florbetapir PET/CT among 6 independent readers for 3 different visual criteria recommended by the 3 companies. We studied the intra-reader variability. Methods: We analyzed 252 18F-Florbetapir PET/CT (228 patients with cognitive impairment and 24 normal cognitive control), obtained from subjects participating at INDIA-FBP study and performed in 2 PET centers. Six independent blinded readers (2 experts in amyloid imaging and 4 who followed reading educational programs but without direct clinical experience) examined each PET-scan three times according to 3 different visual reading criteria, classifying them with a subjective confidence level of 5 point score (4:definitely positive; 3:doubt positive; 2:doubt; 1:doubt negative or 0:definitely negative) and then attributing a numeric score for each region analyzed that formed anoverallobjective score. We compared the mean of the results obtained by the 2 experts readers (considered as “gold standard”) with those obtained by the others 4 readers. Results: Substantial to almost perfect agreement was observed for visual interpretation by the six readers’ confidence level (mean weighted Kappa) for all 3 radiotracers: mwK1⁄40.82 for 18F-Florbetapir lecture; mwK1⁄40.84 for 18F-Florbetaben; mwK1⁄40.85 for 18F-Flutemetamol. Overall objective score showed a low inter-readers variability: mwK1⁄40.73 for 18F-Florbetapir lecture; mwK1⁄40.82 for 18F-Florbetaben and mean Spearman’s rho1⁄40.82 for 18F-Flutemetamol. Comparison between the “gold standard” showed a high confidence level: wK ranging 0.80-0.94. Intra-rater agreement has Spearman’s rho ranging 0.89-1.00 (p<0.001) about confidence level and Spearman’srho ranging 0.75-0.90 (p<0.001) about overall score. Conclusions: There is a high inter-rater and intra-reader agreement using the three different visual criteria for amyloid images. All the criteria proposed are useful and valid to determine the positivity or negativity of amyloid deposition and when compared with a “gold standard” there aren’t statistical differences between the three criteria.