Poster Presentations: P2P382

P2-101 AMYLOID AND VASCULAR COGNITIVE

IMPAIRMENT: A PILOT STUDY OF FREQUENCY

AND IMPACT

Elizabeth Dao, Ging-Yuek Robin Hsiung, Vesna Sossi, Claudia Jacova,

Teresa Liu-Ambrose, University of British Columbia, Vancouver, British

Columbia, Canada. Contact e-mail: elizabeth.dao@hiphealth.ca

Background: Traditionally, Alzheimer’s disease (AD) and vascular cogni-

tive impairment (VCI) were considered to be distinct and unrelated; how-

ever, increasing evidence is demonstrating an overlap between AD and

VCI pathology. As such, the current criteria for the clinical diagnosis of

VCI may not distinguish those with cognitive impairment due to subcortical

ischemic small vessel disease from those with cognitive impairment due to

mixed vascular and AD pathology (mixed dementia - MD). Furthermore, it

is unclear how co-existing amyloid pathology may affect cognitive function

in peoplewith VCI. The purpose of this pilot study was to determine the fre-

quency of MD in patients diagnosed with VCI. In addition, we investigated

how co-existing amyloid pathology may affect cognitive function in people

with VCI.Methods:We conducted a planned sub-analysis of a randomized

controlled trial investigating the effect of targeted aerobic exercise training

on cognitive function in people with VCI. Elevan participants - 8 partici-

pants with VCI and 3 normal controls - underwent a PiB-PET scan to esti-

mate amyloid burden. Participants with VCI who exhibited PiB uptake 2

standard deviations above the mean of controls were considered to be

PiB-positive and to haveMD.To determine the effect of co-existing amyloid

pathology on cognitive function we collected the following measures: 1)

ADAS-Cog; 2) EXIT-25; and 3) a) Digits Forward and Backwards Test,

b) Stroop-Colour Word Test, and c) Trail Making Test (Part B-Part A). To

determine the associations between PiB uptake and cognitive function we

conducted correlational analysis using Pearson correlation coefficients.

Results: Five (62.5%) participants with VCI were PiB-positive and 3

(37.5%) participants were PiB-negative. Increased PiB retention was signif-

icantly correlated with reduced cognitive performance in the ADAS-Cog

(r¼0.849, p¼0.008) and the Trail Making Test (r¼0.861, p¼0.006).

Conclusions: Neuroimaging with PiB-PET showed 62.5% of those with

a clinical diagnosis of VCI have co-existing amyloid pathology. Critically,

increased amyloid binding was associated with increased cognitive deficits

in domains typically affected by both AD and VCI. Thus, our results suggest

that those with VCI and co-existing amyloid pathology demonstrate more

diverse cognitive deficits. M ore research is needed to develop reliable

and valid measures for the diagnosis of MD.

P2-102 CORTICAL BETA-AMYLOID AND

MICROSTRUCTURAL PROPERTIES OF THE

CORPUS CALLOSUM IN PEOPLE WITH

MILD-TO-MODERATE ALZHEIMER’S DISEASE

Kristian Steen Frederiksen1, Nina Reislev2, SteenHasselbalch3, Ian Law4,

Karine Madsen5, Tim B. Dyrby2, Ellen Garde6, Hartwig Siebner7,

Gunhild Waldemar8, 1Rigshospitalet – Dept. of Neurology, Copenhagen,

Denmark; 2Danish Center for Magnetic Resonance, Copenhagen,

Denmark; 3Neurobiology Research Unit N9201, Copenhagen, Denmark;4Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark;5Copenhagen University Hospital, Copenhagen, Denmark; 6Copenhagen

University Hospital, Hvidovre, Denmark; 7Danish Research Center for

Magnetic Resonance, Copenhagen, Denmark; 8Rigshospitalet -

Copenhagen University Hospital, Copenhagen, Denmark.

Contact e-mail: Kristian.Steen.Frederiksen@regionh.dk

Background: Several studies have reported microstructural changes of the

corpus callosum (CC) in AD. The pathological basis of these changes re-

mains uncertain.We investigated whether fractional anisotropy (FA) as

a marker of microstructural properties of the CC assessed by diffusion-

weighted imaging (DWI) is associated with cortical beta-amyloid measured

by amyloid imaging.Methods: Patients with mild to moderate Alzheimer’s

dementia (according to NINCDS-ADRDA criteria and ICD-10 criteria),

were recruited from the Memory Clinic (Copenhagen University Hospital,

Rigshospitalet). Patients underwent 11C-PiB-PET for assessment of cortical

beta-amyloid and MRI (Siemens Magnetom Trio 3T) including DWI for

quantification of white matter microstructure. DWI preprocessing included

correction of susceptibility and movement artefacts. Following, DWI data

were fitted to the tensor model, and FAmaps were created.Masks wereman-

ually delineated on the midsagittal slice of the FA maps, encompassing the

anterior 1/6th of the CC (projecting to prefrontal cortex) and the posterior 1/

4th (projecting to parietal, temporal, occipital cortex) based on the assumed

cortical projections of callosal axons. Following, filters which included vox-

els with FA > 0.4 and with the 1st eigenvector oriented left-right were ap-

plied, to ensure that only voxels belonging to the CC were included. For

quantification of beta-amyloid standard uptake value ratios (SUVR) from

prefrontal cortex, and parietal, occipital and temporal cortex were used in

analysis. Multivariable linear regression analysis with FA in anterior and

posterior CC (in two different models) as outcome variable and PiB

SUVR values as predictor variable, and age, gender and MMSE score as

covariates, were carried out. Results: In total, 32 patients (Age, years:

69.2 (6 7.3); Gender (f/m): 14/18; MMSE: 24.8 (3.5)) were included in

the study. Mean FA across subjects for anterior CC was 0.74 (6 0.038)

and for posterior CC 0,71 (6 0.068). No significant associations between

FA in anterior CC and prefrontal 11C-PiB uptake or between FA in posterior

CC and parietal-temporal-occipital cortex 11C-PiB uptake was found.

Conclusions: Cortical deposition of beta-amyloid does not seem to be the

underlying mechanism of microstructural changes in the CC in AD patients.

However, whether beta-amyloid may affect the CC through cortical atrophy

remains undecided. Future studies should assess whether cortical atrophy

may contribute.

P2-103 CORRELATIONOF 18F-FDG PET FINDINGSWITH

NEUROPSYCHOLOGICAL RESULTS IN

POSTERIOR CORTICAL ATROPHYAND TYPICAL

ALZHEIMER’S DISEASE

Byeong C. Kim1, Seong-Min Choi2, Hyun Jung Jung3, Ho-Chun Song4,

Woong Yoon5, 1Dept. of Neurology, Chonnam National University Medical

School, Gwangju, South Korea; 2Chonnam National Univ. Medical School,

Gwaongju, South Korea; 3Chonnam National Univ. Hosp., Gwangju, South

Korea; 4Chonnam National University Medical School, Gwangju, South

Korea; 5Chonnam National University Medical School, Gwangju, South

Korea. Contact e-mail: byeckim@gmail.com

Background: Posterior cortical atrophy (PCA) is a progressive degenera-

tive brain disease where the lesion is particularly focused at the cerebral re-

gion responsible for visual processing. The underlying cause of PCA is

Alzheimer’s disease (AD) in the majority of cases. Despite being caused

by the same disease process, the clinical characteristics of PCA and typical

AD are very different. This study aimed to know the regions of glucose hy-

pometabolism and whether each region is correlated with the neuropsycho-

logical results in PCA and typical AD. Methods: Seven patients with PCA

and 11 patients with typical AD underwent 18 F-FDG PET and neuropsy-

chological test (Seoul Neuropsychological Screen Battery), and 18 F-

FDG PETalso was made in 18 age- and sex- matched normal elderly people

for control group. Imaging data were analyzed using statistical parametric

mapping software (SPM2) with MATLAB version 6.5 for group analysis,

and using FALBA software for regional activity.We also compare neuropsy-

chological data between PCA and typical AD using PASW 18 software.

Results: As compared to normal subjects, patient with PCA showed signif-

icant cerebral glucose hypometabolism in the occipital, parietal, and poste-

rior part of temporal lobe, and patients with typical AD have significant

reduction of cerebral glucose metabolism in cingulated gyrus, precuneus,

temporal, and frontal lobe (uncorrected p <0.001). Global cerebral glucose

metabolism of PCA is similar with typical AD. In neuropsychological re-

sults, RCFT copy and the Stroop Word test are more severe impaired in

PCA than in typical AD (p<0.05). In contrast, typical AD had a significantly

lower score in comparison to PCA in digit span forward and verbal memory

(p < 0.05). Conclusions: The regional glucose hypometabolism in patients

with PCA seems to be a different pattern as compared with that in typical