Cortical thickness comparison between PiB-positive and PiB-negative healthy control patients

S210

Poster Presentation’s P1

Tapio Viljanen2, Kjell Nagren2, Mony De Leon1, 1NYU Center for Brain Health, New York, New York, United States; 2Turku PET Centre, Turku, Finland; 3NYU School of Medicine, New York, New York, United States; 4 Cornell University, New York, New York, United States. Background: Having a parent affected with late-onset Alzheimer’s disease (LOAD) is a major risk factor among cognitively normal (NL) individuals. This study examines whether NL with LOAD-parents show preclinical evidence of brain AD, as reflected in increased fibrillar amyloid-beta (Aß) deposition on C-Pittsburgh Compound B (PiB)-PET, a major hallmark of AD pathology, and reduced glucose metabolism on 18 F-fluorodeoxyglucose (FDG)-PET, a proxy for neuronal dysfunction. Methods: Forty-five 40-80 year-old NL with 8F-FDG and 11C-PIB PET examinations were examined, including 15 NL with a maternal history (MH), 15 NL with a paternal history (PH), and 15 NL with negative family history of LOAD (NH). For all cases, the parents’ AD diagnosis was clinician certified. Automated regions-of-interest, statistical parametric mapping, voxelwise inter-modality correlations and logistic regression were used to examine cerebral-to-cerebellar PiB and FDG standardized uptake value ratios across groups. Results: Groups were comparable for clinical and demographical measures. The MH group showed higher PiB retention and lower metabolism in AD-regions compared to NH and PH, while the PH group showed milder PiB increases and no metabolic reductions compared to NH. Results remained significant controlling for age, gender, education and ApoE. Metabolism and PiB retention were negatively correlated locally in PCC, frontal and parieto-temporal regions in MHPY, whereas no correlations were observed in NH and PH. The combination of Aß deposition and metabolism improved group separation over either measure alone, yielding
65% accuracy for MH vs NH and PH (P < 0.05). Conclusions: Among NL with LOAD-parents, only MH show co-occurring Aß increases and hypometabolism in AD-vulnerable regions, suggesting that these subjects may be at a MHPY increased risk for AD than PH. Present findings may motivate further research on familial transmission and parent-of-origin effects in LOAD, and indicate a great need for early intervention trials targeting adult children of LOAD-parents.

Background: New diagnostic criteria have been developed for the detection of Preclinical AD using biomarkers in cognitively normal (CN) older adults. We implemented these criteria using an MRI biomarker previously demonstrated to be associated with Alzheimer’s disease (AD) dementia to test the hypothesis that individuals classified as Preclinical AD using this marker would be at elevated risk for cognitive decline consistent with symptoms of early AD. Methods: The ADNI dataset was interrogated for CN individuals. MRI data were processed to obtain cortical thickness measures using a previously published set of a priori regions of interest to derive a single composite measure known as the “AD-signature” (ADsig). Each individual was then classified as ADsig-Low consistent with Preclinical AD (> 1S.D. below the mean), ADsig-Average (within 1 S.D. of the mean), or ADsig-High (> 1 S.D. above the mean). A three-year Cognitive Decline outcome was defined a priori using change in CDRSum-ofBoxes and selected neuropsychological measures. We hypothesized that Preclinical AD individuals would be at elevated risk of Cognitive Decline. Results: Of the 125 individuals who were CN at baseline, the19 classified as Preclinical AD using the MRI biomarker were at markedly increased risk of Cognitive Decline, which developed in 21% of them compared with 7% of ADsig-Average and 0% of ADsig-High groups (p ¼ 0.03). A logistic regression model demonstrated that every 1 S.D. of cortical thinning was associated with a nearly tripled risk of Cognitive Decline (p ¼ 0.02). In addition, of those for whom baseline cerebrospinal fluid (CSF) data were available, 60% of the Preclinical AD group had CSF characteristics consistent with AD while 36% of the AD-sig Average and 19% of the AD-sig High groups had such CSF characteristics (p ¼ 0.1). Conclusions: The present data supports our hypothesis that this approach to the detection of Preclinical ADidentified in single NC individuals using this quantitative AD-signature MRI biomarker-may provide investigators with a population enriched for AD pathobiology and at relatively high risk for imminent cognitive decline consistent with prodromal AD.

P1-312

CORTICAL THICKNESS COMPARISON BETWEEN PIB-POSITIVE AND PIB-NEGATIVE HEALTHY CONTROL PATIENTS

Vincent Dore1, Pierrick Bourgeat1, Jurgen Fripp1, Oscar Acosta2, Gael Chetelat3, Cassandra Szoeke4, Kathryn Ellis5, Ralph Martins6, Victor Villemagne7, Colin Masters8, David Ames5, Christopher Rowe7, Olivier Salvado1, 1CSIRO Preventative Health National Research Flagship ICTC, The Australian e-Health Research Centre-BioMedical, Brisbane, Australia; 2Universit
e de Rennes 1, LTSI, Rennes, France; 3Department of Nuclear Medicine and Centre for PET, and Department of Medicine, University of Melbourne, Austin Hospital, Melbourne, Australia; 4CSIRO Parkville, Melbourne, Australia; 5Academic Unit for Psychiatry of Old Age, Department of Psychiatry, The University of Melbourne, St. Vincent’s Aged Psychiatry Service, St George’s Hospital, Melbourne, Australia; 6Centre of Excellence for Alzheimer’s Disease Research & Care, School of Exercise Biomedical and Health Sciences,, Perth, Australia; 7Austin Health, Melbourne, Australia; 8The Mental Health Research Institute, Melbourne, Australia.

P1-311

TESTING THE NEW DIAGNOSTIC CRITERIA FOR PRECLINICAL ALZHEIMER’S DISEASE: MRI BIOMARKER OF ALZHEIMER’S DISEASERELATED ATROPHY IN COGNITIVELY NORMAL INDIVIDUALS PREDICTS COGNITIVE DECLINE

Brad Dickerson1, David Wolk2, 1MGH/Harvard Medical School, Charlestown, Massachusetts, United States; 2University of Pennsylvania, Philadelphia, Pennsylvania, United States.

Background: Studies have showed that B-amyloid plaques are likely to only exhibit local affects on the cortex at early stages of the disease before or when early cognition impairments occur. Understanding when and where neurodegeneration starts in the cortex may provide insights into the pathogenesis of AD. In this study, we focus on elderly Healthy Control (HC). While the majority of subjects in this group have low neocortical PiB retention (PiB SURV< 1.5), some present high PiB retention (PiB SURV>¼1.5), which is often seen as prodromal AD. In this abstract, we first identify discriminating regions using AD/PiB-negative HC analysis to then find difference between PiB-positive HC and PiB-negative HC. Methods: 119 subjects underwent a MRI scan and a 11C-PIB PET scans as part of the

Poster Presentation’s P1 Australian Imaging, Biomarker and Lifestyle study. PiB scans were normalized using the standardized uptake value ratio (SUVR) method. In the subject cohort, 29 patients were AD, while the other 90 were HC; 33 of whom were PiB-positive and 57 were PiB-negative. We used a surface based approach to identify the regions where the cortical thickness in AD patients was significantly lower than PiB-negative HC. Individual mean cortical thicknesses were calculated in the significant patch of each AAL region. T-test analysis was performed between the 3 groups (AD, PiB-negative HC, PiB-positive HC) in each tessellated region while controlling for age. Results: In the discriminating regions of AD/PiB-negative (Figure 1), we showed that PiB-positive HC had significantly lower cortical thickness than PiB-negative HC in the hippocampus and amygdala region, in the precuneus and in the superior temporal gyrus in the left hemisphere Table 1). The regions in the right hemisphere showed the same patterns but were not significant. Conclusions: The results showed that atrophy patterns in PiB-positive HC group were similar to AD group but to a lesser extent, suggesting an effect of amyloid plaques early on in asymptomatic individuals.

P1-313

PREDICTION OF MCI CONVERTERS IN THE ADNI COHORT USING PATTERNS OF CORTICAL THINNING

Simon Eskildsen1, Vladimir Fonov1, Pierrick Coupe1, Lasse Ostergaard2, Louis Collins1, 1Montreal Neurological Institute, Montreal, Quebec, Canada; 2Aalborg University, Aalborg, Denmark. Background: Structural neuro-imaging is seen as one possible surrogate biomarker for diagnosing and predicting AD. However, image processing techniques have so far not been able to accurately predict conversion to AD in patients with MCI[1]. In this study we investigated the possibility of using patterns of cortical thinning for predicting AD in a group of subjects with MCI. Methods: 1.5T T1w MRI baseline data were selected from the ADNI database (AD ¼ 150, MCI ¼ 325, controls ¼ 190). Cortical thickness was automatically calculated using FACE[2] and mapped to an average cortical surface of 100 AD patients (template surface) [3]. Four MCI subgroups were constructed: MCIs that converted to AD within 12 months (MCI12¼56), 24 months (MCI24¼105), and 36 months (MCI36¼117) from baseline, and MCIs that did not progress to AD during

S211

Table Classification results and size of test sets (nl, n2) for each classifier. Classifier

nl

n2

Correct Rate Sensitivity Specificity McNemar’s test

MCI12 vs MCInc MCI24 vs MCInc MCI36 vs MCInc AD vs NC AD vs MCI MCI vs NC

28

104 61%

59%

62%

P ¼ 0.207

52

104 63%

79%

56%

P ¼ 0.272

58

104 64%

66%

63%

P ¼ 0.276

75 94 82% 75 162 67% 162 94 64%

84% 77% 65%

81% 62% 62%

P ¼ 0.004 P ¼ 0.013 P ¼ 0.032

three years (MCI non-converters ¼ 208). A statistical parametric map of differences in cortical thickness between MCI12 and MCI non-converters was constructed, thresholded at p ¼ 0.01, and filtered (Figure). The resulting thinning pattern on the template cortical surface was used as a mask to sample cortical thickness measurements with the purpose of classifying MCI subjects into converters and non-converters using linear discriminant analysis. The converters and non-converters were divided randomly into training and test sets of equal sizes and used in the classifier. McNemar’s chi-square test was used to assess whether the classifier performed better than a random classifier. Results: The table shows correct classification rates, sensitivity, and specificity for the classifier along with results of McNemar’s test (a ¼ 0.05). The correct classification rate of converters and non-converters was above 61%. However, none of the classifications performed significantly better than a random classifier. Classification of AD, MCI, and controls performed significantly better than a random classifier (p  0.03). Conclusions: Using patterns of characteristic cortical thinning in MCI converters compared to MCI non-converters demonstrated promising results for the prediction of patients with prodromal AD progressing towards clinically definite AD. The classification results are better than results of comparable methods recently published [1]. Still, the sensitivity of the method needs to be increased to be clinically applicable. References: [1] Cuingnet et al., NeuroImage; In Press. [2] Eskildsen et al., NeuroImage 2009; 45(3):713-721. [3] Fonov et al., NeuroImage 2011, 54(1):313-327.

P1-314

AMYLOID DEPOSITION PREDICTS REGIONAL AND QUANTITATIVE METABOLIC DECLINE IN PATIENTS WITH MILD ALZHEIMER’S DISEASE

Stefan F€orster1, Timo Grimmer2, Gjermund Henriksen1, Behrooz Yousefi1, Hans F€orstl2, Alexander Kurz2, Alexander Drzezga1, 1Dept. Nuclear Medicine, TU Munich, Munich, Germany; 2Dept. Psychiatry, TU Munich, Munich, Germany.

Figure. Left: Statistical map of cortical thinning in MCI12 compared to MCI non-converters rendered on the partially flattened AD template surface. Right: The resulting pattern after thresholding and filtering.

Background: Similar anatomical distributions of fibrillar amyloid deposition (measured by[11C]PIB-PET) and brain hypometabolism (measured by[18F]FDG-PET) were reported in numerous AD studies. However, there is a lack of longitudinal studies evaluating the causal relationships of these two different pathological markers in the same AD population. Therefore we aimed to evaluate the predictive value of amyloid deposition relating to regional and quantitative metabolic decline in the same AD patients during a 2-year follow-up. Methods: Fifteen patients with mild probable AD (mean age 68 6 8 yrs., 5 f) underwent