How do we define amyloid positivity in an asymptomatic population ...

2 downloads 0 Views 157KB Size Report
CLINICAL PET EXAMINATIONS. Brian Rubin1, Yi Su1, Jonathan McConathy1, Tyler Blazey1,. Rosana Ponisio1, Jin Weng1, Russ Hornbeck1, Anne Fagan1,.
P6

Oral Sessions: IC-O2: Normal Aging and Early Detection of Alzheimer’s Disease

ORAL SESSIONS: IC-02 NORMAL AGING AND EARLY DETECTION OF ALZHEIMER’S DISEASE IC-O2-01

HOW DO WE DEFINE AMYLOID POSITIVITY IN AN ASYMPTOMATIC POPULATION? COMPARISON OF CSF, QUANTITATIVE PET AND CLINICAL PET EXAMINATIONS

Brian Rubin1, Yi Su1, Jonathan McConathy1, Tyler Blazey1, Rosana Ponisio1, Jin Weng1, Russ Hornbeck1, Anne Fagan1, Chengjie Xiong1, John Morris2, Tammie Benzinger1, 1Washington University School of Medicine, St. Louis, Missouri, United States; 2 Washington University, St. Louis, Missouri, United States. Contact e-mail: [email protected] Background: It has been proposed that the first stage of Alzheimer’s disease (AD) is marked by amyloidosis, which can be measured by both cerebrospinal fluid (CSF Ab42) and PET, including [11C]Pittsburgh Compound B (PIB) and18F-florbetapir (AV45), for which a clinical reading system has been developed to identify moderate to frequent plaques in evaluating patients with dementia. This study compares each of these markers in a largely asymptomatic population. Methods: Thirty-six participants (mean age ¼ 65.1 yrs; 35 CDR 0, 1 CDR 0.5) were prospectively enrolled in this study. Each participant completed MR, PIB-PET, and AV45-PET. Twenty-nine had all PET data available for analysis. FreeSurfer was used to segment brain MRIs for quantitative analysis of the PET scans. For PIB, SUVR and mean cortical binding potential (MCBP) were calculated with cerebellar cortex as reference. For AV45, SUVR was computed using the entire cerebellum as reference. PIB MCBP of 0.18 was used to define PIB positivity. Two trained radiologists (JMc & RP) performed clinical ratings based on the AV45 scans using FDA approved interpretation criteria. Thirty-three cases had both clinical ratings. Twenty-eight participants had available

Fig 1. MCSUVR PiB vs. AV45 (trendline for PiB MCSUVR >1.2)

Fig 2. PiB MCSUVR vs. CSF AB42.

CSF data; a cutoff value of 550 was established for CSF AB42 positivity. Results: Strong correlations were observed between PiB SUVR and AV45 SUVR both regionally (r¼0.835 for precuneus) and globally (r¼0.757 for MCSUVR). For cases with intermediate and positive PIB MCSUVR (1.2 and higher) these correlations were r¼0.968 globally and r¼0.960 for the precuneus. Quantitative PIB analysis identified 7 amyloid+ cases while the readers reported 4 amyloid+ cases and 2 cases, respectively. All clinical positive scans were also quantitatively positive. For CSF AB42, six were rated as positive, of which two were quantitatively positive by PIB MCBP. Eighteen were rated negative by CSF AB42, three of which were quantitatively positive by PIB. Conclusions: High correlations were observed between PET amyloid imaging tracers. Discrepancies were found in 3 of 33 cases comparing clinical/quantitative PET, and 7 of 24 comparing CSF/PET. Discrepancies cannot be explained simply by differences in test sensitivities. Further studies in these asymptomatic populations are warranted, especially as clinical trials are designed to target this cohort.

IC-O2-02

DETECTING COGNITIVE EVIDENCE OF PRECLINICAL ALZHEIMER’S DISEASE

Dorene Rentz1, Trey Hedden2, Elizabeth Mormino3, Rebecca Amariglio4, John Becker3, Aaron Schultz5, Gad Marshall6, Reisa Sperling1, Keith Johnson4, 1Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, United States; 2Martinos Center for Biomedical Imaging at Harvard University, Charlestown, Massachusetts, United States; 3 Massachusetts General Hospital, Boston, Massachusetts, United States; 4 Massachusetts General Hospital/ Harvard Medical School, Boston, Massachusetts, United States; 5Massachusetts General Hospital, Charlestown, Massachusetts, United States; 6Brigham and Women’s Hospital, Center for Alzheimer Research and Treatment, Boston, Massachusetts, United States. Contact e-mail: [email protected] Background: Recently published commentaries propose that the time to intervene in Alzheimer’s disease (AD) is during a preclinical stage when clinically normal individuals (CN) demonstrate evidence of amyloidosis, neuronal injury and very subtle cognitive decline. We sought to determine whether tests of memory and executive function in CN subjects relate to synaptic integrity using 18 F-flourodeoxyglucose (FDG PET) and amyloid deposition with C 11 Pittsburgh Compound B (PiB PET). Methods: CN subjects (N¼129), ages 65 to 87 (mean age ¼ 73.7 6 5.9; MMSE ¼ 29.1 6 0.9) with Clinical Dementia Rating scores of 0 were administered executive function and memory tests, along with PiB-PET and FDG-PET. Factor scores for episodic memory (EM) and executive function (EF) were created from a confirmatory factor analysis. Multiple linear regressions explored the relationship between factor scores, PiB retention and FDG metabolism and the interaction of these terms with education, a proxy of cognitive reserve. Results: EM was inversely related to PiB retention (beta ¼ -0.856, p¼0.035) and positively related to FDG metabolism (beta ¼2.970, p¼0.006) in an aggregate of cortical regions vulnerable to amyloid deposition in AD. PiB and FDG measures were significant independent predictors of EM. EF was significantly related to FDG metabolism (beta ¼2.449, p¼0.014) but not to PiB retention. While PiB and FDG were not significantly related (r¼ -0.074, p¼0.402), the main effect of education with FDG metabolism was significant (R 2 ¼ 0.238, p