DT‐01–06: Baseline FDG‐PET and Volumetric MRI Predicts Alzheimer's Disease Conversion from Mild Cognitive Impairment: an ADNI Study
Kewei Chen,Cynthia Stonnington,Napatkamon Ayutyanont,Cole Reschke,Pradeep Thyyagura,Hillary Protas,Xiaofen Liu,Auttawut Roontiva,Stephanie Parks,Robert Bauer III,Wendy Lee,Robert Koeppe,William Jagust,Norman Foster,Michael Weiner,Adam Fleisher,Eric Reiman
DOI: https://doi.org/10.1016/j.jalz.2013.08.007
2013-01-01
Abstract:The feasibility of using imaging based biomarker data in the early stage of the AD (or preclinical phase of AD) such as Mild Cognitive Impairment (MCI) due to AD has been investigated in multiple studies recently. In this study, using data from Alzheimer's disease neuroimaging initiative (ADNI), we examine baseline measurement of FDG-PET, structural MRI each alone and in combination with cognitive tests in distinguishing individuals with MCI who converted to AD and those who did not aiming to establish their usefulness in as predictors for progression to AD. Total of 139 ADNI participants who were diagnosed as MCI were included in this study. Among them, 78 (75.8±7.0 years old) developed incident AD during the subsequent 36 months, and the remaining (75.3±8.0) did not during the same period. All subjects had baseline FDG-PET and MRI data. In addition to voxel-wise analysis for each of the two imaging modalities, FDG-PET hypometabolic convergence index (HCI) and MRI hippocampal (HCV) volume were used each separately or in combination together also with Alzheimer's Dementia Assessment Scale-modified (ADAS-mod) to examine the statistical power in distinguishing the two groups using Receiver Operating Curve (ROC). The two group did not differ in mean age, education and gender ratio. However, they differed significantly in APOE4 ratio (p=0.03) and cognitive tests (ADAS-MOD p=1.1e-9, AVLT-LTM p=3.7e-7, MMSE p=0.001). Voxel-based analysis for both baseline FDG PET and MRI VBM demonstrated reduced glucose uptake and regional gray matter volumes in converters compared to non-converters in the anterior cingulate, medial prefrontal cortex, posterior cingulate and precuneus regions. HCI, HCV and ADAS-mod each separately distinguished converters from non-converters (sensitivity/specificity =80%/64%, 78%/71% and 70%/64% for HCI, HCV and ADAS-mod). When combining these 3 indices together, the sensitivity and specificity were both increased and more balanced (sensitivity=82%, and specificity=80%). FDG PET measured glucose uptake, MRI measured hippocampal volume and ADAS-mod at baseline can all distinguish MCI converters from non-converters and with increased statistical power when combined. Our findings validate several previous studies that demonstrated imaging based biomarkers as predictors of future clinical decline.