Early frame of PiB and FDG in autosomal dominant Alzheimer's disease: Similarity, discrepancy, and clinical implication


      Alzheimer’s disease (AD), the leading cause of dementia in the elderly, can affect individuals in their thirties in autosomal dominant form. The Imaging Core of the Dominantly Inherited Alzheimer Network (DIAN) aims to characterize transition from preclinical to symptomatic disease using imaging biomarkers. Decreases in cerebral glucose metabolism in the parietal lobe are detectable 10 years before the estimated year of symptom onset (EYO) (Benzinger, Blazey et al., 2013) and may represent synaptic dysfunction. In sporadic AD, studies have shown that early perfusion frames of amyloid imaging with [11C]-Pittsburgh Compound B PiB (ePiB) correlate well with glucose metabolism (Rostomian et al., 2011). Here, we evaluated whether ePiB is a reasonable surrogate marker for synaptic dysfunction, in comparison to glucose metabolism hypometabolism, and how ePiB changes with the disease progression.


      DIAN participants (n=110), including 65 asymptomatic and symptomatic mutation carriers (MC), underwent full dynamic PiB-PET and also had [18F]-fluorodeoxyglucose (FDG) PET and volumetric brain MRI. The MRI was used to register the PET images. A standardized uptake value ratio (SUVR) from MR segmented PiB and FDG regions. An ePiB image with 1-9 min time frames was selected. Voxel-wise spatial correlation between FDG and ePiB was performed for each participant. The mutation and cognitive status were taken into account in the analyses. For each imaging modality, relationship with EYO was evaluated with linear mixed models on specific regions such as inferior parietal and precuneus cortices.


      FDG and ePiB were visually similar and showed high spatial correlation with an average of 0.8±0.04 regardless of the mutation or cognitive status. As we have previously found, the association between FDG and EYO significantly differs between MC and non-carrier groups (p-value<0.001 and p-value<0.01 for inferior parietal and precuneus, respectively). However, these associations were not significant between ePiB and EYO.


      Our findings show that ePiB is strongly correlated with FDG within the same individual. However, ePiB does not display the same sensitivity as FDG to reflect disease progression in this population. Further studies are needed to fully determine the utility of ePiB measurements in clinic.