MICROGLIAL ACTIVATION AND TAU BURDEN PREDICT COGNITIVE DECLINE IN ALZHEIMER’S DISEASE
Maura Malpetti,Rogier A. Kievit,Luca Passamonti,P. Simon Jones,Kamen A. Tsvetanov,Timothy Rittman,Elijah Mak,Nicolas Nicastro,W. Richard Bevan-Jones,Li Su,Young T. Hong,Tim D. Fryer,Franklin Aigbirhio,John T. O'Brien,James B. Rowe
DOI: https://doi.org/10.1093/brain/awaa088
2019-01-01
Alzheimer s & Dementia
Abstract:Neurodegeneration, tau pathology, and neuroinflammation are key etiopathogenic processes of Alzheimer's disease (AD). Understanding whether these features predict cognitive decline in AD is crucial to developing prognostic measures. We studied the independent prognostic value of atrophy (structural MRI), tau burden ([18F]AV-1451 PET), and neuroinflammation ([11C]PK-11195 PET) in predicting longitudinal cognitive decline in patients with AD dementia, and amyloid positive Mild Cognitive Impairment (MCI+). 29 healthy controls (HC) and 26 patients with AD pathology (12 AD and 14 MCI+) underwent a cross-sectional assessment with structural MRI, dynamic [11C]PK-11195, and [18F]AV-1451 PET imaging as part of the NIMROD study (NeuroImaging of MemoRy and Other related Disorders). An annual cognitive follow-up (using the Revised Addenbrooke's Cognitive Examination (ACE-R)) was performed for two years. Regional grey matter volumes were derived from structural MRI in AD-specific regions, while [18F]AV-1451 and [11C]PK-11195 non-displaceable specific binding were calculated for these volumes to quantify tau pathology and neuroinflammation. To reduce dimensionality of the data, we used a Principal Component Analysis (PCA) on each imaging method. A Latent Growth Curve model (LGCM) was used to test for the independent predictive value of the PCA-derived neuroimaging measures on longitudinal cognitive decline (Figure 1). PCA identified a global component for grey matter volumes, and two components for each PET ligand: one in anterior temporal regions, and another in posterior temporo-parietal regions (Figure 2). The LGCM using MRI (χ2(4)=2.82, p=0.59, RMSEA=0.00, CFI=1.00) showed a positive association between grey-matter volumes and longitudinal ACE-R scores (Std all=0.56, p<0.001). LGCM with [11C]PK-11195 (χ2(6)=1.50, p=0.96, RMSEA=0.00, CFI=1.00) indicated a positive association with 2-year cognitive decline in both anterior (Std all=−0.42, p<0.001) and posterior (Std all=−0.35, p=0.016) components. Furthermore, in LGCM with [18F]AV-1451 (χ2(6)=10.63, p=0.10, RMSEA=0.12, CFI=0.98), longitudinal cognitive performance correlated negatively with values in the posterior (Std all=−0.61, p=0.01) and anterior (Std all=−0.43, p<0.001) components.