Neuropathologic assessment of participants in two multi-center longitudinal observational studies: the Alzheimer Disease Neuroimaging Initiative (ADNI) and the Dominantly Inherited Alzheimer Network (DIAN)

Abstract

It has been hypothesized that the relatively rare autosomal dominant Alzheimer disease (ADAD) may be a useful model of the more frequent, sporadic, late-onset AD (LOAD). Individuals with ADAD have a predictable age at onset and the biomarker profile of ADAD participants in the preclinical stage may be used to predict disease progression and clinical onset. However, the extent to which the pathogenesis and neuropathology of ADAD overlaps with that of LOAD is equivocal. To address this uncertainty, two multicenter longitudinal observational studies, the Alzheimer Disease Neuroimaging Initiative (ADNI) and the Dominantly Inherited Alzheimer Network (DIAN), leveraged the expertise and resources of the existing Knight Alzheimer Disease Research Center (ADRC) at Washington University School of Medicine, St. Louis, Missouri, USA, to establish a Neuropathology Core (NPC). The ADNI/DIAN-NPC is systematically examining the brains of all participants who come to autopsy at the 59 ADNI sites in the USA and Canada and the 14 DIAN sites in the USA (eight), Australia (three), UK (one) and Germany (two). By 2014, 41 ADNI and 24 DIAN autopsies (involving nine participants and 15 family members) had been performed. The autopsy rate in the ADNI cohort in the most recent year was 93% (total since NPC inception: 70%). In summary, the ADNI/DIAN NPC has implemented a standard protocol for all sites to solicit permission for brain autopsy and to send brain tissue to the NPC for a standardized, uniform and state-of-the-art neuropathologic assessment. The benefit to ADNI and DIAN of the implementation of the NPC is very clear. The NPC provides final "gold standard" neuropathological diagnoses and data against which the antecedent observations and measurements of ADNI and DIAN can be compared.

Keywords: PET-PiB amyloid imaging; autosomal dominant Alzheimer disease; late-onset Alzheimer disease; neuropathologic diagnostic criteria; neuropathologic heat map.

Figure 1

Comparison of histology and immunohistochemistry in formalin-fixed, paraffin-embedded (A–C) and snap-frozen brain tissue (D–F). Mild to moderate neuronal loss and gliosis in the right frozen occipital lobe (D) are comparable in the left formalin-fixed occipital lobe (A) (hematoxylin and eosin; Brodmann area 18). Amyloid burden (Aβ plaques and cerebral amyloid angiopathy; Aβ(10D5) immunohistochemistry) and tauopathy (neurofibrillary tangles, neuropil threads and neuritic plaques; PHF1 immunohistochemistry) appear more severe in the frozen right occipital lobe (E and F) than in the formalin-fixed left lobe (B and C). Bars: A and D, 500 µm; B, C, E, F, 100 µm.

Figure 2

Heatmaps summarizing the semiquantitative neuropathologic grading of the left hemibrain of ADNI participants. AD, Alzheimer disease; DLB, dementia with Lewy bodies. Heatmaps summarizing the semiquantitative neuropathological grading of molecular pathologies from left to right: Diffuse Plaques, diffuse Aβ plaques; Tau NFT, phospho-tau immunoreactive neurofibrillary tangles; Lewy Bodies, phospho-α-synuclein immunoreactive Lewy bodies; and TDP Dep., phospho-TDP-43-immunoreactive neuronal cytoplasmic inclusions. Adapted from [26].

Figure 3

Semiquantitative neuropathologic heat maps facilitate PET-PiB amyloid imaging – Aβ plaque correlations. Upper panel: Coronal view (left) and lateral view (right) show PET-PiB binding in arbitrary units in a DIAN participant. Lower panel: Semiquantitative histopathologic data of diffuse plaques (left) and cored/compact plaques (right) from the left hemibrain; data are mirrored to generate a whole coronal slice at the same level as in the upper panel. Areas in gray not assessed.