Preclinical AD Workgroup staging: pathological correlates and potential challenges - PubMed (original) (raw)
Preclinical AD Workgroup staging: pathological correlates and potential challenges
Gregory A Jicha et al. Neurobiol Aging. 2012 Mar.
Abstract
The National Institute on Aging Preclinical Alzheimer's disease Workgroup (PADW) has issued a preliminary report with recommendations for classifying preclinical Alzheimer's disease (pAD) according to 3 early disease stages. Here we examine the PADW recommendations in relation to neuropathological features in a large, consecutive series of cognitively intact elderly persons, autopsied within a year after cognitive testing (n = 126 cognitively intact patients with mean age 83.7 years at death). Subjects were grouped based on a hypothetical construct correlating pathological features with PADW stages. Many cognitively intact individuals were classifiable as pAD (53/126 or 43%), as expected based on epidemiological and biomarker studies. Of these, most (48%) were in "stage 3", which corresponds to amyloid pathology with early neurodegeneration. As with prior studies, our data indicate that the development of neocortical neurofibrillary tangles is the key pathological event that is not observed in pAD cases: Braak stages III or IV pathology are hence not truly a substrate for "intermediate likelihood" that cognitive impairment is due to Alzheimer's disease (AD). We also stress the importance of comorbid non-Alzheimer's disease brain pathologies (hippocampal sclerosis, neocortical alpha-synucleinopathy, cerebrovascular disease, and brains with hippocampal neurofibrillary tangles but no cortical amyloid plaques) that can contribute to the development of cognitive impairment, or which may serve as confounds in the application of the PADW recommendations. While the final recommendations from the PADW working group have not yet been released, this preliminary analysis provides a perspective on those recommendations from a neuropathological point of view.
Copyright © 2012 Elsevier Inc. All rights reserved.
Figures
Fig. 1
Flow chart to demonstrate the individuals included and excluded in the assessment of preclinical Alzheimer’s disease (pAD) in correlation with the pAD Workgroup recommendations (Alzheimer’s Association, 2010). Note that all individuals with eventual diagnosis of pAD or mild cognitive impairment (MCI) were recruited while initially cognitively normal. * Of the 14 individuals included as early Alzheimer’s disease (AD; Braak stage V), 12 were recruited while neurologically normal but 2 were cognitively impaired when recruited.
Fig. 2
Representative histopathological features that may correspond with the evolution of preclinical Alzheimer’s disease (pAD) in correlation with the preliminary recommendations of the Preclinical Alzheimer’s Disease Workgroup (stages described at right). Stage 0 (A and B) refers to cases without cerebral amyloidosis. There are no amyloid plaques —(A) using Aβ immunohistochemistry (IHC) — or neurofibrillary changes — (B) PHF-1/phospho-tau IHC in a low-power photomicrograph of hippocampal CA1 (counterstained with hematoxylin). In stage 1, there are diffuse amyloid plaques, detectable using Aβ IHC (C), but these plaques lack neuritic component that would be detectable using a Bielschowsky silver stain (D). In stage 2, there are both diffuse amyloid plaques (E) and also some plaques with a neuritic component (blue arrow in F). By stage 3, there are typically features seen in fulminant Alzheimer’s disease (AD), but with lower densities especially in the neocortex. Medial temporal lobe regions such as the amygdala (G) may have many neuritic amyloid plaques and tangles as shown here with a sensitive Gallyas silver stain. However, one usually does not see many neuritic amyloid plaques in the neocortex (upper left of H) without also seeing some neurofibrillary tangles (lower right of H). Scale bars: (A) 500 µm; (B) 1 mm; (C–F) 100 µm; (G) 50 µm; (H) 30 µm.
Fig. 3
Grid with information related to how patients were classified in the current study. Only patients that died while cognitively intact with cognitive testing within a year of death were included in this study and have results that are relevant to this chart. The grid shows that individual patients can be categorized according to the extent of neurofibrillary pathology using the Braak staging. Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) scores provide information about the density of neuritic amyloid plaques. In the earliest part of the disease, without either neurofibrillary tangles or neuritic amyloid plaques (blue portion of the chart), a case is determined to be stage 0 if there are no diffuse plaques but a case is stage 1 if the brain has enough diffuse amyloid plaques to meet the Khachaturian criteria (Khachaturian, 1985). Stage 0-N refers to cases with a modicum of neurofibrillary tangles in the hippocampus, without amyloid plaques in neocortex, as described in a prior study (Nelson et al., 2009a).
Fig. 4
Pie chart shows relative numbers of cases that fall into each of the preclinical Alzheimer’s disease (pAD) stages (total n = 126). A relatively large proportion of are “stage 0” (59/126, or 47%) which indicates no cortical amyloidosis. Cases with diffuse plaques but lacking substantial numbers of neuritic plaques (“stage 1”) are relatively few. Cases with mixed” pathology were found in 10/54 (19%) of stage 1–3 pAD cases, indicating subclinical levels of cerebrovascular disease, cortical Lewy bodies, or hippocampal sclerosis.
Fig. 5
Groups in the different preclinical Alzheimer’s disease (pAD) stages lack significant differences on many cognitive tests. Shown here are results for Logical Memory: Immediate Recall and Animal Fluency sections of the Universal Dataset test battery. Error bars = standard deviation. All cases with pAD were tested within a year of death. * p < 0.05 or ** p < 0.001.
Fig. 6
Early Braak stages are not helpful interval variables in the context of preclinical or early Alzheimer’s disease (AD). Shown here is a chart of final Mini Mental State Examination (MMSE) scores plotted in correlation with Braak staging of neurofibrillary degeneration. From an initial sample of University of Kentucky Alzheimer’s Disease Center (UK-ADC) autopsies that included demented and nondemented subjects (n = 612), cases were excluded that had frontotemporal dementia pathology, alpha-synucleinopathy such as dementia with Lewy bodies, vascular dementia, and any case that lacked Braak staging or MMSE scores. Cases were not excluded from this analysis based on interval between final MMSE test and death. Although in a person with documented dementia, a Braak stage less than V may indicate some impact from the disease, this ordinal variable cannot confidently predict global cognitive status in the earliest stages of Alzheimer’s disease. This is probably because cognitive impact from AD pathology may be difficult to detect before there are abundant neocortical neurofibrillary tangles (NFTs) (Arriagada et al., 1992a). p values show result of comparing mean MMSE scores in each Braak stage versus Braak stage 0 using unpaired Student t test. NS, not significant; NS*, average final MMSE scores trended lower for Braak stage III and IV (p ~ 0.04), but this was not significant after correction for multiple comparisons.
Fig. 7
Some prevalent brain pathologies in older individuals do not map well onto the preliminary Preclinical Alzheimer’s disease Workgroup (PADW) recommendations. Brains with medial temporal lobe neurofibrillary tangles (NFTs) but no amyloid plaques in the cortex, which we refer to as stage 0-N, comprise over 10% of preclinical Alzheimer’s disease (pAD) cases (A and B). This brain from a 78-year-old male (final Mini Mental State Examination [MMSE] score = 29) showed moderate densities of NFTs in the hippocampal formation and entorhinal cortex (A, stained with Gallyas stain) but there was no Aβ-positive plaques detected (B), in contrast to what is seen in Alzheimer’s disease (AD) brains using the same stain (C, at same magnification). Whereas stage 0-N was seen in 13/126 cognitively intact individuals, we found only a single case with hippocampal sclerosis in nondemented individuals, so this form of non-AD hippocampal atrophy appears more relevant to mild cognitive impairment and demented states. (D) An hematoxylin and eosin stained brain from an 88-year-old female with Clinical Dementia Rating (CDR) = 1 and clinical diagnosis of presumed early AD. The hippocampus appears shrunken, especially the subiculum (arrow). Immunohistochemical stain for TDP-43 (E; nuclei counterstained with hematoxylin) showed aberrant pattern of staining with cytoplasmic TDP-43 inclusions (arrows). Scale bars: (A and E) 25 µm, (B and C) 100 µm; (D) 2 mm.
References
- Consensus recommendations for the postmortem diagnosis of Alzheimer’s disease. The National Institute on Aging, and Reagan Institute Working Group on Diagnostic Criteria for the Neuropathological Assessment of Alzheimer’s Disease. Neurobiol. Aging. 1997;18(4 Suppl):S1–S2. -PubMed
- Aarsland D, Kurz M, Beyer M, Bronnick K, Piepenstock Nore S, Ballard C. Early discriminatory diagnosis of dementia with Lewy bodies. The emerging role of CSF and imaging biomarkers. Dement. Geriatr. Cogn. Disord. 2008;25:195–205. -PubMed
- Albert MS, Moss MB, Tanzi R, Jones K. Preclinical prediction of AD using neuropsychological tests. J. Int. Neuropsychol. Soc. 2001;7:631–639. -PubMed
- Alzheimer’s Association. Preliminary Recommendations by Preclinical Alzheimer’s Disease Workgroup. 2010 Available at: www.alz.org/research/diagnostic_criteria/
- Arriagada PV, Growdon JH, Hedley-Whyte ET, Hyman BT. Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology. 1992a;42:631–639. -PubMed
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Medical