Amyloid β Deposition and Suspected Non-Alzheimer Pathophysiology and Cognitive Decline Patterns for 12 Years in Oldest Old Participants Without Dementia | Dementia and Cognitive Impairment | JAMA Neurology | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Timing of Neuroimaging Relative to Neuropsychological Assessments
Timing of Neuroimaging Relative to Neuropsychological Assessments

Participants were evaluated longitudinally from 2000 to 2016. Annual cognitive testing was performed with the exception of a 3- to 4-year gap from 2000 to 2004. Classification of biomarker status was based on Pittsburgh Compound B positron emission tomography and magnetic resonance imaging performed in 2009.

Figure 2.  California Verbal Learning Test Delayed Recall (Verbal Memory) Performance Over Time, by Biomarker Group
California Verbal Learning Test Delayed Recall (Verbal Memory) Performance Over Time, by Biomarker Group

Decline in California Verbal Learning Test delayed recall performance over the follow-up period by the 4 biomarker groups. Higher score is better. Aβ indicates amyloid β; ND, neurodegeneration.

Table 1.  Demographics and Neuropsychological Test Scores at the Time of Neuroimaginga
Demographics and Neuropsychological Test Scores at the Time of Neuroimaginga
Table 2.  Estimates of the Difference in Rates of Annualized or Accelerated Cognitive Test Score Change, by Biomarker Group Relative to Aβ/ND Groupa,b,c
Estimates of the Difference in Rates of Annualized or Accelerated Cognitive Test Score Change, by Biomarker Group Relative to Aβ−/ND− Groupa,b,c
Table 3.  Estimates of the Difference in Rates of Annualized or Accelerated Cognitive Test Score Change, by Biomarker Group Relative to the Aβ/ND Group in Participants With Normal Cognition at Neuroimaginga,b,c
Estimates of the Difference in Rates of Annualized or Accelerated Cognitive Test Score Change, by Biomarker Group Relative to the Aβ−/ND− Group in Participants With Normal Cognition at Neuroimaginga,b,c
1.
Jack  CR  Jr, Wiste  HJ, Knopman  DS,  et al.  Rates of β-amyloid accumulation are independent of hippocampal neurodegeneration.  Neurology. 2014;82(18):1605-1612.PubMedGoogle ScholarCrossref
2.
Sperling  RA, Aisen  PS, Beckett  LA,  et al.  Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease.  Alzheimers Dement. 2011;7(3):280-292.PubMedGoogle ScholarCrossref
3.
Jack  CR  Jr, Knopman  DS, Weigand  SD,  et al.  An operational approach to National Institute on Aging–Alzheimer’s Association criteria for preclinical Alzheimer disease.  Ann Neurol. 2012;71(6):765-775.PubMedGoogle ScholarCrossref
4.
Toledo  JB, Weiner  MW, Wolk  DA,  et al; Alzheimer’s Disease Neuroimaging Initiative.  Neuronal injury biomarkers and prognosis in ADNI subjects with normal cognition.  Acta Neuropathol Commun. 2014;2:26.PubMedGoogle ScholarCrossref
5.
Petersen  RC, Aisen  P, Boeve  BF,  et al.  Mild cognitive impairment due to Alzheimer disease in the community.  Ann Neurol. 2013;74(2):199-208.PubMedGoogle Scholar
6.
Caroli  A, Prestia  A, Galluzzi  S,  et al; Alzheimer’s Disease Neuroimaging Initiative.  Mild cognitive impairment with suspected nonamyloid pathology (SNAP): prediction of progression.  Neurology. 2015;84(5):508-515.PubMedGoogle ScholarCrossref
7.
Mormino  EC, Betensky  RA, Hedden  T,  et al.  Synergistic effect of β-amyloid and neurodegeneration on cognitive decline in clinically normal individuals.  JAMA Neurol. 2014;71(11):1379-1385.PubMedGoogle ScholarCrossref
8.
Vos  SJ, Verhey  F, Frölich  L,  et al; Alzheimer’s Disease Neuroimaging Initiative.  Prevalence and prognosis of Alzheimer’s disease at the mild cognitive impairment stage.  Brain. 2015;138(pt 5):1327-1338.PubMedGoogle ScholarCrossref
9.
Roe  CM, Fagan  AM, Grant  EA,  et al.  Amyloid imaging and CSF biomarkers in predicting cognitive impairment up to 7.5 years later.  Neurology. 2013;80(19):1784-1791.PubMedGoogle ScholarCrossref
10.
Wirth  M, Villeneuve  S, Haase  CM,  et al.  Associations between Alzheimer disease biomarkers, neurodegeneration, and cognition in cognitively normal older people.  JAMA Neurol. 2013;70(12):1512-1519.PubMedGoogle Scholar
11.
Jack  CR  Jr, Knopman  DS, Chételat  G,  et al.  Suspected non-Alzheimer disease pathophysiology—concept and controversy.  Nat Rev Neurol. 2016;12(2):117-124.PubMedGoogle ScholarCrossref
12.
Vos  SJ, Xiong  C, Visser  PJ,  et al.  Preclinical Alzheimer’s disease and its outcome: a longitudinal cohort study.  Lancet Neurol. 2013;12(10):957-965.PubMedGoogle ScholarCrossref
13.
van Harten  AC, Smits  LL, Teunissen  CE,  et al.  Preclinical AD predicts decline in memory and executive functions in subjective complaints.  Neurology. 2013;81(16):1409-1416.PubMedGoogle ScholarCrossref
14.
Wisse  LEM, Butala  N, Das  SR,  et al; Alzheimer’s Disease Neuroimaging Initiative.  Suspected non-AD pathology in mild cognitive impairment.  Neurobiol Aging. 2015;36(12):3152-3162.PubMedGoogle ScholarCrossref
15.
Prestia  A, Caroli  A, van der Flier  WM,  et al.  Prediction of dementia in MCI patients based on core diagnostic markers for Alzheimer disease.  Neurology. 2013;80(11):1048-1056.PubMedGoogle ScholarCrossref
16.
Knopman  DS, Jack  CR  Jr, Wiste  HJ,  et al.  Short-term clinical outcomes for stages of NIA-AA preclinical Alzheimer disease.  Neurology. 2012;78(20):1576-1582.PubMedGoogle ScholarCrossref
17.
Soldan  A, Pettigrew  C, Cai  Q,  et al; BIOCARD Research Team.  Hypothetical preclinical Alzheimer disease groups and longitudinal cognitive change.  JAMA Neurol. 2016;73(6):698-705.PubMedGoogle ScholarCrossref
18.
Burnham  SC, Bourgeat  P, Doré  V,  et al; AIBL Research Group.  Clinical and cognitive trajectories in cognitively healthy elderly individuals with suspected non-Alzheimer’s disease pathophysiology (SNAP) or Alzheimer’s disease pathology: a longitudinal study.  Lancet Neurol. 2016;15(10):1044-1053.PubMedGoogle ScholarCrossref
19.
Mormino  EC, Papp  KV, Rentz  DM,  et al.  Heterogeneity in suspected non–Alzheimer disease pathophysiology among clinically normal older individuals.  JAMA Neurol. 2016;73(10):1185-1191.PubMedGoogle ScholarCrossref
20.
Bennett  DA, Schneider  JA, Arvanitakis  Z,  et al.  Neuropathology of older persons without cognitive impairment from two community-based studies.  Neurology. 2006;66(12):1837-1844.PubMedGoogle ScholarCrossref
21.
Aizenstein  HJ, Nebes  RD, Saxton  JA,  et al.  Frequent amyloid deposition without significant cognitive impairment among the elderly.  Arch Neurol. 2008;65(11):1509-1517.PubMedGoogle ScholarCrossref
22.
Sonnen  JA, Santa Cruz  K, Hemmy  LS,  et al.  Ecology of the aging human brain.  Arch Neurol. 2011;68(8):1049-1056.PubMedGoogle ScholarCrossref
23.
DeKosky  ST, Williamson  JD, Fitzpatrick  AL,  et al; Ginkgo Evaluation of Memory (GEM) Study Investigators.  Ginkgo biloba for prevention of dementia: a randomized controlled trial.  JAMA. 2008;300(19):2253-2262.PubMedGoogle ScholarCrossref
24.
DeKosky  ST, Fitzpatrick  A, Ives  DG,  et al; GEMS Investigators.  The Ginkgo Evaluation of Memory (GEM) study: design and baseline data of a randomized trial of Ginkgo biloba extract in prevention of dementia.  Contemp Clin Trials. 2006;27(3):238-253.PubMedGoogle ScholarCrossref
25.
Snitz  BE, Weissfeld  LA, Lopez  OL,  et al.  Cognitive trajectories associated with β-amyloid deposition in the oldest-old without dementia.  Neurology. 2013;80(15):1378-1384.PubMedGoogle ScholarCrossref
26.
Lezak  MD, Howieson  DB, Loring  DW.  Neuropsychological Assessment. 4th ed. New York, NY: Oxford University Press; 2004.
27.
Becker  JT, Boller  F, Saxton  J, McGonigle-Gibson  KL.  Normal rates of forgetting of verbal and non-verbal material in Alzheimer’s disease.  Cortex. 1987;23(1):59-72.PubMedGoogle ScholarCrossref
28.
Trenerry  MR, Crosson  B, DeBoe  J, Leber  WR.  Stroop Neuropsychological Screening Test. Odessa, FL: Psychological Assessment Resources; 1989.
29.
Lopez  OL, Becker  JT, Jagust  WJ,  et al.  Neuropsychological characteristics of mild cognitive impairment subgroups.  J Neurol Neurosurg Psychiatry. 2006;77(2):159-165.PubMedGoogle ScholarCrossref
30.
Saxton  J, Ratcliff  G, Munro  CA,  et al.  Normative data on the Boston Naming Test and two equivalent 30-item short forms.  Clin Neuropsychol. 2000;14(4):526-534.PubMedGoogle ScholarCrossref
31.
Fai  AH-T, Cornelius  PL.  Approximate F-tests of multiple degree of freedom hypotheses in generalized least squares analyses of unbalanced split-plot experiments.  J Stat Comput Simul. 1996;54(4):363-378.Google ScholarCrossref
32.
Wirth  M, Oh  H, Mormino  EC, Markley  C, Landau  SM, Jagust  WJ.  The effect of amyloid β on cognitive decline is modulated by neural integrity in cognitively normal elderly.  Alzheimers Dement. 2013;9(6):687-698.e1.PubMedGoogle ScholarCrossref
33.
Jack  CR  Jr, Wiste  HJ, Weigand  SD,  et al.  Age-specific population frequencies of cerebral β-amyloidosis and neurodegeneration among people with normal cognitive function aged 50-89 years: a cross-sectional study.  Lancet Neurol. 2014;13(10):997-1005.PubMedGoogle ScholarCrossref
34.
Braak  H, Braak  E.  Frequency of stages of Alzheimer-related lesions in different age categories.  Neurobiol Aging. 1997;18(4):351-357.PubMedGoogle ScholarCrossref
35.
Raz  N, Lindenberger  U, Rodrigue  KM,  et al.  Regional brain changes in aging healthy adults: general trends, individual differences and modifiers.  Cereb Cortex. 2005;15(11):1676-1689.PubMedGoogle ScholarCrossref
36.
Jack  CR  Jr, Dickson  DW, Parisi  JE,  et al.  Antemortem MRI findings correlate with hippocampal neuropathology in typical aging and dementia.  Neurology. 2002;58(5):750-757.PubMedGoogle ScholarCrossref
37.
Whitwell  JL, Jack  CR  Jr, Parisi  JE,  et al.  Does TDP-43 type confer a distinct pattern of atrophy in frontotemporal lobar degeneration?  Neurology. 2010;75(24):2212-2220.PubMedGoogle ScholarCrossref
38.
Jack  CR  Jr.  PART and SNAP.  Acta Neuropathol. 2014;128(6):773-776. PubMedGoogle ScholarCrossref
39.
Di Paola  M, Caltagirone  C, Fadda  L, Sabatini  U, Serra  L, Carlesimo  GA.  Hippocampal atrophy is the critical brain change in patients with hypoxic amnesia.  Hippocampus. 2008;18(7):719-728.PubMedGoogle ScholarCrossref
40.
Mormino  EC, Papp  KV.  Cognitive decline in preclinical stage 2 Alzheimer disease and implications for prevention trials.  JAMA Neurol. 2016;73(6):640-642.PubMedGoogle ScholarCrossref
41.
Bäckman  L, Jones  S, Berger  AK, Laukka  EJ, Small  BJ.  Cognitive impairment in preclinical Alzheimer’s disease: a meta-analysis.  Neuropsychology. 2005;19(4):520-531.PubMedGoogle ScholarCrossref
42.
Saxton  J, Lopez  OL, Ratcliff  G,  et al.  Preclinical Alzheimer disease: neuropsychological test performance 1.5 to 8 years prior to onset.  Neurology. 2004;63(12):2341-2347.PubMedGoogle ScholarCrossref
43.
Masur  DM, Sliwinski  M, Lipton  RB, Blau  AD, Crystal  HA.  Neuropsychological prediction of dementia and the absence of dementia in healthy elderly persons.  Neurology. 1994;44(8):1427-1432.PubMedGoogle ScholarCrossref
44.
Klunk  WE, Perani  D.  Amyloid and neurodegeneration: converging and diverging paths.  Neurology. 2013;81(20):1728-1729.PubMedGoogle ScholarCrossref
45.
Crary  JF, Trojanowski  JQ, Schneider  JA,  et al.  Primary age-related tauopathy (PART): a common pathology associated with human aging.  Acta Neuropathol. 2014;128(6):755-766.PubMedGoogle ScholarCrossref
46.
Gordon  BA, Blazey  T, Su  Y,  et al.  Longitudinal β-amyloid deposition and hippocampal volume in preclinical Alzheimer disease and suspected non–Alzheimer disease pathophysiology.  JAMA Neurol. 2016;73(10):1192-1200.PubMedGoogle ScholarCrossref
47.
Savva  GM, Wharton  SB, Ince  PG, Forster  G, Matthews  FE, Brayne  C; Medical Research Council Cognitive Function and Ageing Study.  Age, neuropathology, and dementia.  N Engl J Med. 2009;360(22):2302-2309.PubMedGoogle ScholarCrossref
48.
Haroutunian  V, Schnaider-Beeri  M, Schmeidler  J,  et al.  Role of the neuropathology of Alzheimer disease in dementia in the oldest-old.  Arch Neurol. 2008;65(9):1211-1217.PubMedGoogle ScholarCrossref
49.
Kawas  CH, Kim  RC, Sonnen  JA, Bullain  SS, Trieu  T, Corrada  MM.  Multiple pathologies are common and related to dementia in the oldest-old: the 90+ Study.  Neurology. 2015;85(6):535-542.PubMedGoogle ScholarCrossref
50.
Richard  E, Schmand  B, Eikelenboom  P, Westendorp  RG, Van Gool  WA.  The Alzheimer myth and biomarker research in dementia.  J Alzheimers Dis. 2012;31(suppl 3):S203-S209.PubMedGoogle Scholar
51.
Kuller  LH, Lopez  OL.  Dementia and Alzheimer’s disease: a new direction: the 2010 Jay L. Foster Memorial Lecture.  Alzheimers Dement. 2011;7(5):540-550.PubMedGoogle ScholarCrossref
52.
Lopez  OL, Klunk  WE, Mathis  C,  et al.  Amyloid, neurodegeneration, and small vessel disease as predictors of dementia in the oldest-old.  Neurology. 2014;83(20):1804-1811.PubMedGoogle ScholarCrossref
Original Investigation
January 2018

Amyloid β Deposition and Suspected Non-Alzheimer Pathophysiology and Cognitive Decline Patterns for 12 Years in Oldest Old Participants Without Dementia

Author Affiliations
  • 1School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania
  • 2Department of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania
  • 3Department of Neurology, University of Pittsburgh, Pittsburgh, Pennsylvania
  • 4Department of Psychiatry, University of Pittsburgh, Pittsburgh, Pennsylvania
  • 5Department of Radiology, University of Pittsburgh, Pittsburgh, Pennsylvania
  • 6now with the Department of Radiology, Massachusetts General Hospital, Boston
  • 7Department of Epidemiology, University of Pittsburgh, Pittsburgh, Pennsylvania
  • 8Department of Human Genetics, University of Pittsburgh, Pittsburgh, Pennsylvania
  • 9Department of Neurology, University of Florida, Gainesville
JAMA Neurol. 2018;75(1):88-96. doi:10.1001/jamaneurol.2017.3029
Key Points

Question  Do domain patterns of long-term cognitive decline in the oldest old differ by neuroimaging biomarker status?

Findings  In this longitudinal study, 175 adults without dementia who were older than 80 years were followed up a mean of 12.2 years, with positron emission tomography and magnetic resonance imaging completed in the middle of the observation period. Isolated hippocampal atrophy was associated only with greater decline in memory, while isolated amyloid β was associated with decline in memory plus language and executive functions.

Meaning  These findings suggest different pathophysiologic processes underlying preclinical biomarkers of Alzheimer disease in the oldest old.

Abstract

Importance  The prevalence of pathologic conditions of the brain associated with Alzheimer disease increases strongly with age. Little is known about the distribution and clinical significance of preclinical biomarker staging in the oldest old, when most individuals without dementia are likely to have positive biomarkers.

Objective  To compare the patterns of long-term cognitive decline in multiple domains by preclinical biomarker status in the oldest old without dementia.

Design, Setting, and Participants  A longitudinal observational study with a mean (SD) of 12.2 (2.2) years (range 7.2-15.1 years) of follow-up was conducted in an academic medical center from August 24, 2000, to January 14, 2016, including and extending observations from the Ginkgo Evaluation of Memory study. A total of 197 adults who had completed the Ginkgo Evaluation of Memory study, were free of dementia, and were able to undergo magnetic resonance imaging were eligible for a neuroimaging study in 2009. Of these patients, 175 were included in the present analyses; 140 (80%) were cognitively normal and 35 (20%) had mild cognitive impairment.

Main Outcomes and Measures  Biomarker groups included amyloid β negative (Aβ)/neurodegeneration negative (ND), amyloid β positive (Aβ+)/ND, Aβ/neurodegeneration positive (ND+), and Aβ+/ND+ based on Pittsburgh Compound B retention and hippocampal volume in 2009. Participants completed baseline neuropsychological testing from 2000 to 2002 and annual testing from 2004 to 2016. Domains included memory, executive function, language, visual-spatial reasoning, and attention and psychomotor speed. Slopes of decline were evaluated with linear mixed models adjusted for age, sex, and years of education.

Results  Of the 175 participants (71 women and 104 men), at imaging, mean (SD) age was 86.0 (2.9) years (range, 82-95 years). A total of 42 participants (24.0%) were Aβ/ND, 32 (18.3%) were Aβ+/ND, 35 (20.0%) were Aβ/ND+, and 66 (37.7%) were Aβ+/ND+. On all cognitive measures, the Aβ+/ND+ group showed the steepest decline. Compared with the Aβ/ND group, the amyloid deposition alone (Aβ+/ND) group showed faster decline on tests of verbal and visual memory (–0.3513; 95% CI, –0.5269 to –0.1756), executive function (0.0158; 95% CI, 0.0013-0.0303), and language (–0.1934; 95% CI, –0.3520 to –0.0348). The Aβ/ND+ group showed faster visual memory decline than the Aβ/ND reference group (–0.3007; 95% CI, –0.4736 to –0.1279).

Conclusions and Relevance  In the oldest old without dementia, presence of either or both Aβ and hippocampal atrophy is typical (>75%). Isolated hippocampal volume atrophy is associated only with greater decline in memory. However, isolated Aβ is associated with decline in memory plus language and executive functions. These findings suggest different underlying pathophysiologic processes in the Aβ+/ND and Aβ/ND+ groups.

Introduction

In recent years, studies of neuroimaging biomarkers of Alzheimer disease (AD) pathologic conditions have demonstrated the independence of amyloid β (Aβ) amyloidosis and neurodegeneration (ND) markers in the preclinical phase. Specifically, Jack et al1 proposed a 2-feature biomarker approach to classify clinically normal older individuals with Aβ accumulation and/or ND. The cross-tabulation results in the following 4 biomarker classifications: Aβ and ND negative (Aβ/ND), Aβ positive and ND negative (Aβ+/ND), Aβ negative and ND positive (Aβ/ND+), and Aβ positive and ND positive (Aβ+/ND+). Corresponding to the National Institute on Aging–Alzheimer’s Association staging of preclinical AD2 and as modified by Jack et al,3/ND maps to stage 0, Aβ+/ND to stage 1, and Aβ+/ND+ to stage 2/3; the Aβ/ND+ classification has been termed suspected non-Alzheimer disease pathophysiology (SNAP). The 2-feature biomarker approach has been applied to study participants with normal cognition (NC) and mild cognitive impairment (MCI).3-10

The characterization of SNAP has triggered debate about its pathophysiologic basis and its role in cognitive decline and progression to dementia.11 Cross-sectionally, cognitive profiles of groups with NC and SNAP do not differ from groups with NC and Aβ/ND and Aβ+/ND at baseline.7,11-13 However, among individuals with MCI, SNAP and Aβ+/ND groups show deficits compared with Aβ/ND groups.14 Regarding the risk of cognitive decline, studies consistently show that Aβ+/ND+ confers the highest risk of cognitive decline and Aβ/ND the lowest risk.9,12-16

However, it is unclear whether the Aβ+/ND and the SNAP groups show specific cognitive decline signatures. Several studies have examined longitudinal composite scores or global cognitive scores,7,17-19 but few have compared patterns of decline in various cognitive domains between SNAP and Aβ-positive participants. Furthermore, little is known about 2-feature biomarker classification in oldest old adults, when prevalence of amyloidosis and neurodegenerative pathologic conditions is at its highest.10,20-22 In this study, we applied the 2-feature biomarker grouping method to compare groups defined by Aβ deposition and evidence of ND (hippocampal atrophy) on cognitive decline across various cognitive domains during long-term follow-up, with a mean of 12.2 years.

Methods
Participants

Participants were a subgroup of the Ginkgo Evaluation of Memory (GEM) study,23,24 conducted from 2000 to 2008, who were enrolled in the subsequent Ginkgo Evaluation of Memory Study (GEMS) Imaging Sub-Study (details of which have been previously described).25 Of 966 participants of the GEM study at the Pittsburgh, Pennsylvania, site, 197 continued in 2009 with the GEMS Imaging Sub-Study, which included Pittsburgh Compound B positron emission tomography (PiB-PET) and magnetic resonance imaging (MRI). The study inclusion criterion was completion of the GEM parent study. Exclusion criteria were dementia at the completion of the GEM parent study and contraindications for neuroimaging.25 Of the 197 participants in the GEMS Imaging Sub-Study, 3 were excluded for technical issues with PiB-PET, 16 were further excluded for lack of T2 MRI imaging data, and 3 were excluded owing to a dementia diagnosis in 2009, resulting in 175 participants for this study. The 175 older adults had a mean (SD) age of 78.0 (2.8) years (range, 75.0-87.8 years) at the initial GEM study baseline visit between 2000 to 2002 and were followed up for a mean (SD) of 12.2 (2.2) years (range, 7.2-15.1 years). At the time of imaging in 2009, the mean (SD) age was 86.0 (2.9) years. Clinical and neuropsychological evaluations were completed annually through the GEM study and the GEMS Imaging Sub-Study. The mean (SD) length of follow-up from 2009 was 4.2 (2.2) years (range, 0-6.9 years). Therefore, in total, study procedures were conducted from August 24, 2000, to January 14, 2016. Participants provided written informed consent for all study procedures as approved by the University of Pittsburgh institutional review board.

Neuroimaging

Positron emission tomography data were acquired for 20 minutes (4 × 5-minute frames) beginning 50 minutes after injection of a mean (SD) of 15 (1.5) mCi of PiB on a Siemens/Computer Technology Imaging emission computerized axial tomography high-resolution plus scanner in 3-dimensional imaging mode (63 planes with slice width of 2.4 mm). Retention of PiB was measured over the 50- to 70-minute scan interval and scaled to the injected dose and body mass to generate standardized uptake values (SUVs). The SUVs were then normalized to the SUV of the cerebellum reference region to generate SUV ratio (SUVR) measures of PiB retention. Using an iterative mild outlier cutoff method based on an independent sample of 62 controls,21 we define Aβ-negative and Aβ-positive status by a global cortical measure (mean of the frontal, anterior cingulate, precuneus, parietal, and temporal cortices) with a cutoff of 1.57 SUVR units.

Structural 1.5-T MRI images (General Electric Signa) were acquired, from which the hippocampal volume (HV) was derived and normalized to total intracranial volume. w-Scores were calculated based on an independent reference group of 77 NC individuals (age range, 45-89 years) adjusted for sex and age. Neurodegeneration negative was defined as a w-score less than −0.9063, reflecting 85% sensitivity to AD generated from an independent sample of 51 individuals with AD. The 175 participants were stratified into the following groups: Aβ/ND, Aβ+/ND, Aβ/ND+ (SNAP), and Aβ+/ND+ based on PiB and MRI findings in 2009 (Figure 1).

Cognitive Assessment

Participants completed the following neuropsychological test battery26 at the initial visit (2000-2002) and annually starting in 2004. Memory: California Verbal Learning Test (CVLT), immediate and delayed recall, and the modified Rey-Osterrieth (R-O) figure test, immediate and delayed recall.27 Executive functions: Trail-Making Test part B (TMT-B) and the Trenerry Stroop Test, interference condition.28 Visual-spatial reasoning: Modified Wechsler Adult Intelligence Scale–Revised Block Design29 and copy condition of the modified R-O figure test. Language: letter verbal fluency, semantic (animals) verbal fluency, and the 30-item Boston Naming Test (BNT).30 Attention and psychomotor speed: Trail-Making Test part A (TMT-A) and the Trenerry Stroop Test, control condition. In 2009 at the time of neuroimaging, a reduced battery was administered, omitting Block Design, BNT, and the Stroop Test (see Figure 1 for study timeline).

The GEM study Cognitive Diagnostic Center23 completed adjudication in the Imaging Sub-Study in 2009, blind to PiB-PET results, taking into account historical serial cognitive assessments from the parent GEM study. Criteria for MCI included 1 to 3 tests impaired at cutoffs of 1.5 SD below means adjusted for age and educational level.

Statistical Analysis

Linear mixed models with a fixed factor of time, quadratic time, and biomarker group as well as their linear and quadratic interactions (when significant) were used to estimate the rate of cognitive decline in neuropsychological test scores for the entire observational period. A random intercept and slope were included with an unstructured covariance to account for the repeated observation structure of the data. The Satterthwaite method31 was used for computing degrees of freedom. Models were adjusted for age, sex, and years of education. Primary analyses included all 175 participants, adjusted for age, sex, and years of education. The Aβ/ND status group × time interaction term was used to reflect difference in the rates of cognitive performance decline over time relative to the reference group, which was Aβ/ND status. The addition of quadratic time and quadratic interaction effects were tested using likelihood ratio test between the models with and without these terms and compared with the corresponding χ2 degrees of freedom. The quadratic interactions reflect accelerating or decelerating decline rates compared with the reference group. Secondary analyses included only participants with normal cognitive status (n = 140) at the time of imaging. A sensitivity analysis examined shifting the baseline of cognitive trajectories to the time of imaging (2009). Since the number of cognitive measurements was reduced, time was modeled as a linear term only in these models.

Results
Demographic and Clinical Status at Baseline and the Time of Imaging

At the time of imaging (2009), 42 participants (24.0%) were classified as Aβ/ND, 32 (18.3%) as Aβ+/ND, 35 (20.0%) as Aβ/ND+, and 66 (37.7%) as Aβ+/ND+. Shown in Table 1, the 4 groups stratified based on Aβ and ND status in 2009 did not differ in age, sex, race, and years of education, although the Aβ+/ND+ group showed a trend toward older age. The Aβ+/ND+ group had more APOE*4 carriers (22 of 62 [35.5%]) compared with the Aβ/ND group (2 of 40 [5.0%]) and Aβ/ND+ group (2 of 34 [5.9%]), with the Aβ+/ND group intermediate (5 of 27 [18.5%]). In 2009, most participants (140 [80.0%]) had NC and the remaining 35 (20.0%) had MCI. The proportion of participants with MCI within each group did not differ significantly. Table 1 also presents neuropsychological test scores at the time of imaging. The Aβ+/ND+ group performed the poorest. On tests of executive function and psychomotor speed (TMT-A and TMT-B), both Aβ+ groups performed comparably. At baseline of the observation period (2000-2002), there were significant differences among the imaging subgroups on Stroop interference (F3,159 = 2.89; P = .04) and control (F3,167 = 3.02; P = .03) conditions, with the Aβ+/ND group showing the poorest performance. On all other baseline cognitive tests, there were no significant differences among imaging subgroups.

Change Over Time in Neuropsychological Test Scores

Table 2 shows the relative rates of annualized cognitive test score change among the Aβ+/ND, Aβ/ND+, and Aβ+/ND+ groups compared with Aβ/ND as the reference group. The Aβ+/ND+ group showed significantly steeper decline in all neuropsychological tests except for phonemic fluency and the control condition (ie, word reading) of the Stroop test. All 3 biomarker-positive groups showed steeper decline in R-O figure immediate (eFigure 1 in the Supplement) and delayed recalls. In addition, the Aβ+/ND group showed steeper decline in CVLT delayed recall (Figure 2), TMT-B (eFigure 2 in the Supplement), BNT, and semantic fluency compared with the Aβ/ND group. In contrast, the Aβ/ND+ group showed steeper decline only in R-O figure immediate and delayed recall compared with the Aβ/ND reference group.

When excluding participants with MCI at the time of imaging, results were highly similar, as shown in Table 3. A total of 38 of the 140 patients with NC (27.1%) were classified as Aβ/ND, 26 (18.6%) were classified as Aβ+/ND, 28 (20.0%) were classified as Aβ/ND+, and 48 (34.3%) were classified as Aβ+/ND+. The Aβ+/ND+ group showed steeper decline in most of the neuropsychological tests compared with the Aβ/ND group (exceptions were: R-O figure copy condition, phonemic fluency, and Stroop control condition). As in the primary analyses, all 3 groups showed steeper decline in R-O figure immediate and delayed recall. In addition, the Aβ+/ND group showed steeper decline in TMT B, CVLT learning trials and delayed recall, and semantic fluency.

In the sensitivity analysis shifting the baseline of cognitive trajectories to the time of imaging, patterns of results were generally similar but with fewer significant contrasts. Differences to the primary analyses (Table 2) were as follows: CVLT delayed recall did not decline more for the Aβ+/ND group; R-O figure delayed recall did not decline more for the Aβ+/ND and Aβ/ND+ groups; TMT-A did not decline more for the Aβ+/ND+ group; TMT-B did not decline more for the Aβ+/ND group; Stroop color-word interference, R-O copy, and Block Design did not decline more for the Aβ+/ND+ group; BNT did not decline more for the Aβ+/ND group; and phonemic fluency declined more for the Aβ+/ND+ group (eTable in the Supplement).

Discussion

This study compares long-term cognitive trajectories for a mean of 12 years of biomarker groups classified by Aβ deposition and hippocampal atrophy near the middle of the cognitive trajectories. Although the study design is not typical, the duration of neuropsychological observation is among the longest,7,17,18,32 to our knowledge. Furthermore, the participants were among the oldest old; thus, the study expands the current literature investigating ND and Aβ deposition in older adults without dementia.

The distribution of the biomarker groups is generally consistent with previous observations and prediction. At the time of imaging, the mean (SD) age of this study cohort was 86.0 (2.9) years, and the proportions of the 4 groups were 24.0% Aβ/ND, 18.3% Aβ+/ND, 20.0% Aβ/ND+, and 37.7% Aβ+/ND+. Jack et al33 reported a similar distribution in 80 individuals who were 85 to 89 years of age: 21% were Aβ/ND, 20% were Aβ+/ND, 25% were Aβ/ND+, and 34% were Aβ+/ND+. Compared with younger old cohorts, our finding of 37.7% Aβ+/ND+ is higher than in many previous studies (9% in the study by Burnham et al,18 17% in the study by Mormino et al,7 13% in the study by Soldan et al,17 and 16% in the study by Jack et al3). Consistent with results of postmortem and imaging studies,34,35 advanced aging is associated with increased Aβ accumulation and hippocampal ND. Other studies observed Aβ/ND+ (ie, SNAP) in approximately 23% of individuals with NC3,4,7,11-13 and approximately 25% of individuals with MCI,11 slightly higher than the 20.0% we report here. Many of these studies defined ND as 18[F] fluorodeoxyglucose (FDG) abnormality on PET results or hippocampal atrophy on MRI findings,33 encompassing more individuals in the Aβ/ND+ group than in our study, which operationalized ND using only hippocampal atrophy. Although HV reduction is not unique to AD (other possible causes include hippocampal sclerosis,36 frontotemporal lobar degeneration,37 argyrophilic grain disease,38 and ischemia39), it is predictive of cognitive status in AD.36

At baseline, we did not observe significant cognitive differences among the 4 groups on most tests, consistent with Mormino et al.7 However, other studies found lower baseline cognition, mostly in memory, for Aβ+/ND+,12,13,17 and another study reported lower baseline global cognitive measures for SNAP.18 Regarding cognitive decline, our findings show that in the oldest old, presence of both Aβ and ND markers together confers the fastest rate of cognitive decline across domains, consistent with results of previous studies7,17,40 and consistent with the National Institute on Aging–Alzheimer’s Association preclinical stage model.3 We also observed that Aβ deposition without ND (Aβ+/ND) is associated with decline in memory, executive functions, and language, while isolated hippocampal atrophy (Aβ/ND+; SNAP) was associated only with decline in memory. Our findings are consistent with those of a study by Burnham et al,18 in which Aβ+/ND showed a steeper decline trajectory for 8 years than did Aβ/ND in verbal memory, while SNAP did not. Owing to the unique length of follow-up in our study, we found the quadratic interaction fit for time can better reflect the accelerated cognitive decline pattern over years, unlike the linear fit used in previous studies to approximate cognitive decline trajectories.7,17,18 This finding suggests that cognitive decline often does not follow a linear decline pattern over time. Although a previous study demonstrated diminished practice effect in the Aβ+/ND and Aβ/ND+ groups compared with the Aβ/ND group,7 the extended length of follow-up and advanced age of participants in this study allows us to observe absolute and relative cognitive decline on most tests.

A previous study reported that Aβ+ participants show steeper decline in visual memory, semantic fluency, and executive function 8 years prior to imaging compared with Aβ participants.25 The present study confirms and extends these findings in that both of the Aβ+ groups (Aβ+/ND and Aβ+/ND+) exhibit steeper decline for a longer observation period, including 8 years of prospective follow-up, on these cognitive measures. The association between Aβ and long-term decline in extramemory cognitive processes, particularly executive functions and language, is consistent with an established literature showing that executive function and semantic fluency deficits are very early predictors of AD.41-43

This study included individuals with both NC and MCI at the time of imaging. When restricting the sample to participants with NC in 2009, however, the results were highly similar, indicating that the larger cohort findings were not driven by the MCI group. Because the imaging performed in 2009 lagged behind the initial cognitive assessment performed between 2000 and 2002, biomarker status at initial intake was unknown. In a sensitivity analysis that excluded all cognitive assessments prior to imaging, the patterns of effects were similar, although there were fewer significant group contrasts, likely owing to reduced power. This analysis conveys the power of baseline imaging to estimate subsequent cognitive decline, most significantly for the oldest old individuals with Aβ+/ND+.

Operationalization with different ND markers contributes significantly to variability among studies of cognitive decline. For example, differences in relative risk for cognitive decline in Aβ+/ND compared with SNAP depend on whether cerebrospinal fluid τ or imaging (HV or FDG) was used to define SNAP.11 Because HV and FDG hypometabolism are not completely concordant,44 this study simplifies the complexity of ND biomarkers by focusing only on HV. Regarding the underlying pathologic findings of SNAP, Crary et al45 noted similarities between primary age-related tauopathy and SNAP and proposed a correspondence. Like SNAP, primary age-related tauopathy has a lower prevalence of APOE*4 (OMIM 107741), its prevalence increases with age, and it prominently involves medial temporal lobe pathologic findings.11,45 However, Mormino et al19 reported comparable τ PET imaging results between SNAP (classified based on HV and FDG hypometabolism) and Aβ/ND groups, challenging the hypothesis of a close correspondence between SNAP and primary age-related tauopathy. Gordon et al,46 investigating longitudinal changes in PiB and HV, found that SNAP was similar to Aβ/ND in the rate of accumulation of Aβ and rate of hippocampal atrophy; thus, they concluded that SNAP is more likely to reflect inherent variability in brain structure than early AD. Our cognitive slope findings support the notion that SNAP is dissimilar to early AD in that no domains other than memory decline in SNAP, whereas isolated Aβ is associated with extramemory decline (namely, language and executive function). We observed some evidence of a progressive process in SNAP (ie, visual memory decline greater than the nonpathologic reference), whereas in the younger cohort (mean age, 66 years), SNAP was not associated with faster rate of hippocampal atrophy than other groups.46

These findings further the understanding of AD pathophysiologic conditions in the oldest old. Postmortem studies indicate that after 80 years of age, presence of Aβ plaques (and neurofibrillary tangles) no longer discriminates between clinical dementia and nondementia cases.47,48 Furthermore, pathologic heterogeneity of dementia increases with advanced age.49 Weakening of associations between Aβ and cognition with advanced age suggests that AD may be a different disease above 80 years of age than below. If true, the argument has been made that Aβ is likely a misguided therapeutic target, as it may reflect an inherent process of advanced aging.50,51 As reported previously in this same oldest old cohort, Aβ deposition was not associated with the incidence of dementia during 2 years among the highest-risk participants with MCI (although it was associated with incidence of dementia among all participants).52 However, the more sensitive cognitive outcomes we now report for a 12-year follow-up indicate that, despite frequent occurrence of Aβ deposition at this age (98 of 175 [56.0%]), Aβ is associated with long-term cognitive decline compared with its absence. The same finding was observed for the frequent occurrence of hippocampal volume reduction (101 of 175 [57.7%]) vs its absence, although cognitive decline was restricted to visual memory. Both biomarkers appear to have measurable cognitive consequences and are hallmarks of decline, even among the oldest old. Although this study does not refute the notion of a diminished role of Aβ with advanced age, results support the hypothesis that Aβ remains functionally consequential in advanced aging and thus remains an important, if not sufficient, pathophysiologic process.

Limitations

This study has some limitations. Because the imaging lagged behind the initial cognitive assessment, biomarker status at initial intake was unknown. In addition, ND was operationalized only by HV, simplifying the complexity of ND biomarkers. Finally, the participants were relatively highly educated and mostly of white European descent; the findings may not generalize to other populations.

Conclusions

In the oldest old, the presence of both Aβ deposition and reduced hippocampal volume is common and confers the greatest risk for cognitive decline across domains. Neither biomarker abnormality is benign in the ninth and 10th decades of life. Isolated Aβ deposition is associated with decline in memory, executive functions, and some aspects of language, while isolated hippocampal atrophy is associated only with decline in memory. Suspected non-Alzheimer disease pathophysiology and isolated Aβ exhibit distinct cognitive decline profiles, which suggests different underlying pathophysiologic processes.

Back to top
Article Information

Corresponding Author: Beth E. Snitz, PhD, Department of Neurology, University of Pittsburgh, 3501 Forbes Ave, Ste 830, Pittsburgh, PA 15215 (snitzbe@upmc.edu).

Accepted for Publication: August 6, 2017.

Published Online: November 6, 2017. doi:10.1001/jamaneurol.2017.3029

Author Contributions: Dr Snitz had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Lopez, Aizenstein, DeKosky, Klunk, Snitz.

Acquisition, analysis, or interpretation of data: Zhao, Tudorascu, Lopez, Cohen, Mathis, Aizenstein, Price, Kuller, Kamboh, DeKosky, Snitz.

Drafting of the manuscript: Zhao, Tudorascu, Price, Snitz.

Critical revision of the manuscript for important intellectual content: Lopez, Cohen, Mathis, Aizenstein, Kuller, Kamboh, DeKosky, Klunk, Snitz.

Statistical analysis: Zhao, Tudorascu, Kuller.

Obtained funding: Lopez, Aizenstein, DeKosky, Klunk, Snitz.

Administrative, technical, or material support: Lopez, Cohen, Mathis, Price, Kuller, Kamboh, DeKosky, Klunk.

Study supervision: Lopez, Aizenstein, Kamboh, Klunk, Snitz.

Conflict of Interest Disclosures: General Electric Healthcare holds a license agreement with the University of Pittsburgh based on the technology described in this article. Drs Mathis and Klunk are coinventors of Pittsburgh Compound B and, as such, have a financial interest in this license agreement. General Electric Healthcare provided no grant support for this study and had no role in the design or interpretation of results or preparation of this manuscript. Dr DeKosky reported serving as a consultant for Amgen, Acumen Pharmaceuticals, Biogen, and Cognition Therapeutics. No other conflicts were reported.

Funding/Support: This study was supported by grant U01AT000162 from the National Center for Complementary and Integrative Health and the Office of Dietary Supplements and by support from the National Institute on Aging, National Heart, Lung, and Blood Institute; program project grant P01 AG025204 from the National Institute on Aging, National Institutes of Health; and grants P50AG05133, AG030653, and AG041718 from the University of Pittsburgh Alzheimer’s Disease Research Center (Dr Kamboh).

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

References
1.
Jack  CR  Jr, Wiste  HJ, Knopman  DS,  et al.  Rates of β-amyloid accumulation are independent of hippocampal neurodegeneration.  Neurology. 2014;82(18):1605-1612.PubMedGoogle ScholarCrossref
2.
Sperling  RA, Aisen  PS, Beckett  LA,  et al.  Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease.  Alzheimers Dement. 2011;7(3):280-292.PubMedGoogle ScholarCrossref
3.
Jack  CR  Jr, Knopman  DS, Weigand  SD,  et al.  An operational approach to National Institute on Aging–Alzheimer’s Association criteria for preclinical Alzheimer disease.  Ann Neurol. 2012;71(6):765-775.PubMedGoogle ScholarCrossref
4.
Toledo  JB, Weiner  MW, Wolk  DA,  et al; Alzheimer’s Disease Neuroimaging Initiative.  Neuronal injury biomarkers and prognosis in ADNI subjects with normal cognition.  Acta Neuropathol Commun. 2014;2:26.PubMedGoogle ScholarCrossref
5.
Petersen  RC, Aisen  P, Boeve  BF,  et al.  Mild cognitive impairment due to Alzheimer disease in the community.  Ann Neurol. 2013;74(2):199-208.PubMedGoogle Scholar
6.
Caroli  A, Prestia  A, Galluzzi  S,  et al; Alzheimer’s Disease Neuroimaging Initiative.  Mild cognitive impairment with suspected nonamyloid pathology (SNAP): prediction of progression.  Neurology. 2015;84(5):508-515.PubMedGoogle ScholarCrossref
7.
Mormino  EC, Betensky  RA, Hedden  T,  et al.  Synergistic effect of β-amyloid and neurodegeneration on cognitive decline in clinically normal individuals.  JAMA Neurol. 2014;71(11):1379-1385.PubMedGoogle ScholarCrossref
8.
Vos  SJ, Verhey  F, Frölich  L,  et al; Alzheimer’s Disease Neuroimaging Initiative.  Prevalence and prognosis of Alzheimer’s disease at the mild cognitive impairment stage.  Brain. 2015;138(pt 5):1327-1338.PubMedGoogle ScholarCrossref
9.
Roe  CM, Fagan  AM, Grant  EA,  et al.  Amyloid imaging and CSF biomarkers in predicting cognitive impairment up to 7.5 years later.  Neurology. 2013;80(19):1784-1791.PubMedGoogle ScholarCrossref
10.
Wirth  M, Villeneuve  S, Haase  CM,  et al.  Associations between Alzheimer disease biomarkers, neurodegeneration, and cognition in cognitively normal older people.  JAMA Neurol. 2013;70(12):1512-1519.PubMedGoogle Scholar
11.
Jack  CR  Jr, Knopman  DS, Chételat  G,  et al.  Suspected non-Alzheimer disease pathophysiology—concept and controversy.  Nat Rev Neurol. 2016;12(2):117-124.PubMedGoogle ScholarCrossref
12.
Vos  SJ, Xiong  C, Visser  PJ,  et al.  Preclinical Alzheimer’s disease and its outcome: a longitudinal cohort study.  Lancet Neurol. 2013;12(10):957-965.PubMedGoogle ScholarCrossref
13.
van Harten  AC, Smits  LL, Teunissen  CE,  et al.  Preclinical AD predicts decline in memory and executive functions in subjective complaints.  Neurology. 2013;81(16):1409-1416.PubMedGoogle ScholarCrossref
14.
Wisse  LEM, Butala  N, Das  SR,  et al; Alzheimer’s Disease Neuroimaging Initiative.  Suspected non-AD pathology in mild cognitive impairment.  Neurobiol Aging. 2015;36(12):3152-3162.PubMedGoogle ScholarCrossref
15.
Prestia  A, Caroli  A, van der Flier  WM,  et al.  Prediction of dementia in MCI patients based on core diagnostic markers for Alzheimer disease.  Neurology. 2013;80(11):1048-1056.PubMedGoogle ScholarCrossref
16.
Knopman  DS, Jack  CR  Jr, Wiste  HJ,  et al.  Short-term clinical outcomes for stages of NIA-AA preclinical Alzheimer disease.  Neurology. 2012;78(20):1576-1582.PubMedGoogle ScholarCrossref
17.
Soldan  A, Pettigrew  C, Cai  Q,  et al; BIOCARD Research Team.  Hypothetical preclinical Alzheimer disease groups and longitudinal cognitive change.  JAMA Neurol. 2016;73(6):698-705.PubMedGoogle ScholarCrossref
18.
Burnham  SC, Bourgeat  P, Doré  V,  et al; AIBL Research Group.  Clinical and cognitive trajectories in cognitively healthy elderly individuals with suspected non-Alzheimer’s disease pathophysiology (SNAP) or Alzheimer’s disease pathology: a longitudinal study.  Lancet Neurol. 2016;15(10):1044-1053.PubMedGoogle ScholarCrossref
19.
Mormino  EC, Papp  KV, Rentz  DM,  et al.  Heterogeneity in suspected non–Alzheimer disease pathophysiology among clinically normal older individuals.  JAMA Neurol. 2016;73(10):1185-1191.PubMedGoogle ScholarCrossref
20.
Bennett  DA, Schneider  JA, Arvanitakis  Z,  et al.  Neuropathology of older persons without cognitive impairment from two community-based studies.  Neurology. 2006;66(12):1837-1844.PubMedGoogle ScholarCrossref
21.
Aizenstein  HJ, Nebes  RD, Saxton  JA,  et al.  Frequent amyloid deposition without significant cognitive impairment among the elderly.  Arch Neurol. 2008;65(11):1509-1517.PubMedGoogle ScholarCrossref
22.
Sonnen  JA, Santa Cruz  K, Hemmy  LS,  et al.  Ecology of the aging human brain.  Arch Neurol. 2011;68(8):1049-1056.PubMedGoogle ScholarCrossref
23.
DeKosky  ST, Williamson  JD, Fitzpatrick  AL,  et al; Ginkgo Evaluation of Memory (GEM) Study Investigators.  Ginkgo biloba for prevention of dementia: a randomized controlled trial.  JAMA. 2008;300(19):2253-2262.PubMedGoogle ScholarCrossref
24.
DeKosky  ST, Fitzpatrick  A, Ives  DG,  et al; GEMS Investigators.  The Ginkgo Evaluation of Memory (GEM) study: design and baseline data of a randomized trial of Ginkgo biloba extract in prevention of dementia.  Contemp Clin Trials. 2006;27(3):238-253.PubMedGoogle ScholarCrossref
25.
Snitz  BE, Weissfeld  LA, Lopez  OL,  et al.  Cognitive trajectories associated with β-amyloid deposition in the oldest-old without dementia.  Neurology. 2013;80(15):1378-1384.PubMedGoogle ScholarCrossref
26.
Lezak  MD, Howieson  DB, Loring  DW.  Neuropsychological Assessment. 4th ed. New York, NY: Oxford University Press; 2004.
27.
Becker  JT, Boller  F, Saxton  J, McGonigle-Gibson  KL.  Normal rates of forgetting of verbal and non-verbal material in Alzheimer’s disease.  Cortex. 1987;23(1):59-72.PubMedGoogle ScholarCrossref
28.
Trenerry  MR, Crosson  B, DeBoe  J, Leber  WR.  Stroop Neuropsychological Screening Test. Odessa, FL: Psychological Assessment Resources; 1989.
29.
Lopez  OL, Becker  JT, Jagust  WJ,  et al.  Neuropsychological characteristics of mild cognitive impairment subgroups.  J Neurol Neurosurg Psychiatry. 2006;77(2):159-165.PubMedGoogle ScholarCrossref
30.
Saxton  J, Ratcliff  G, Munro  CA,  et al.  Normative data on the Boston Naming Test and two equivalent 30-item short forms.  Clin Neuropsychol. 2000;14(4):526-534.PubMedGoogle ScholarCrossref
31.
Fai  AH-T, Cornelius  PL.  Approximate F-tests of multiple degree of freedom hypotheses in generalized least squares analyses of unbalanced split-plot experiments.  J Stat Comput Simul. 1996;54(4):363-378.Google ScholarCrossref
32.
Wirth  M, Oh  H, Mormino  EC, Markley  C, Landau  SM, Jagust  WJ.  The effect of amyloid β on cognitive decline is modulated by neural integrity in cognitively normal elderly.  Alzheimers Dement. 2013;9(6):687-698.e1.PubMedGoogle ScholarCrossref
33.
Jack  CR  Jr, Wiste  HJ, Weigand  SD,  et al.  Age-specific population frequencies of cerebral β-amyloidosis and neurodegeneration among people with normal cognitive function aged 50-89 years: a cross-sectional study.  Lancet Neurol. 2014;13(10):997-1005.PubMedGoogle ScholarCrossref
34.
Braak  H, Braak  E.  Frequency of stages of Alzheimer-related lesions in different age categories.  Neurobiol Aging. 1997;18(4):351-357.PubMedGoogle ScholarCrossref
35.
Raz  N, Lindenberger  U, Rodrigue  KM,  et al.  Regional brain changes in aging healthy adults: general trends, individual differences and modifiers.  Cereb Cortex. 2005;15(11):1676-1689.PubMedGoogle ScholarCrossref
36.
Jack  CR  Jr, Dickson  DW, Parisi  JE,  et al.  Antemortem MRI findings correlate with hippocampal neuropathology in typical aging and dementia.  Neurology. 2002;58(5):750-757.PubMedGoogle ScholarCrossref
37.
Whitwell  JL, Jack  CR  Jr, Parisi  JE,  et al.  Does TDP-43 type confer a distinct pattern of atrophy in frontotemporal lobar degeneration?  Neurology. 2010;75(24):2212-2220.PubMedGoogle ScholarCrossref
38.
Jack  CR  Jr.  PART and SNAP.  Acta Neuropathol. 2014;128(6):773-776. PubMedGoogle ScholarCrossref
39.
Di Paola  M, Caltagirone  C, Fadda  L, Sabatini  U, Serra  L, Carlesimo  GA.  Hippocampal atrophy is the critical brain change in patients with hypoxic amnesia.  Hippocampus. 2008;18(7):719-728.PubMedGoogle ScholarCrossref
40.
Mormino  EC, Papp  KV.  Cognitive decline in preclinical stage 2 Alzheimer disease and implications for prevention trials.  JAMA Neurol. 2016;73(6):640-642.PubMedGoogle ScholarCrossref
41.
Bäckman  L, Jones  S, Berger  AK, Laukka  EJ, Small  BJ.  Cognitive impairment in preclinical Alzheimer’s disease: a meta-analysis.  Neuropsychology. 2005;19(4):520-531.PubMedGoogle ScholarCrossref
42.
Saxton  J, Lopez  OL, Ratcliff  G,  et al.  Preclinical Alzheimer disease: neuropsychological test performance 1.5 to 8 years prior to onset.  Neurology. 2004;63(12):2341-2347.PubMedGoogle ScholarCrossref
43.
Masur  DM, Sliwinski  M, Lipton  RB, Blau  AD, Crystal  HA.  Neuropsychological prediction of dementia and the absence of dementia in healthy elderly persons.  Neurology. 1994;44(8):1427-1432.PubMedGoogle ScholarCrossref
44.
Klunk  WE, Perani  D.  Amyloid and neurodegeneration: converging and diverging paths.  Neurology. 2013;81(20):1728-1729.PubMedGoogle ScholarCrossref
45.
Crary  JF, Trojanowski  JQ, Schneider  JA,  et al.  Primary age-related tauopathy (PART): a common pathology associated with human aging.  Acta Neuropathol. 2014;128(6):755-766.PubMedGoogle ScholarCrossref
46.
Gordon  BA, Blazey  T, Su  Y,  et al.  Longitudinal β-amyloid deposition and hippocampal volume in preclinical Alzheimer disease and suspected non–Alzheimer disease pathophysiology.  JAMA Neurol. 2016;73(10):1192-1200.PubMedGoogle ScholarCrossref
47.
Savva  GM, Wharton  SB, Ince  PG, Forster  G, Matthews  FE, Brayne  C; Medical Research Council Cognitive Function and Ageing Study.  Age, neuropathology, and dementia.  N Engl J Med. 2009;360(22):2302-2309.PubMedGoogle ScholarCrossref
48.
Haroutunian  V, Schnaider-Beeri  M, Schmeidler  J,  et al.  Role of the neuropathology of Alzheimer disease in dementia in the oldest-old.  Arch Neurol. 2008;65(9):1211-1217.PubMedGoogle ScholarCrossref
49.
Kawas  CH, Kim  RC, Sonnen  JA, Bullain  SS, Trieu  T, Corrada  MM.  Multiple pathologies are common and related to dementia in the oldest-old: the 90+ Study.  Neurology. 2015;85(6):535-542.PubMedGoogle ScholarCrossref
50.
Richard  E, Schmand  B, Eikelenboom  P, Westendorp  RG, Van Gool  WA.  The Alzheimer myth and biomarker research in dementia.  J Alzheimers Dis. 2012;31(suppl 3):S203-S209.PubMedGoogle Scholar
51.
Kuller  LH, Lopez  OL.  Dementia and Alzheimer’s disease: a new direction: the 2010 Jay L. Foster Memorial Lecture.  Alzheimers Dement. 2011;7(5):540-550.PubMedGoogle ScholarCrossref
52.
Lopez  OL, Klunk  WE, Mathis  C,  et al.  Amyloid, neurodegeneration, and small vessel disease as predictors of dementia in the oldest-old.  Neurology. 2014;83(20):1804-1811.PubMedGoogle ScholarCrossref
×