Prclinical Stages Nics Adra 2011

8/9/2019 Prclinical Stages Nics Adra 2011

http://slidepdf.com/reader/full/prclinical-stages-nics-adra-2011 1/13

Toward dening the preclinical stages of Alzheimer’s disease:Recommendations from the National Institute on Aging and the

Alzheimer’s Association workgroupReisa A. Sperlin ga ,*, Paul S. Aisen b , Laurel A. Becket tc , David A. Bennett d , Suzanne Craf te,Anne M. Fagan f , Takeshi Iwatsub og , Clifford R. Jack h , Jef frey Kaye i, Thomas J. Montin e j,Denise C. Park k , Eric M. Reiman l, Christopher C. Rowe m , Eric Siemers n , Yaakov Stern o ,

Kristine Yaffe p , Maria C. Carrillo q , Bill Thies q , Marcelle Morrison-Bogorad r, Molly V. Wagster r,Creighton H. Phelps r

a Center for Alzheimer Research and Treatment, Department of Neurology, Brigham and Women’s Hospital, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA

b Department of Neurosciences, University of California San Diego, San Diego, CA, USAc Division of Biostatistics, School of Medicine, University of California, Davis, CA, USA

d Rush Alzheimer’s Disease Center, Rush University Medical Center, Chicago, IL, USAeGeriatric Research, Education, and Clinical Center, Veterans Affairs Puget Sound; Department of Psychiatry and Behavioral Sciences,

University of Washington School of Medicine, Seattle, WA, USA f Department of Neurology, Washington University School of Medicine, St. Louis, MO, USA

g Department of Neuropathology, Graduate School of Medicine, University of Tokyo, Tokyo, Japanh Department of Radiology, Mayo Clinic Minnesota, Rochester, MN, USA

i Departments of Neurology and Biomedical Engineering, Layton Aging & Alzheimer’s Disease Center, Oregon Center for Aging & Technology,Oregon Health & Science University and Portland Veteran’s Affairs Medical Center, Portland, OR, USA

j Department of Pathology, University of Washington, Seattle, WA, USAk Center for Vital Longevity, University of Texas at Dallas, Dallas, TX, USA

l Banner Alzheimer’s Institute, Phoenix, AZ, USAm Austin Health, University of Melbourne, Melbourne, Australia

nEli Lilly and Company, Indianapolis, IN, USAoCognitive Neuroscience Division, Taub Institute, Columbia University College of Physicians and Surgeons, New York, NY, USA

p Departments of Psychiatry, Neurology, and Epidemiology and Biostatistics, University of California San Francisco, San Francisco VA Medical Center,San Francisco, CA, USA

q Alzheimer’s Association, Chicago, IL, USAr Division of Neuroscience, National Institute on Aging, Bethesda, MD, USA

Abstract The pathophysiological process of Alzheimer’s disease (AD) is thought to begin many yearsbefore the diagnosis of AD dementia. This long “preclinical” phase of AD would provide a criticalopportunity for therapeutic intervention; however, we need to further elucidate the link between the

pathological cascade of AD and the emergence of clinical symptoms. The National Institute onAging and the Alzheimer’s Association convened an international workgroup to review the bio-marker, epidemiological, and neuropsychological evidence, and to develop recommendations todetermine the factors which best predict the risk of progression from “normal” cognition tomild cognitive impairment and AD dementia. We propose a conceptual framework and operationalresearch criteria, based on the prevailing scientic evidence to date, to test and rene these modelswith longitudinal clinical research studies. These recommendations are solely intended for researchpurposes and do not have any clinical implications at this time. It is hoped that these recommen-

*Corresponding author. Tel.: 1 1-617-732-8085; Fax: 1 1-617-264-5212.E-mail address: [email protected]

1552-5260/$ - see front matter 2011 The Alzheimer’s Association. All rights reserved.doi:10.1016/j.jalz.2011.03.003

Alzheimer’s & Dementia - (2011) 1–13

mailto:[email protected]

http://dx.doi.org/10.1016/j.jalz.2011.03.003






mailto:[email protected]



dations will provide a common rubric to advance the study of preclinical AD, and ultimately, aidthe eld in moving toward earlier intervention at a stage of AD when some disease-modifying ther-apies may be most efcacious.

2011 The Alzheimer’s Association. All rights reserved.

Keywords: Preclinical Alzheimer’s disease; Biomarker; Amyloid; Neurodegeneration; Prevention

1. Introduction

Converging evidence from both genetic at-risk cohortsand clinically normal older individuals suggests that thepathophysiological process of Alzheimer’s disease (AD)begins years, if not decades, before the diagnosis of clinicaldementia [1]. Recent advances in neuroimaging, cerebrospi-nal uid (CSF) assays, and other biomarkers now provide theability to detect evidence of the AD pathophysiological pro-cess in vivo. Emerging data in clinically normal older indi-

viduals suggest that biomarker evidence of amyloid beta(Ab ) accumulation is associated with functional and struc-tural brain alterations, consistent with the patterns of abnor-mality seen in patients with mild cognitive impairment(MCI) and AD dementia. Furthermore, clinical cohortstudies suggest that there may be very subtle cognitivealterations that are detectable years before meeting criteriafor MCI, and that predict progression to AD dementia. It isalso clear, however, that some older individuals with thepathophysiological process of AD may not become symp-tomatic during their lifetime. Thus, it is critical to better de-ne the biomarker and/or cognitive prole that best predicts

progression from the preclinical to the clinical stages of MCIand AD dementia. The long preclinical phase of AD pro-vides a critical opportunity for potential intervention withdisease-modifying therapy, if we are able to elucidate thelink between the pathophysiological process of AD andthe emergence of the clinical syndrome.

A recent report on the economic implications of the im-pending epidemic of AD, as the “baby boomer” generationages, suggests that more than 13.5 million individuals justin the United States will manifest AD dementia by the year2050 (http://www.alz.org/alzheimers_disease_trajectory.asp). A hypothetical intervention that delayed the onset of

AD dementia by 5 years would result in a 57% reductionin the number of patients with AD dementia, and reducethe projected Medicare costs of AD from $627 to $344billion dollars. Screening and treatment programs institutedfor other diseases, such as cholesterol screening for cardio-vascular and cerebrovascular disease and colonoscopy forcolorectal cancer, have already been associated with a de-crease in mortality because of these conditions. The currentlifetime risk of AD dementia for a 65-year-old is estimated tobe at 10.5%. Recent statistical models suggest that a screen-ing instrument for markersof the pathophysiological processof AD (with 90% sensitivity and specicity) and a treatment

that slows down progression by 50% would reduce that risk to 5.7%.

Both laboratory work and recent disappointing clinicaltrial results raise the possibility that therapeutic interven-tions applied earlier in the course of AD would be morelikely to achieve disease modication. Studies with trans-genic mouse models suggest that A b -modifying therapiesmay have limited effect after neuronal degeneration has be-gun. Several recent clinical trials involving the stages of mildto moderate dementia have failed to demonstrate clinicalbenet, even in the setting of biomarker or autopsy evidenceof decreased A b burden. Although the eld is already mov-ing to earlier clinical trials at the stage of MCI, it is possiblethat similar to cardiac disease and cancer treatment, ADwould be optimally treated before signicant cognitiveimpairment, in the “presymptomatic” or “preclinical” stagesof AD. Secondary prevention studies, which would treat“normal” or asymptomatic individuals or those with subtleevidence of impairment due to AD so as to delay the onsetof full-blown clinical symptoms, are already in the planningstages. The overarching therapeutic objective of thesepreclinical studies would be to treat early pathological pro-cesses (e.g., lower A b burden or decrease neurobrillary tan-gle pathology) to prevent subsequent neurodegeneration and

eventual cognitive decline.For these reasons, our working group sought to examine

the evidence for a denable preclinical stage of AD, and toreview the biomarker, epidemiological, and neuropsycho-logical factors that best predict the risk of progression fromasymptomatic to MCIandAD dementia.To narrow thescopeof our task, we chose to specically focus on predictors of cognitivedecline thought to be due to the pathophysiologicalprocess of AD. We did not address cognitive aging in theabsence of recognized pathological changes in the brain, orcognitivedecline becauseof other common age-related braindiseases; however, we readily acknowledge that these brain

diseases, in particular, cerebrovascular disease, Lewy bodydisease, and other neurodegenerative processes, may signif-icantly inuence clinical manifestations of AD and possiblyits pathophysiology. Although there are likely lifelongcharacteristics and midlife risk factors that inuence thelikelihood of developing cognitive impairment late in life,for feasibility in current studies, we chose to focus onthe 10-year period before the emergence of cognitiveimpairment.

Furthermore,we proposea research framework to providea common language to advance the scientic understandingof the preclinical stages of AD and a foundation for the eval-

uation of preclinical AD treatments. These criteria areintended purely for research purposes, and have no clinical

R.A. Sperling et al. / Alzheimer’s & Dementia - (2011) 1–132

http://www.alz.org/alzheimers_disease_trajectory.asp








for the study of individuals who have evidence of early ADpathological changes but do not meet clinical criteria forMCI or dementia. It is likely that even this preclinicalstage of the disease represents a continuum fromcompletely asymptomatic individuals with biomarkerevidence suggestive of AD-P at risk for progression to ADdementia to biomarker-positive individuals who are alreadydemonstrating very subtle decline but not yet meeting stan-dardized criteria for MCI (refer to accompanying MCI work-group recommendations by Albert et al). This latter group of individuals might be classied as “Not normal, not MCI” butwould be included under the rubric of preclinical AD(Fig. 1). Importantly, this continuum of preclinical ADwould also encompass (1) individuals who carry one ormore apolipoprotein E ( APOE ) 3 4 alleles who are knownto have an increased risk of developing AD dementia, at the point they are AD-P biomarker-positive , and (2) carriersof autosomal dominant mutations, who are in the presymp-tomatic biomarker-positive stage of their illness, and whowill almost certainly manifest clinical symptoms and prog-ress to dementia.

Our group carefully considered several monikers to bestcapture this stage of the disease, including “asymptomatic,”“presymptomatic,” “latent,” “premanifest,” and “preclini-cal.” The term “preclinical” was felt to best encompassthis conceptual phase of the disease process but is not meantto imply that all individuals who have evidence of early ADpathology will necessarily progress to clinical AD dementia.Individuals who are biomarker positive but cognitively nor-mal might currently be dened as “asymptomatic at risk for

AD dementia.” Indeed, our goal is to better dene the factorswhich best predict cognitive decline in biomarker-positiveindividuals, so as to move toward an accurate prole of pre-clinical AD.

4. Models of the pathophysiological sequence of AD

To facilitate the discussion of the concept of a preclinicalstage of AD, we propose a theoretical model of the patho-physiological cascade of AD ( Fig. 2). It is important toacknowledge that this model, although based on the prevail-ing evidence, may be incorrect, is certainly incomplete, and

will evolve as additional laboratory and clinical studies arecompleted. Indeed, this model should be viewed as an initialattempt to bring together multiple areas of research into ourbest estimate of a more coherent whole.

The proposed model of AD views A b peptide accumula-tion as a key early event in the pathophysiological process of AD. However, we acknowledge that the etiology of ADremains uncertain, and some investigators have proposedthat synaptic, mitochondrial, metabolic, inammatory, neu-ronal, cytoskeletal, and other age-related alterations mayplay an even earlier, or more central, role than A b peptidesin the pathogenesis of AD [6,7]. There also remains

signicant debate in the eld as to whether abnormalprocessing versus clearance of A b 42 is the etiologic event

in sporadic, late-onset AD [8]. Some investigators have sug-gested that sequestration of A b into brillar forms may evenserve as a protective mechanism against oligomeric species,which may be the more synaptotoxic forms of A b [9–11] .However, of all the known autosomal dominant, earlyonset forms of AD are thought to be, at least in part, dueto alterations in amyloid precursor protein ( APP )production or cleavage. Similarly, trisomy 21 invariably

Fig. 1. Model of the clinical trajectory of Alzheimer’s disease (AD). Thestage of preclinical AD precedes mild cognitive impairment (MCI) andencompassesthe spectrum of presymptomaticautosomal dominant mutationcarriers, asymptomatic biomarker-positive older individuals at risk for pro-gression to MCI due to AD and AD dementia, as well as biomarker-positiveindividuals who have demonstrated subtle decline from their own baselinethat exceeds that expected in typical aging, but would not yet meet criteria

for MCI. Note that this diagram represents a hypothetical model for thepathological-clinical continuum of AD but does not imply that allindividuals with biomarker evidence of AD-pathophysiological processwill progress to the clinical phases of the illness.

Fig. 2. Hypotheticalmodel of the Alzheimer’s disease (AD)pathophysiolog-ical sequence leading to cognitive impairment. This model postulates thatamyloid beta (A b ) accumulation is an “upstream” event in the cascade thatis associated with “downstream” synaptic dysfunction, neurodegeneration,and eventual neuronal loss. Note that although recent work from animalmodels suggests thatspecicformsof A b maycauseboth functional andmor-phological synaptic changes, it remains unknown whether A b is sufcient toincite the neurodegenerative process in sporadic late-onset AD. Age andgenetics, as well as other specic host factors, such as brain and cognitivereserve, or other brain diseases may inuence the response to A b and/orthe pace of progression toward the clinical manifestations of AD.




results in AD-P in individuals who have three intact copiesof the APP coding region located on chromosome 21.Finally, APOE , the major genetic risk factor for late-onsetAD, has been implicated in amyloid trafcking and plaqueclearance. Both autopsy and biomarker studies (see later inthe text) similarly suggest that A b 42 accumulation increaseswith advanced aging, the greatest risk factor for developingAD. At this point, it remains unclear whether it is meaning-ful or feasible to make the distinction between A b as a risk factor for developing the clinical syndrome of AD versus A baccumulation as an early detectable stage of AD becausecurrent evidence suggests that both concepts are plausible.

Also, it is clear that synaptic depletion, intracellular hy-perphosphorylated forms of tau, and neuronal loss invariablyoccur in AD, and at autopsy, these markers seem to correlatebetter than plaque counts or total A b load with clinicalimpairment. Although we present evidence later that thepresence of markers of “upstream” A b accumulation is asso-ciated with markers of “downstream” pathological change,including abnormal tau, neural dysfunction, glial activation,and neuronal loss and atrophy, it remains to be proven thatAb accumulation is sufcient to incite the downstream path-ological cascade of AD. It remains unknown whether thisneurodegenerative process could be related to direct synaptictoxicity due to oligomeric forms of A b , disruption of axonaltrajectories from brillar forms of A b , or a “second hit” thatresults in synaptic dysfunction, neurodegeneration, neuro-brillary tangle formation, and eventually neuronal loss.

Epidemiological data suggest there are signicant modu-lating factors that may alter the pace of the clinical expression

of AD-P, although evidence that these factors alter the under-lying pathophysiological process itself is less secure. Largecohort studies have implicated multiple health factors thatmay increase the risk for developing cognitive decline andde-mentia thought tobecaused by AD [12]. In particular, vascularrisk factors such as hypertension, hypercholesterolemia, anddiabetes have been associated with an increased risk of dementia, and may contribute directly to the effect of ADpathology on the aging brain [13,14] . Depressive sym-ptomatology, apathy, and chronic psychological distresshave also been linked to increased risk of manifesting MCIand dementia [15–17] . It also remains unclear whether there

are specic environmental exposures, such as head trauma,that may inuence the progression of the pathophysiologicalsequence or the clinical expression of the pathology. On thepositive side, there is some evidence that engagement inspecic activities, including cognitive, physical, leisure, andsocial activity, may be associated with decreased risk of MCI and AD dementia [18].

The temporal lag between the appearance of AD-P and theemergence of AD-C also may be altered by factors such asbrain or cognitive reserve [19]. The concept of reserve wasoriginally invoked to provide an explanation for the observa-tion that the extent of AD histopathological changes at

autopsy did not always align with the degree of clinicalimpairment, and can be thought of as the ability to tolerate

higher levels of brain injurywithout exhibiting clinical symp-toms. “Brain reserve” refers to the capacity of the brain towithstandpathological insult, perhaps because of greater syn-aptic density or larger number of healthy neurons, such thatsufcient neural substrate remains to support normal func-tion. In contrast, “cognitive reserve” is thought to representthe ability to engage alternate brain networks or cognitivestrategies to cope with the effects of encroaching pathology.It is not clear, however, that the data support a sharp demarca-tion between these twoconstructs because many factors, suchas higher socioeconomic status or engagement in cognitivelystimulating activities, may contribute to both forms of reserve. Higher education and socioeconomic status havebeen associated with lower age-adjusted incidence of ADdiagnosis. Recent studies suggest that high reserve may pri-marily inuence the capability of individuals to tolerate theirAD-P for longer periods, but may also be associated withrapid decline after a “tipping point” is reached and compen-satory mechanisms begin to fail [20,21] .

5. Biomarker model of the preclinical stage of AD

A biomarker model has been recently proposed in whichthe most widely validated biomarkers of AD-P becomeabnormal and likewise reach a ceiling in an ordered manner[22]. This biomarker model parallels the hypothetical patho-physiological sequence of AD discussed previously, and isparticularly relevant to tracking the preclinical stages of AD (Fig. 3). Biomarkers of brain A b amyloidosis includereductions in CSF A b 42 and increased amyloid tracer reten-

tion on positron emission tomography (PET) imaging. Ele-vated CSF tau is not specic to AD and is thought to bea biomarker of neuronal injury. Decreased uorodeoxyglu-cose 18F (FDG) uptake on PET with a temporoparietal pat-tern of hypometabolism is a biomarker of AD-relatedsynaptic dysfunction. Brain atrophy on structural magneticresonance imaging (MRI) in a characteristic pattern involv-ing the medial temporal lobes, paralimbic and temporoparie-tal cortices is a biomarker of AD-related neurodegeneration.

This biomarker model was adapted from the originalgraph proposed by Jack et al [22] to expand the preclinicalphase, and has the following features: (1) A b accumulation

biomarkers become abnormal rst and a substantial A b loadaccumulates before the appearance of clinical symptoms.The lag phase between A b accumulation and clinical symp-toms remains to be quantied, but current theories suggestthat the lag may be for more than a decade. Similar to the hy-pothetical pathophysiological model described previously,interindividual differences in this time lag are likely causedby differences in brain reserve, cognitive reserve, and theadded contributions of coexisting pathologies. Note that inthis biomarker model, brain A b accumulation is necessarybut not sufcient to produce the clinical symptoms of MCIand dementia, (2) biomarkers of synaptic dysfunction,

including FDG and functional MRI (fMRI), may demon-strate abnormalities very early, particularly in APOE gene

R.A. Sperling et al. / Alzheimer’s & Dementia - (2011) 1–13 5



3 4 allele carriers, who may manifest functional abnormali-ties before detectable A b deposition [23–25] . The severityand change over time in these synaptic markers correlatewith clinical symptoms during MCI and AD dementia, (3)

structural MRI is thought to become abnormal a bit later,as a marker of neuronal loss, and MRI retains a closerelationship with cognitive performance through theclinical phases of MCI and dementia [26], (4) none of thebiomarkers is static; rates of change in each biomarkerchange over time and follow a nonlinear time course, whichis hypothesized to be sigmoid shaped, and (5) anatomic in-formation from imaging biomarkers provides useful diseasestaging information in that the topography of disease-relatedimaging abnormalities changes in a characteristic mannerwith disease progression.

6. Biomarker and autopsy evidence linking ADpathology to early symptomatology

Several multicenter biomarker initiatives, including theAlzheimer’s Disease Neuroimaging Initiative; the AustralianImaging, Biomarkers and Lifestyle Flagship Study of Aging;as well as major biomarker studies in preclinical populationsat several academic centers, are ongoing. These studies havealready provided preliminary evidence that biomarker abnor-malities consistent with AD pathophysiological process aredetectable before the emergence of overt clinical symptom-atology and are predictive of subsequent cognitive decline.

Many of the recent studies have focused on markers of A busing either CSF assays of A b 42 or PET amyloid imaging

with radioactive tracers that bind to brillar forms of A b .Both CSFandPETamyloidimagingstudiessuggestthata sub-stantialproportion of clinically normal older individuals dem-onstrate evidence of A b accumulation [27–32] . The exact

proportion of “amyloid-positive” normal individuals isdependent on the age and genetic background of the cohort,but ranges from approximately 20% to 40% and is veryconsonant with large postmortem series [33,34] . Furth-ermore, there is evidence that the AD-P detected at autopsyis related to episodic memory performance even within the“normal” range [35]. Interestingly, the percentage of “amy-loid-positive”normalindividualsat autopsydetectedat a givenage closely parallels the percentage of individuals diagnosedwith AD dementia a decade later [36,37] (Fig. 4). Similarly,genetic at-risk cohorts demonstrate evidence of A b accumula-tion many years before detectable cognitive impairment

[38–41] . These data support the hypothesis that there isa lengthy temporal lag between the appearance of detectableAD-P and the emergence of AD-C.

Multiplegroups have nowreportedthat cognitively normalolder individuals with low CSF A b 1–42 or high PET amyloidbinding demonstrate disruption of functional networks[42–44] and decreased brain volume [45–49] , consistentwith the patterns seen in AD. There have been variablereports in the previously published data thus far, regardingwhether A b -positive individuals demonstrate lowerneuropsychological test scores at the time of biomarkerstudy [50–54] , which may represent heterogeneity in where

these individuals fall on the preclinical continuum, thecognitive measures evaluated, and the degree of cognitive

Fig. 3. Hypothetical model of dynamic biomarkers of the AD expanded to explicate the preclinical phase: A b as identied by cerebrospinal uid A b 42 assay orPET amyloid imaging. Synaptic dysfunction evidenced by uorodeoxyglucose (F18) positron emission tomography (FDG-PET) or functional magneticresonance imaging (fMRI), with a dashed line to indicate that synaptic dysfunction may be detectable in carriers of the 3 4 allele of the apolipoprotein Egene beforedetectable A b deposition. Neuronal injuryis evidenced by cerebrospinaluid tauor phospho-tau,brain structureis evidencedby structural magneticresonance imaging. Biomarkers change from normal to maximally abnormal ( y-axis) as a function of disease stage ( x-axis). The temporal trajectory of two keyindicators used to stage the disease clinically, cognitive and behavioral measures, and clinical function are also illustrated. Figure adapted with permission fromJack et al [22].




reserve in the cohorts. A few early studies have reported that

Ab positivity in clinically normal older individuals isassociated with an increased rate of atrophy [55] and an in-creased risk of cognitive decline and progression to dementia[56–62] . Multiple studies focused on other biomarkers,including volumetric MRI, FDG-PET, or plasma biomarkers,in cohorts of clinically normal older individuals have also re-ported evidence that these markers are predictive of cognitivedecline (refer [63,64] for recent examples). Additionallongitudinal studies are clearly needed to conrm thesendings and to elucidate the combination of factors that bestpredict likelihood and rate of decline, and to betterunderstand individual diff-erences in risk for decline.

As a complement to longitudinal studies in the populationat risk by virtue of age, researchers continue to detect andtrack the biological and cognitive changes associated withthe predisposition to AD in cognitively normal people at dif-ferential genetic risk for AD alone or in conjunction withother risk factors (such as a person’s reported family historyof the disease). To date, the best established genetic risk fac-tors for AD include common allelic variants of APOE ; themajor late-onset AD susceptibility gene; uncommon early-onset AD-causing mutations in the presenilin 1, presenilin2, and APP genes; and trisomy 21 (Down syndrome). Bio-marker studies in presymptomatic carriers of these genetic

risk factors have revealed evidence of A b accumulation on

CSF and PET amyloid imaging, as well as FDG-PET hypo-metabolism, fMRI abnormalities, and brain atrophy that mayprecede symptoms by more than a decade.

7. Cognitive studies

Despite the clear potential of biomarkers for detectingevidence of the AD pathophysiological process, it is impor-tant not to lose sight of the potential that behavioral markershold for early identication. Tests developed by bothneuropsychological and cognitive aging researchers haveprovided evidence that normal aging is accompanied by de-clines in speed of information processing, executive function(working memory, task switching, inhibitory function), andreasoning. Studies that have conducted assessments of cog-nitive function at multiple time points before dementia havealso shown consistently a long period of gradual cognitivedecline in episodic memory as well as nonmemory domainsprogressing up to a decade before onset of dementia. Impor-tantly, in studies that have modeled the curve of cognitivechange versus time, the preclinical trajectory suggests notonly a long- and slow rate of presymptomatic change butalso a period of acceleration of performance decrementthat may begin several years before MCI onset [65]. Recentstudies also suggest that self-report of subtle cognitivedecline, even in the absence of signicant objective impair-ment on testing, may portend future decline in older individ-uals. Despite the existence of multiple studies spanningthousands of participants, the promise of both subjectiveand objective cognitive measures for assessing risk of

progression to AD in individual elders has not yet been fullyrealized. It is likely that measured change in cognition overtime will be more sensitive than any one-time measure.Additional longitudinal studies of older individuals, perhapscombining biomarkers with measures sensitive to detectingvery subtle cognitive decline, are clearly needed.

8. Caveats

Although the aforementioned studies provide compellingevidence that markers of A b in “normal” older individualsare associated with other brain alterations consistent those

seen in AD dementia, and that specic factors may

Fig. 4. Postulated temporal lag of approximately a decade between the de-position of A b (% of individuals with amyloid plaques in a large autopsy se-ries [68]) and the clinical syndrome of AD dementia (estimated prevalencefrom three epidemiological studies [69–71]). Figure courtesy of Mark Mintun and John Morris, Washington University.

Table 1Staging categories for preclinical AD research

Stage Description A b (PET or CSF)Markers of neuronal injury(tau, FDG, sMRI)

Evidence of subtlecognitive change

Stage 1 Asymptomatic cerebralamyloidosis

Positive Negative Negative

Stage 2 Asymptomatic amyloidosis1 “downstream” neurodegeneration

Positive Positive Negative

Stage 3 Amyloidosis 1 neuronal injury1 subtle cognitive/behavioraldecline

Positive Positive Positive

Abbreviations: AD, Alzheimer’s disease; A b , amyloid beta; PET, positron emission tomography; CSF, cerebrospinal uid; FDG, uorodeoxyglucose (18F);sMRI, structural magnetic resonance imaging.




accurately predict those individuals who are at a higher risk of progression to AD-C, it is important to note several poten-tial confounding issues in the majority of these studies. It islikely that many of these studies suffer from cohort biases. Inparticular, the biomarker and cognitive studies likely are notrepresentative of the general older population because theyare typically “samples of convenience,” that is, volunteer co-horts who tend to come from highly educated and socioeco-nomic status backgrounds. These individuals may also beless likely to harbor typical age-related comorbidities thatmay inuence the rate of cognitive decline. Older individ-uals who are willing to participate in such intensive studiesmay also represent the “volunteer gene,” and may be moreactively engaged than the typical aging population. Con-versely, these cohorts may include individuals who self-select for this research because of subjective concerns abouttheir own memory function or positive family history, as re-ected by the high rate of APOE 3 4 carriers in some of thesecohorts.

It is also important to note that although these biomarkershave revolutionized the eld of early AD, these markers aremerely “proxies” for the underlying disease and may not fullyreect the biological processes in the living brain. For exam-ple, both CSF and PET amyloid imaging markers seem to beestimates of the deposition of brillar forms of A b , and maynot provide information about oligomeric forms, which maybe the relevant species for synaptic toxicity. Similarly, ourproxy measurements for synaptic dysfunction, such as fMRIor FDG-PET, are indirect measurements of neural function.Other markers of neurodegeneration such as CSF tau and vol-

umetric MRI are not specic to the AD process. Finally, it isimportant to acknowledge that the relationship between bio-markers and cognition may vary signicantly across age andgenetic cohorts. In particular, the dissociation between thepresence or absence of AD-P and clinical symptomatologyin the oldest-old needs to be better understood.

Finally, it is important to re-emphasize that although A bdeposition and neuritic plaque formation are required for thediagnosis of denite AD, and that current evidence suggeststhat Ab accumulation is an early detectable stage of thepathological-clinical continuum of AD, the role of A b asthe etiologic agent in sporadic late-onset AD remains to be

proven. There may be pathophysiological events that are “up-stream” of A b accumulation yet to be discovered, and the re-lationship between A b and neurodegeneration is not yet clear.In particular, the failure of biologically active A b -loweringtherapies to demonstrate clinical benet thus far is of concern.Thus, it is important to continue research in alternativepathophysiological pathways and therapeutic avenues.

9. Draft operational research framework for stagingpreclinical AD

To facilitate future studies, we propose draft operational

research criteria to dene study cohorts at risk for develop-ing AD dementia for use in (1) longitudinal natural history

studies to determine whether the presence of A b markers,either in isolation or in combination with additional markersof neurodegeneration, is predictive of cognitive decline inclinically normal older individuals, and (2) clinical trialsof potential disease-modifying agents to investigate effectson biomarker progression and/or the emergence of clinicalsymptoms.

We emphasize again that this framework is not intendedto serve as diagnostic criteria for clinical purposes. Use of these biomarkers in the clinical setting is currently unwar-ranted because many individuals who satisfy the proposedresearch criteria may not develop the clinical features of AD in their lifetime. Inappropriate use of this informationin this context could be associated with unwarranted concernbecause there is currently insufcient information to relatepreclinical biomarker evidence of AD to subsequent ratesof clinical progression with any certainty.

These research criteria are based on the postulate that ADis characterized by a sequence of biological events thatbegins far in advance of clinical dementia. On the basis of current evidence from both genetic at-risk and older cohortstudies, we put forth the hypothesis that A b accumulation,or the stage of cerebral amyloidosis, is currently one of theearliest measurable stages of AD, and occurs before anyother evidence of cognitive symptomatology. We postulatethat the presence of biomarker “positivity” for A b in clini-cally normal older individuals, particularly in combinationwith evidence of abnormality on other biomarkers of AD-P, may have implications for the subsequent course of AD-C and the responsiveness to treatments targeting AD-P.

Recognizing that the preclinical stages of AD representa continuum, including individuals who may never progressbeyond the stage of A b accumulation, we further suggest thefollowing staging schema (see Table 1), which may proveuseful in dening research cohorts to test specic hypothe-ses. Research cohorts could be selected on the basis of thesestaging criteria, to optimize the ability to ascertain the spe-cic outcomes important for a given type (e.g., natural his-tory or treatment trial) and duration of the study. Evidenceof “downstream” biomarkers or subtle cognitive symptomsin addition to evidence of A b accumulation may increasethe likelihood of rapid emergence of cognitivesymptomatol-

ogy and clinical decline to MCI within several years. Thepresence of one or more of these additional biomarkerswould indicate that individuals are already experiencingearly neurodegeneration, and as such, it is possible thatamyloid-modifying therapies may be less efcacious afterthe downstream pathological process is set in motion. Thereare specic circumstances, however, such as pharmaceuticalindustry trials that may require a cognitive or clinical end-point, rather than relying solely on biomarker outcomes. Inthese cases, it may be advantageous to enrich the study pop-ulation with individuals in late preclinical stages of AD withevidence of very subtle cognitive change, who would be

most likely to rapidly decline and manifest MCI withina short period (see Fig. 5). We recognize that these stages




will likely require further modication as new ndingsemerge, and that the feasibility of delineating these stagesin recruiting clinical research cohorts remains unclear. Itmay be easiest to recruit individuals on the basis of A b pos-itivity and perform post hoc analyses to determine thepredictive value of specic combinations of biomarker ab-normalities. These proposed research criteria are intendedto facilitate the standardized collection of new data to betterdene the spectrum of preclinical AD, and to elucidate theendophenotype of individuals who are most likely toprogress toward AD-C.

9.1. Stage 1: The stage of asymptomatic cerebralamyloidosis

These individuals have biomarker evidence of A baccumulation with elevated tracer retention on PET amy-loid imaging and/or low A b 42 in CSF assay, but no detect-able evidence of additional brain alterations suggestive of neurodegeneration or subtle cognitive and/or behavioralsymptomatology. The standards for determining “amy-

loid-positivity” are still evolving (refer to the next section).Although recent work suggests there may be a CSF A b 42

cutoff value that is predictive of progression from MCIto AD dementia [66], it is unknown whether a similarthreshold will be optimal in prediction of decline in indi-viduals with normal or near normal cognition. Similarly,using PET imaging techniques, it remains unknownwhether a summary numeric threshold within an aggregatecortical region or within specic anatomic region will pro-vide the most useful predictive value. Recent data suggestthat although CSF A b 42 is strongly inversely correlatedwith quantitative PET amyloid imaging measures (distribu-

tion value ratio or standardized uptake value), there aresome individuals who demonstrate decreased CSF A b 42

and who would not be considered amyloid positive onPET scans [67]. It remains unclear whether this nding re-ects different thresholds used across these techniques or if decreased CSF A b 42 is an earlier marker of accumulation.In addition, there may be genetic effects that are specic toCSF or PET markers of A b .

As mentioned previously, we note that the currently avail-able CSF and PET imaging biomarkers of A b primarily pro-vide evidence of amyloid accumulation and deposition of brillar forms of amyloid. Although limited, current datasuggest that soluble or oligomeric forms of A b are likelyin equilibrium with plaques, which may serve as reservoirs,but it remains unknown whether there is an identiable pre-plaque stage in which only soluble forms of A b are present.Because laboratory data increasingly suggest that oligo-meric forms of amyloid may be critical in the pathologicalcascade, there is ongoing work to develop CSF and plasmaassays for oligomeric forms of A b . There are also emergingdata from genetic-risk cohorts that suggest early synapticchangesmay be present before evidence of amyloid accumu-lation using currently available amyloid markers. Thus, it

may be possible in the future to detect a stage of diseasethat precedes stage 1.

9.2. Stage 2: Amyloid positivity 1 evidence of synapticdysfunction and/or early neurodegeneration

These individuals have evidence of amyloid positivityand presence of one or more markers of “downstream”AD-P-related neuronal injury. The current markers of neuro-nal injury with the greatest validation are: (1) elevated CSFtau or phospho-tau, (2) hypometabolism in an AD-like pat-tern (i.e., posterior cingulate, precuneus, and/or temporopar-

ietal cortices) on FDG-PET, and (3) cortical thinning/graymatter loss in a specic anatomic distribution (i.e., lateral

Fig. 5. Graphic representation of the proposed staging framework for preclinical AD. Note that some individuals will not progress beyond Stage 1 or Stage 2.Individuals in Stage 3 are postulated to be more likely to progress to MCI and AD dementia. Abbreviations: AD, Alzheimer’s disease; Ab, amyloid beta; PET,

position emission tomography; CSF, cerebrospinal uid; FDG, uorodeoxyglucose, fMRI, functional magnetic resonance imaging, sMRI, structural magneticresonance imaging.




and medial parietal, posterior cingulate, and lateral temporalcortices) and/or hippocampal atrophy on volumetric MRI.Future markers may also include fMRI measures of defaultnetwork connectivity. Although previous studies have dem-onstrated that, on average, amyloid-positive individualsdemonstrate signicantly greater abnormalities on thesemarkers as compared with amyloid-negative individuals,there is signicant interindividual variability. We hypothe-size that amyloid-positive individuals with evidence of earlyneurodegeneration may be farther down the trajectory (i.e.,in later stages of preclinical AD). It remains unclear whetherit will be feasible to detect differences among these otherbiomarkers of AD-P, but there is some evidence that earlysynaptic dysfunction, as assessed by functional imagingtechniques such as FDG-PET and fMRI, may be detectablebefore volumetric loss.

9.3. Stage 3: Amyloid positivity 1 evidence of neurodegeneration 1 subtle cognitive decline

We postulate that individuals with biomarker evidence of amyloidaccumulation,earlyneurodegeneration, and evidenceof subtle cognitive decline are in the last stage of preclinicalAD, and are approaching the border zone with the proposedclinical criteria for MCI. These individuals may demonstrateevidence of decline from their own baseline (particularly if proxies of cognitive reserve are taken into consideration),even if they still perform within the “normal” range on stan-dard cognitive measures. There is emerging evidence thatmoresensitive cognitive measures, particularly withchalleng-

ing episodic memory measures, may detectvery subtle cogni-tive impairment in amyloid-positive individuals. It remainsunclear whether self-complaint of memory decline or othersubtle neurobehavioral changes will be a useful predictor of progression, but it is possible that the combination of bio-markersand subjective assessment of subtle change will proveto be useful.

10. Need for additional study

We propose a general framework with biomarker criteriafor the study of the preclinical phase of AD; however, more

work is needed to clarify the optimal CSF assays, PET orMRI analytic techniques, and in particular, the specicthresholds needed to meet these criteria. There are signi-cant challenges in implementing standardized biomarker“cut-off” values across centers, studies, and countries.Work to standardize and validate both uid-based and imag-ing biomarker thresholds is ongoing in multiple academicand pharmaceutical industry laboratories, as well as inseveral multicenter initiatives. These criteria will need tobe validated in large multicenter natural history studies, oras provisional criteria for the planning of preventative clini-cal trials. For instance, it will be important to establish the

test–retest andcross-center reliability of biomarker measure-ments, further characterize the sequence of biomarker

changes, and the extent to which these biomarkers predictsubsequent clinical decline or clinical benet. In particular,there is an important need to evaluate methods for determin-ing “amyloid-positivity” because it remains unclear whetherthere is a biologically relevant continuum of A b accumula-tion, or whether there is a clear threshold or “cut-off” valuethat could be dened on the basis of predictive value forsubsequent clinical decline, as has been suggested in severalCSF studies [28,66] . It also remains unknown whether thesethresholds should be adjusted for age or genotype. Afterthese thresholds are established, it may be most feasible toselect research cohorts for large studies solely on the basisof “amyloid-positivity” on CSF or PET amyloid imaging,and to use additional biomarker and cognitive measures forpost hoc analyses to determine additional predictive value.

Although recent advances in biomarkers have revolution-ized our ability to detect evidence of early AD-P there is stilla need for novel biomarker development. In particular, al-though the current biomarkers provide evidence of A bdeposition, an in vivo marker of oligomeric forms of A bwould be of great value. Imaging markers of intraneuronalpathology, including specic markers of specic forms of tau/tangles and alpha-synuclein, are also needed. In addi-tion, more sensitive imaging biomarkers that can detect earlysynaptic dysfunction and functional and structural discon-nection, such as fMRI and diffusion tensor imaging, mayone day prove to be useful to track early response toamyloid-lowering therapies. Finally, we may be able to usethe currently available biomarkers as a new “gold standard”to re-evaluate simple blood and urine markers that were dis-

carded on the basis of excessive overlap between clinicallynormal and AD patients. The signicant proportion of clin-ically normal individuals who are “amyloid-positive” onboth CSF and PET imaging may have confounded previousstudies attempting to differentiate “normal” controls frompatients with AD.

Similarly, additional work is required to identify andvalidate neuropsychological and neurobehavioral measuresto detect the earliest clinical manifestations of AD. We needto develop sensitive measures in multiple cognitive and be-havioral domains that will reveal evidence of early synapticdysfunction in neural networks vulnerable to AD pathology.

We also need to develop measures of very early functionalchanges in other domains, including social interaction,mood, psychomotor aspects of function, and decisionmaking.These measures would allow us to link better the pathologicalprocesses to the emergence of clinical symptoms, and may beparticularly useful to monitor response to potential disease-modifying therapies in these very early stages.

The proposed criteria apply primarily to individuals at risk by virtue of advanced age because inclusion criteria for trialsin autosomal dominant mutation carriers and homozygous APOE 3 4 carriers will be likely dened primarily on geneticstatus. Trials in genetic-risk populations might use these crite-

ria to stage individuals within the preclinical phase of AD. Ingenetic-risk cohorts, it may even be possible to detect an even




earlier stage of presymptomatic AD, before the point whenthere is already detectable cerebral amyloidosis. SeveralFDG-PET and fMRI studies have suggested that evidenceof synaptic dysfunction may be present in young andmiddle-aged APOE 3 4 carriers (see Fig. 3), and there maybe other biological alterations that are present before signi-cant deposition of brillar forms of amyloid that would bepreferentially responsive to presymptomatic intervention.

The emerging concept of preclinical AD and the role of biomarkers in the detection and tracking in this stage of the disease have important implications for the developmentof effective treatments. Therapies for preclinical AD wouldbe intended to postpone, reduce the risk of, or completelyprevent the clinical stages of the disorder. As recently noted,the use of clinical endpoints in clinical trials of such treat-ments would require large numbers of healthy volunteers,large amounts of money, and many years of study.Researchers have raised the possibility of evaluating bio-marker endpoints for these treatments in cognitively normalpeople at increased risk for AD because these studies mightbe performed more rapidly than otherwise possible. Subjectsenrolled in these studies could include individuals with auto-somal dominant mutation carriers (with essentially a 100%chance of developing clinical AD) or those at increasedrisk of developing sporadic AD (e.g., APOE 3 4 carriers orsubjects with biomarker evidence of preclinical AD pathol-ogy). The use of biomarkers rather than clinical outcomescould accelerate progress in these trials; however, regulatoryagencies must be assured that a given biomarker is “reason-ably likely” to predict a clinically meaningful outcome be-

fore they would grant approval for treatments tested intrials using biomarkers as surrogate endpoints. Researchstrategies have been proposed to provide this evidence byembedding the most promising biomarkers in preclinicalAD trials of people at the highest imminent risk of clinicalonset to establish a link between a biomarker effect andthe onset of clinical symptoms of AD. We envision thetime when the scientic means and accelerated regulatoryapproval pathway support multiple preclinical AD trials us-ing biomarkers to identify subjects and provide shorter termoutcomes, such that demonstrably effective treatments toward off the clinical stages of AD are found as quickly as

possible. There are several burgeoning efforts to designand conduct clinical trials in both genetic at-risk andamyloid-positive older individuals, including the Domi-nantly Inherited Alzheimer Network (study of familialAD), the Alzheimer Prevention Initiative, and Anti-Amyloid Treatment in Asymptomatic AD (A4) trial beingconsidered by the Alzheimer’s Disease Cooperative Study.

Finally, the ethical and practical implications surroundingthe issues of future implementation of making a “diagnosis”of AD at a preclinical stage need to be studied, should thepostulates put forth previously prove to be correct. Althoughat this point our recommendations are strictly for research

purposes only , the public controversy surrounding the iden-tication of asymptomatic individuals with evidence of AD-

P raised several important points that the eld must consider.In particular, the poignant question of “why would anyonewant to know they have AD a decade before they might de-velop symptoms, if there is nothing they can do about it?”should be carefully considered well before any resultsfrom research is translated into clinical practice. First, theremay be important reasons, including social and nancialplanning, why some individuals would want to know theirlikelihood of developing AD dementia within the nextdecade, even in the absence of an available disease-modifying therapy. It is our hope, however, that the advancesin preclinical detection of AD-P will enable earlier, moreeffective treatment, just as nearly all of therapeutic gainsin cancer, cardiovascular disease, osteoporosis, and diabetesinvolve treatment before signicant clinical symptoms arepresent. It is entirely possible that promising drugs, particu-larly amyloid-modifying agents, will fail to affect the clini-cal course of AD at the stage of dementia or even MCI, whenthe neurodegenerative process is well entrenched, but maybe efcacious at the earliest stages of the AD-P, before theonset of symptoms.

The denitive studies to determine whether the majorityof asymptomatic individuals with evidence of AD-P areindeed destined to develop AD dementia, to elucidate thebiomarker and/or cognitive endophenotype that is mostpredictive of cognitive decline, and to determine whetherintervention with potential disease-modifying therapies inthe preclinical stages of AD will prevent dementia arelikely to take more than a decade to fully accomplish.Thus, we must move quickly to test the postulates put forth

previously, and adjust our models and study designs as newdata become available. Because potential biologically ac-tive treatments may be associated with small but signicantrisk of adverse side effects, we will need to determinewhether we can predict the emergence of cognitive symp-toms with sufcient certainty to appropriately weigh therisk/benet ratios to begin treatment in asymptomatic indi-viduals. It is clear that many questions remain to beanswered, and that there may be additional factors whichwill inuence the probability of developing clinicalAD. However, the considerable progress made over thepast two decades now enables a strategic path forward to

test these hypotheses, move the eld toward earlier inter-vention, and ultimately, toward the prevention of ADdementia.

Acknowledgments

The chair (Reisa Sperling) acknowledges the invaluableassistance of Dr. Cerise Elliott at National Institute on Ag-ing, as well as thoughtful input solicited from several indi-viduals, in particular, Drs. Keith Johnson, Dorene Rentz,Peter Davies, Deborah Blacker, Steve Salloway, Sanjay As-

thana, and Dennis Selkoe, as well as the helpful public com-mentary provided by our colleagues in the eld.




Reisa Sperling has servedas a site investigator and/orcon-sultant to several companies developing imaging biomarkersand pharmacological treatments for early AD, includingAvid, Bayer, Bristol-Myers-Squibb, Elan, Eisai, Janssen,Pzer, and Wyeth. Paul Aisen serves on a scientic advisoryboard for NeuroPhage; serves as a consultant to Elan Corpo-ration, Wyeth, Eisai Inc., Bristol-Myers Squibb, Eli Lilly andCompany, NeuroPhage, Merck & Co., Roche, Amgen, Ab-bott, Pzer Inc., Novartis, Bayer, Astellas, Dainippon, Bio-marin, Solvay, Otsuka, Daiichi, AstraZeneca, Janssen, andMedivation Inc.; receives research support from Pzer Inc.and Baxter International Inc.; and has received stock optionsfrom Medivation Inc. and NeuroPhage. Clifford Jack servesas a consultant for Eli Lilly, Eisai, and

Elan; is an investigatorin clinical trials sponsored by Baxter and Pzer Inc.; andowns stock in Johnson and Johnson. Denise Park has re-ceived research support from Avid Pharmaceuticals. EricSiemers is an employee of Eli Lilly and Company, which ac-quired Avid Pharmaceuticals. Yaakov Stern has consulted toBayer Pharmaceuticals and has received research supportfrom Bayer, Janssen, Eli Lilly, and Elan; Maria Carrillo isan employee of the Alzheimer’s Association and reports noconicts; Bill Thies is an employee of the Alzheimer’s Asso-ciation and reports no conicts; Creighton Phelps is an em-ployee of the U.S. Government and reports no conicts.

References

[1] Morris JC. Early-stage and preclinical Alzheimer disease. AlzheimerDis Assoc Disord 2005;19:163–5.

[2] Dubois B, Feldman HH, Jacova C, Cummings JL, Dekosky ST, Bar-berger-Gateau P, et al. Revising the denition of Alzheimer’s disease:a new lexicon. Lancet Neurol 2010;9:1118–27.

[3] Jack CR Jr, AlbertM, Knopman DS,McKhann GM,Sperling RA,Car-rillo M, et al. Introduction to revised criteria for the diagnosis of Alz-heimer’s disease: National Institute on Aging and the Alzheimer’sAssociation workgroup. Alzheimers Dement 2011;7:in press.

[4] AlbertMS, DeKosky ST, DicksonD,DuboisB,FeldmanHH, FoxNC,etal. The diagnosis of mild cognitive impairment due to Alzheimer’sdisease: Recommendations fromtheNational Instituteon Agingand Alz-heimer’s Association workgroup. Alzheimers Dement 2011;7:in press.

[5] McKhann GM,Knopman DS,Chertkow H, Hyman BT, Jack Jr CR, Ka-was CH, et al. The diagnosis of dementia due to Alzheimer’s disease:Recommendations from the National Institute on Aging and the Alz-

heimer’s Association workgroup. Alzheimers Dement 2011;7:in press.[6] Pimplikar SW, Nixon RA, Robakis NK, Shen J, Tsai LH. Amyloid-independent mechanisms in Alzheimer’s disease pathogenesis.J Neurosci 2010;30:14946–54.

[7] Herrup K. Reimagining Alzheimer’s disease–anage-based hypothesis.J Neurosci 2010;30:16755–62.

[8] Mawuenyega KG, Sigurdson W, Ovod V, Munsell L, Kasten T,Morris JC, et al. Decreased Clearance of CNS {beta}-Amyloid in Alz-heimer’s Disease. Science 2010;330:1774.

[9] Selkoe DJ. Alzheimer’s disease is a synaptic failure. Science 2002;298:789–91.

[10] Lee HG, Casadesus G, Zhu X, Takeda A, Perry G, Smith MA.Challenging the amyloid cascade hypothesis: senile plaques andamyloid-beta as protective adaptations to Alzheimer disease. Ann NY Acad Sci 2004;1019:1–4.

[11] Shankar GM, Li S, Mehta TH, Garcia-Munoz A, Shepardson NE,Smith I, et al. Amyloid-beta protein dimers isolated directly from Alz-

heimer’s brains impair synaptic plasticityand memory.Nat Med 2008;14:837–42.

[12] Yaffe K, Fiocco AJ, Lindquist K, Vittinghoff E, Simonsick EM,Newman AB, et al. Predictors of maintaining cognitive functionin older adults: the Health ABC study. Neurology 2009;72:2029–35.

[13] Arvanitakis Z, Wilson RS, Bienias JL, Evans DA, Bennett DA. Diabe-

tes mellitus and risk of Alzheimer disease and decline in cognitivefunction. Arch Neurol 2004;61:661–6.

[14] Craft S. The role of metabolic disorders in Alzheimer disease andvascular dementia: two roads converged. Arch Neurol 2009;66:300–5.

[15] Ganguli M, Du Y, Dodge HH, Ratcliff GG, Chang CC. Depressivesymptoms and cognitive decline in late life: a prospective epidemio-logical study. Arch Gen Psychiatry 2006;63:153–60.

[16] Wilson RS, Arnold SE, Schneider JA, Kelly JF, Tang Y, Bennett DA.Chronic psychological distress and risk of Alzheimer’s disease in oldage. Neuroepidemiology 2006;27:143–53.

[17] Onyike CU, Sheppard JM, Tschanz JT, Norton MC, Green RC,Steinberg M, et al. Epidemiology of apathy in older adults: the CacheCounty Study. Am J Geriatr Psychiatry 2007;15:365–75.

[18] Wilson RS, Scherr PA, Schneider JA, Tang Y, Bennett DA. Relation of cognitive activity to risk of developing Alzheimer disease. Neurology2007;69:1911–20.

[19] Stern Y. Cognitive reserve. Neuropsychologia 2009;47:2015–28.[20] Fotenos AF, Mintun MA, Snyder AZ, Morris JC, Buckner RL. Brain

volume decline in aging: evidence for a relation between socioeco-nomic status, preclinical Alzheimer disease, and reserve. Arch Neurol2008;65:113–20.

[21] Wilson RS, Barnes LL, Aggarwal NT, Boyle PA, Hebert LE, Mendesde Leon CF, et al. Cognitive activity and the cognitive morbidity of Alzheimer disease. Neurology 2010;75:990–6.

[22] Jack CR Jr, Knopman DS,Jagust WJ,ShawLM, Aisen PS,WeinerMW,et al. Hypothetical model of dynamic biomarkers of the Alzheimer’spathological cascade. Lancet Neurol 2010;9:119–28.

[23] Reiman EM, Chen K, Alexander GE, Caselli RJ,Bandy D, Osborne D,et al. Functional brain abnormalities in young adults at genetic risk for

late-onset Alzheimer’s dementia. Proc Natl Acad Sci U S A 2004;101:284–9.

[24] Filippini N, MacIntosh BJ, Hough MG, Goodwin GM, Frisoni GB,Smith SM, et al. Distinct patterns of brain activity in young carriersof the APOE-epsilon4 allele. Proc Natl Acad Sci U S A 2003;106:7209–14.

[25] Sheline YI, Morris JC, Snyder AZ, Price JL, Yan Z, D’Angelo G, et al.APOE4 allele disrupts resting state fMRI connectivity in the absenceof amyloid plaques or decreased CSF Abeta42. J Neurosci 2010;30:17035–40.

[26] Vemuri P, Wiste HJ, Weigand SD, Knopman DS, Trojanowski JQ,Shaw LM, et al. Serial MRI and CSF biomarkers in normal aging,MCI, and AD. Neurology 2010;75:143–51.

[27] Rowe CC, Ellis KA, Rimajova M, Bourgeat P, Pike KE, Jones G, et al.

Amyloid imaging results from the Australian Imaging, Biomarkersand Lifestyle (AIBL) study of aging. Neurobiol Aging 2010;31:1275–83.

[28] Mintun MA, Larossa GN, Sheline YI, Dence CS, Lee SY, Mach RH,et al. [11C]PIB in a nondemented population: potential antecedentmarker of Alzheimer disease. Neurology 2006;67:446–52.

[29] Jack CR Jr, Lowe VJ, Senjem ML, Weigand SD, Kemp BJ,Shiung MM, et al. 11C PiB and structural MRI provide complemen-tary information in imaging of Alzheimer’s disease and amnesticmild cognitive impairment. Brain 2008;131(Pt 3):665–80.

[30] Gomperts SN, Rentz DM, Moran E, Becker JA, Locascio JJ,Klunk WE, et al. Imaging amyloid deposition in Lewy body diseases.Neurology 2008;71:903–10.

[31] De Meyer G, Shapiro F, Vanderstichele H, Vanmechelen E,Engelborghs S, De Deyn PP, et al. Diagnosis-independent Alzheimerdisease biomarker signature in cognitively normal elderly people.Arch Neurol 2010;67:949–56.




Documents

Prclinical Stages Nics Adra 2011