Received: 17 April 2020 Accepted: 4 June 2020 Published online: 13 July 2020 DOI: 10.1002/dad2.12062 PERSPECTIVE The AT(N) framework for Alzheimer’s disease in adults with Down syndrome Michael S. Rafii1 Beau M. Ances2 Nicole Schupf3,4,5,6 Sharon J. Krinsky-McHale7 Mark Mapstone8 Wayne Silverman9 Ira Lott9 William Klunk10 Elizabeth Head11 Brad Christian12 Florence Lai13 H. Diana Rosas14 Shahid Zaman15,16 Melissa E. Petersen17 Andre Strydom18 Juan Fortea19 Benjamin Handen10 Sid O’Bryant20 1 Alzheimer’s Therapeutic Research Institute (ATRI), Keck School of Medicine, University of Southern California, San Diego, California, USA 2 Center for Advanced Medicine Neuroscience, Washington University School of Medicine in St. Louis, St. Louis, Missouri, USA 3 Taub Institute for Research on Alzheimer’s Disease and the Aging Brain/G.H. Sergievsky Center, Columbia University Irving Medical Center, New York, New York, USA 4 Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, New York, USA 5 Department of Neurology, Neurological Institute of New York, Columbia University Irving Medical Center, New York, New York, USA 6 Department of Psychiatry, Columbia University Medical Center, New York, New York, USA 7 Department of Psychology, NYS Institute for Basic Research in Developmental Disabilities, Staten Island, New York, USA 8 Department of Neurology, University of California, Irvine, California, USA 9 Department of Pediatrics, School of Medicine, University of California, Irvine, California, USA 10 Department of Psychiatry, University of Pittsburgh, Pittsburgh, Pennsylvania, USA 11 Department of Pathology, Gillespie Neuroscience Research Facility, University of California, Irvine, California, USA 12 Department of Medical Physics and Psychiatry, University of Wisconsin Madison, Madison, Wisconsin, USA 13 Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Charlestown, Massachusetts, USA 14 Departments of Neurology and Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, Massachusetts, USA 15 Department of Psychiatry, School of Clinical Medicine, University of Cambridge, Cambridge, UK 16 Cambridgeshire and Peterborough NHS Foundation Trust, Fulbourn Hospital, Cambridge, UK 17 Department of Family Medicine and Institute for Translational Research, University of North Texas Health Science Center, Fort Worth, Texas, USA 18 Department of Forensic and Neurodevelopmental Sciences, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, UK 19 Sant Pau Memory Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Biomedical Research Institute Sant Pau, Universitat Autònoma de Barcelona, Barcelona, Spain 20 Institute for Translational Research and Department of Pharmacology and Neuroscience, University of North Texas Health Science Center, Fort Worth, Texas, USA Correspondence Michael S. Rafii, Alzheimer’s Therapeutic Abstract Research Institute (ATRI), Keck School of Medicine, University of Southern California, The National Institute on Aging in conjunction with the Alzheimer’s Association (NIA- 9860 Mesa Rim Road, San Diego, CA 92121 AA) recently proposed a biological framework for defining the Alzheimer’s disease E-mail: mrafii@usc.edu (AD) continuum. This new framework is based upon the key AD biomarkers (amy- loid, tau, neurodegeneration, AT[N]) instead of clinical symptoms and represents the latest understanding that the pathological processes underlying AD begin decades before the manifestation of symptoms. By using these same biomarkers, individuals with Down syndrome (DS), who are genetically predisposed to developing AD, can also Alzheimer’s Dement. 2020;12:e12062. wileyonlinelibrary.com/journal/dad2 © 2020 the Alzheimer’s Association 1 of 10 https://doi.org/10.1002/dad2.12062 2 of 10 RAFII ET AL . be placed more precisely along the AD continuum. The A/T(N) framework is there- fore thought to provide an objective manner by which to select and enrich samples for clinical trials. This new framework is highly flexible and allows the addition of newly confirmed AD biomarkers into the existing AT(N) groups. As biomarkers for other pathological processes are validated, they can also be added to the AT(N) classification scheme, which will allow for better characterization and staging of AD in DS. These biological classifications can then be merged with clinical staging for an examination of factors that impact the biological and clinical progression of the disease. Here, we leverage previously published guidelines for the AT(N) framework to generate such a plan for AD among adults with DS. KEYWORDS Alzheimer’s disease, biomarkers, Down syndrome 1 INTRODUCTION In 2013, the Food and Drug Administration (FDA) released draft guidance on drug development for AD. The guidance built on the under- A core purpose for the generation of the amyloid, tau, neurodegener- standing that AD is a progressive disease with clinical symptoms of ation (AT[N]) framework was to “enable a more precise approach to dementia appearing decades after the AD pathophysiological process interventional trials where specific pathways can be targeted in the dis- has begun and proposed a disease classification that acknowledged ease progress and in the appropriate people.”1 Recently, the National three stages of AD: the preclinical, prodromal, and dementia stages.8 Institute on Aging and Alzheimer’s Association (NIA-AA) proposed a In 2018, the FDA revised the draft guidance and expanded the taxon- “research framework” based on the AT(N) model2 for observational omy of AD by recognizing four stages.9 These include: Stage 1: “Pre- and interventional research on Alzheimer’s disease (AD).3 Differently clinical AD”; Stage 2: “Preclinical/ Prodromal AD”; Stage 3: “Prodromal from the prior NIA-AA diagnostic criteria,4 this framework defines AD AD”; and Stage 4: “AD dementia.” We are now poised to study these as a biological rather than a clinical construct, characterized by extra- stages of AD in Down syndrome (DSAD) using the most advanced AD cellular deposits of amyloid-beta peptide (Aβ; “A”), intraneuronal aggre- biomarkers available to refine the AT(N) classification for use in this gates of hyperphosphorylated tau (“T”) and neurodegeneration (“[N]”). population. The “N” is placed in parentheses to emphasize that the biomarkers in the (N) group are fundamentally different from “A” and “T” biomark- ers because they are: (1) not specific for neurodegeneration due to AD, 2 APPLICATION OF AT(N) TO THE DS (2) may be attributed to other possible comorbid conditions, and (3) POPULATION do not map onto neuropathologic findings used to diagnose AD. The AT(N) model considers A, T, and (N) status relatively independent from To date, the AT(N) framework has been applied in limited ways to one another with a known sequential order. However, the model then other populations that are at risk for AD as a method to expand this combines the clinically defined diagnostic classifications with AT(N) model and to enrich clinical trials for AD.10 Given the unique features biomarker status for consistent terminology for research use. described below of AD among adults with DS, the framework may have Briefly, negative amyloid and tau along with the absence of neu- utility for rapid advancement of precision medicine approaches to rodegeneration (A−T−[N]–) defines the normal biomarker profile, novel clinical trials in this population. DS is, by definition, a genetically and amyloid negativity with either positivity for tau or presence of determined form of AD as recognized in the International Work Group neurodegeneration corresponds to suspected non-Alzheimer’s pathol- on Criteria for AD (IWG-2).11 In DS, one of the most common forms ogy. β-Amyloidosis (A+) is sufficient to identify the Alzheimer’s of intellectual disability, the underlying genetic link between trisomy continuum. Within this continuum, A+T−(N)– denotes Alzheimer’s 21 and AD has been convincingly established.12-15 By age 40 years, pathologic change (preclinical AD), while A+T+ (with or without [N]+) all adults with DS exhibit some degree of elevated brain amyloid.16-18 establishes definite AD. The AT(N) framework has the potential to The leading explanation for this link is tied to the triplication of enrich clinical trials with individuals who show objective evidence that chromosome 21 (trisomy 21) and the resulting overexpression of the they are on the AD continuum while also permitting the staging of amyloid precursor protein (APP) gene coded on this chromosome.19 individual patients and providing prognosis as well as stratification for The excessive production of Aβ as a result is key to the pathogen- precision-based clinical trials. The AT(N) classification system has been esis of AD in adults with DS.20 Although other genes coded on studied by multiple groups and has demonstrated utility in classifying chromosome 21 may contribute to the early emergence of dementia individuals with late-onset sporadic AD on the basis of biomarkers.5-7 and the phenomenon of accelerated aging seen in adults with DS,21 RAFII ET AL . 3 of 10 forms of partial trisomy 21 which do not result in triplication of APP (ie, the APP-containing portion of chromosome 21 is not present in the RESEARCH IN CONTEXT third copy) are not associated with clinical and pathological signs of 1. Systematic review: Alzheimer’s disease (AD) is highly AD.14,15 prevalent in Down syndrome (DS). The National Institute Despite these consistent AD neuropathologic changes, the timing of on Aging in conjunction with the Alzheimer’s Associa- the development of dementia as part of AD in DS is quite variable,22 tion (NIA-AA) recently proposed a biological framework suggesting the presence of other genetic and environmental risk and for defining the AD continuum. This new framework is protective factors. Individuals with DS have a lifetime risk for demen- based on key AD biomarkers (amyloid, tau, neurodegen- tia in excess of 90%, and DS is now acknowledged to be a genetic form eration, AT[N]) instead of clinical symptoms and repre- of AD similar to the much less common autosomal-dominant causes of sents the latest understanding that the pathological pro- AD.23,24 Although the development of dementia is not inevitable in all cesses underlying AD begin decades before the manifes- adults with DS, the risk increases incrementally with age.25 Further- tation of symptoms. more, as in the late-onset form in the general population, the AT(N) 2. Interpretation: These biological classifications can then classification of adults with DS will be strongly influenced by the age be merged with clinical staging for an examination of fac- of the individual. tors that impact the biological and clinical progression of Identifying cognitive impairment at an early stage of the AD con- the disease. We leverage previously published guidelines tinuum has become an increasingly important goal in AD research, as for the AT(N) framework to generate such a plan for AD it is widely believed that the greatest chance for therapeutic success among adults with DS. will be obtained by intervening early in the disease, before widespread 3. Future directions: Further work on longitudinal AD and irreversible neurodegeneration has occurred.26 As a result, the biomarkers in DS should help clarify whether the AT(N) AT(N) framework describes AD across its full spectrum (ie, preclinical classification system can be applied to individuals with DS to dementia) in terms of biomarker positivity/negativity and is agnos- both for clinical trial stratification as well as for use as a tic with respect to clinical symptoms. As more longitudinal data are potential staging and prognostic tool in the clinic, repre- collected in DS, correlations between the distinct AT(N) classifications senting a fundamental tool for precision medicine. with clinical and cognitive status will be possible as well as a richer understanding of the rates of change in each biomarker category: amy- loid, tau, and neurodegeneration across the AD conitnuum in DS. This more precise assessment will facilitate primary, secondary, and tertiary prevention trials for ADin individuals with DS.27-29 3 CLINICAL ASSESSMENT OF COGNITIVE In the general population30-32 as well as in DS33-36 the construct STATUS of mild cognitive impairment (ie, prodromal AD) as well as the iden- tification of disease in the preclinical stage (eg, accumulation of The defining feature of all causes of dementia is a decline from the amyloid in a cognitively stable individual) is central to the clinical diag- baseline level of function and performance of daily skills. Although this nostic formulation of AD. Mild cognitive impairment (MCI) in the gen- may be straightforward to establish in the general population, it can eral population, as well as in DS, is generally regarded as the border- be a much more complicated task in adults with DS because of life- land between the cognitive changes of aging and early dementia where long intellectual impairment and significant variance in baseline cog- there is measurable decline in memory as well as some decline on nitive functioning.45,46 This is especially true for older adults with DS instrumental activities of daily living (iADLs) but preservation of basic due to various factors impacting living arrangements in which there activities of daily functioning.22,37-40 The characterization of preclini- may be poor record keeping since childhood, lack of continuity in staff cal and prodromal AD is now possible with the advancement of state- members supervising adults with DS over time, and a large number of-the-art biomarker modalities such as amyloid and tau assessment of physicians/health-care providers throughout his or her life span. using positron emission tomography (PET) imaging and cerebrospinal In the absence of a personal historian who can accurately and com- fluid (CSF) measures as well as emerging plasma biomarkers such as prehensively attest to an individual’s baseline level of functioning, the mass spectral Aβ assays41,42 in the absence of or minimal cognitive assessment of a reported cognitive and behavioral change may be decline. exponentially more difficult.47,48 The early signs of dementia in adults Here we propose an application of the AT(N) framework for the full with DS can be subtle and often require an astute observer to iden- characterization of the AD continuum in DS using both state-of-the-art tify these changes. Often, individuals with DS are served by numerous biomarkers and clinical assessments. Given the variability in cognitive caregivers throughout their lifetime, and often newly involved care- assessments, the inclusion of biomarkers may facilitate the evaluation givers will presume that the current level of observed ability repre- of potential efficacy of therapy in this population. Forthcoming data sents the individual’s baseline level of functioning and, thus, miss signs from the Alzheimer’s Biomarker Consortium–Down Syndrome (ABC- of early decline that has already occurred. DS)43 and the European Horizon21 consortium44 will inform the diag- In the clinical setting, accurate assessment of cognition and function nostic accuracy and prognostic potential of AT(N) in DS. depend upon a comprehensive history, which for individuals with DS 4 of 10 RAFII ET AL . must be done in the context of knowledge and quantification of their MRI, as well as biofluid markers in CSF and plasma.59 There exist historic level of intellectual disability.49 It is important that a thorough remarkable similarities between AD biomarkers in DS and other popu- history be obtained to compile evidence consistent with an emerging lations with AD.60-64 Greater hippocampal atrophy is associated with a cognitive impairment while probing for potential factors contributing greater amyloid load.61 Cognitive and functional measures do not cor- to decline. Pertinent historical information is useful from personal relate as strongly with amyloid deposition as they do with abnormali- accounts of caregivers and family members who have known the ties on 18F-fluourodeoxyglucose (FDG) and tau PET.61,65,66 individual for an extended length of time.50 In addition, other sources of information, such as previous neuropsychological testing or school Individual Education Plan information, can greatly assist in accurately 4.1 Amyloid (A) characterizing an individual’s baseline level of functioning. In addition, medical history, medications, family history, social history, review of Amyloid PET positivity as observed using PET imaging in DS seems systems, laboratory evaluations, and brain imaging will be essential to to resemble autosomal dominant AD more closely than sporadic AD. rule out comorbidities that can masquerade as AD-related cognitive Specifically, Pittsburgh compound B (PIB) demonstrates an early and impairment. Objective evidence of memory decline will be essential predominant basal ganglia signal60,62,67,68 for the diagnosis of MCI-DS and dementia. A number of cognitive although other tracers (eg, florbetabir) have shown a pattern more assessment instruments are currently being evaluated in natural similar to sporadic AD.61 The similarity of AD in DS and ADAD is history studies of AD in DS, including the ABC–DS.43 At this time, thought to result from overproduction of Aβ. APP overproduction in DS there is no single cognitive instrument that has been longitudinally leads to baseline plasma levels of Aβ40 and Aβ42 and Aβ42/Aβ40 ratios validated in the context of AD biomarkers in DS but many are being which are higher than those in non-DS individuals.64,69,70 A positive presently intensely researched.51-55 correlation of tau and a negative correlation of CSF Aβ1-42 have been Once arriving at the suspected clinical diagnosis of MCI or dementia, reported with age71 and several studies have documented correlations the AT(N) framework can be used to stage an individual with DS along of the changes in amyloid in DS with AD.72-75 Higher levels of Aβ42 the AD continuum with respect to extent of underlying biomarker or the Aβ42/Aβ40 ratio appear to be associated with the onset of AD changes (Table 1). This staging can be used to provide expected clinical in DS,76,77 although this is not entirely consistent in the literature.78 prognosis, including an estimated duration of independent functioning, CSF Aβ42 levels are first increased in early life and then become lower time to dementia, and to also enrich for more homogenous samples in with age, representing deposition of Aβ into plaques.64,79,80 Most stud- clinical trials. The proposed clinical staging of the cognitive continuum ies seem to suggest that as with sporadic and autosomal-dominant AD, was adapted from previously published guidelines for preclinical AD,56 pathophysiological changes associated with AD in DS occur approxi- MCI,57 and AD dementia.58 mately two decades before the onset of symptoms of dementia. The difficulties with MCI diagnoses in the general population are Blood-based biomarkers have clear advantages as biomarkers as well established. MCI in adults with DS (MCI-DS) is an even more they are easily accessible. Individuals with DS have higher baseline challenging diagnosis and cross-sectional assessments can be unreli- plasma Aβ1-42 and Aβ1-40 concentrations compared to individuals able. Therefore, longitudinal assessments are optimal and required. without DS81 due to the extra copy of the APP gene and the result- Additional work is needed to determine optimal psychometric assess- ing overproduction of APP and Aβ. There have been a limited num- ment instruments, cutoff scores, and/or combinations of instruments ber of CSF studies in individuals with DS which show elevated levels in this population for refinement of the MCI designation. Specifically, of Aβ42 early in life, but with age, CSF Aβ42 levels decline (as expected the following points will need to be considered as the concept of MCI- with their deposition into plaques) while CSF tau levels progressively DS evolves and will be informed by forthcoming data from ABC–DS: increase.64,81 (1) Identification of the most informative cognitive assessment instru- ments for MCI varying based on severity of ID. (2) Quantification of decline needed to represent a clinically meaningful change. (3) Rela- 4.2 Tau (T) tionship between cognitive assessments and rates of change in various AD biomarkers. In order to confirm that MCI-DS is in fact prodromal Neurofibrillary tangles (NFTs)which are comprised of abnormal tau, AD, the use of biomarkers to confirm AD as the underlying etiology will are a key pathological hallmark of AD and correlate with the emer- be required. gence of clinical symptoms more closely than amyloid plaques. This relationship has also been demonstrated in post mortem pathology of DS brains, where NFTs correlate with cognitive decline.82 Tau PET sig- 4 BIOMARKER ASSESSMENTS OF AMYLOID, nal in the DS brain appears to be similar to sporadic AD and can be TAU, AND NEURODEGENERATION IN DS assessed using standard Braak staging.66 Specifically, tau deposition in adults with DS has been studied using the PET tracer (18 F) AV-1451.66 Over the past few years, substantial progress has been made in elu- Abnormal tau distribution (in the form of NFTs) first involves the medial cidating the natural history of AD in people with DS using the latest temporal cortices and then spreads posteriorly,66 similar in manner to biomarkers including amyloid and tau PET imaging, volumetric brain that observed in sporadic AD. More recently, plasma and CSF tau have RAFII ET AL . 5 of 10 TA B L E 1 Clinical staging of cognitive continuum—Diagnostic recommendations Cognitively Stable (Preclinical Alzheimer’s Disease) 1. No report of cognitive decline that is greater than what would be expected with aging, based on informant or clinician report—subjective decline taken into account, but not a requirement 2. No objective evidence of cognitive decline that is greater than what would be expected with aging per se on formal neuropsychological testing (brief cognitive screening instruments are insufficient with this population) a. If first assessment, impairment is defined compared to estimated premorbid level of functioning (eg, functional measure, IQ measure) (i) Confidence of diagnosis is less b. If prior testing is available, impairment is defined based on decline from prior testing levels (i) Confidence of diagnosis is high 3. Preservation of premorbid level of functional abilities based on reliable informant report, unless functional decline is related to age-associated frailty unrelated to AD (ie, muscle weakness, etc.). Mild Cognitive Impairment–Down Syndrome (Prodromal AD) 1. Report of decline in cognitive functioning as reported by reliable informant or clinician AND 2. Objective evidence of impairment in one or more cognitive domains, based on extensive neurocognitive testing (brief cognitive screening instruments are insufficient with this population) a. Cognitive impairment may be in any of the following domains: (i) Impaired ability to acquire and remember new information (ii) Impaired reasoning and handling of complex tasks from premorbid level (iii) Impaired visuospatial abilities (iv) Impaired language functions (v) Changes in personality, behavior, or other neuropsychiatric symptoms that may include uncharacteristic fluctuations in mood (eg, agitation), depression, changes in motivation, apathy, social withdrawal, loss of interest in previous activities, among others b. If first assessment, impairment is defined as compared to estimated premorbid level of functioning (eg, functional measure, IQ measure) (i) Confidence of diagnosis is lower c. If prior cognitive testing is available, impairment is defined based on decline from prior testing levels 3. Confidence of diagnosis is high 4. Cognitive changes are not better explained by other factors such as significant life event (eg, environmental change, medical illness, etc.) 5. Preservation of premorbid level of basic functional abilities (basic ADLs) based on reliable informant report. There may be declines in iADLs in the DS population Dementia in Down Syndrome 1. Cognitive concern reflecting a change as reported by reliable informant or clinician report–subjective decline taken into account if present, but not a requirement 2. Objective evidence of impairment in two or more cognitive domains, based on formal neurocognitive testing (brief cognitive screening instruments are insufficient with this population) a. Cognitive impairment may be in any of the following domains: (i) Impaired ability to acquire and remember new information (ii) Impaired reasoning and handling of complex tasks from premorbid level (iii) Impaired visuospatial abilities (iv) Impaired language functions (v) Changes in personality, behavior, or other neuropsychiatric symptoms that may include: uncharacteristic fluctuations in mood (eg, agitation), changes in motivation, apathy, social withdrawal, loss of interest in previous activities, among others b. If first assessment, impairment is defined as compared to estimated premorbid level of functioning (i) Confidence of diagnosis is less c. If prior testing is available, impairment is defined based on decline from prior testing levels (i) Confidence of diagnosis is high 3. Cognitive changes are not better explained by other factors such as significant life event (eg, environmental change), or active medical or mental illness, etc. 4. Changes in cognition and/or neuropsychiatric/behavioral symptoms interfere with previous level of daily functioning (basic ADLs) based on informant and/or clinician report – subjective decline taken into account but not a requirement. Abbreviations: AD, Alzheimer’s disease; ADL, activities of daily living; DS, Down syndrome; iADL, instrumental activities of daily living. been studied in individuals with DS, with their levels correlating with autosomal dominant forms of AD. More recently, plasma neurofil- AD dementia in DS.64,78,83-85 ament light chain (NfL) levels (also a marker of neurodegeneration) have been shown to correlate with clinical status of AD in DS92 as well as standard AD biomarkers such as amyloid PET and tau PET.93 4.3 Neurodegeneration (N) Specifically, plasma NfL levels appear to increase with age in but can still distinguish between normal aging and AD81,92 . Plasma NfL levels Markers of AD-specific neurodegeneration include regional have also been shown to correlate with other markers of neurode- hypometabolism on FDG PET986-88 or hippocampal atrophy89-91 generation such as hypometabolism on FDG PET and hippocampal have been studied in DS and parallel findings from the sporadic and atrophy, as well as cognitive and functional decline.93 6 of 10 RAFII ET AL . TA B L E 2 Biomarker classification AT(N) pathology among adults elevated tau or any neurodegeneration (A+/T−/N−) would be cate- with DS gorized as preclinical AD. An individual who has symptoms consistent **Biomarker classification is independent of Consensus Clinical with MCI-DS who has elevated brain amyloid but no elevated tau or Staging of Cognitive Continuum evidence of neurodegeneration (A+/T+/N) would be classified as Pro- AT(N) Biomarker Grouping dromal AD. Finally, an individual with MCI-DS but who is A−/T+/N− A: Aggregated Aβ or associated pathophysiologic state would be considered as having a neurodegenerative disease other CSF Aβ42 , or Aβ42 /Aβ40 ratio Amyloid PET than AD (non-AD) as the basis for their symptoms. Therefore, by use of T: Aggregated tau (neurofibrillary tangles) or associated the A/T(N) framework, it is anticipated that we will be able to conduct pathophysiologic state clinical trials in a more finely characterized participant sample. CSF phosphorylated tau Tau PET (N): Neurodegeneration or neuronal injury Anatomic MRI 5 ESTIMATING A+/T+(N)+ PREVALENCE IN DS FDG PET CSF total tau We intend to look at AT(N) classification across the different clin- ical diagnostic categories, that is, Cognitively Stable, Mild Cogni- Abbreviations: Aβ, amyloid beta; AT(N), amyloid, tau, neurodegeneration; CSF, cerebrospinal fluid; FDG, 18F-fluorodeoxyglucose; MRI, magnetic res- tive Impairment, and Dementia in the ABC-DS Study to calculate onance imaging; PET, positron emission tomography A+/T+/(N)+ prevalence and to correlate the various classifica- tions with clinical and cognitive status. Based on a review of the Thus, various biomarker modalities (eg, imaging, biofluids) can be literature,20,36,41,42,59-93 we estimate that, between ages 35 to 55 used to characterize individuals with DS as exhibiting amyloid, tau, or years, there will be 80% A+, 40% T+, and 10% (N)+ for cognitively neurodegeneration “positivity” in the AT(N) classification scheme. The stable adults with DS; 80% A+, 60% T+, and 20% (N)+ for MCI-DS; biomarkers currently being used to characterize individuals with DS and 80% A+, 80% T+, and 60% (N)+ for dementia in the DS group are listed in Table 2. (Figure 1). The AT(N) classification scheme can then be applied to such individ- There may be limitations specific to the A/T(N) classification sys- uals as depicted in Table 3. Briefly, individuals with DS with stable cog- tem. For example, amyloid imaging may underestimate true amy- nition who have no elevations in brain amyloid or tau and no evidence loid positivity. In addition, some biomarkers indicating tau pathol- of neurodegeneration would be classified as A−/T−/N−. Therefore, ogy may become positive at different stages of the disease (ie, CSF they would not be on the AD continuum. However, an individual with becoming abnormal before PET imaging). And, there appears to be a stable cognition who has elevated brain amyloid but no evidence of potential discrepancy between timing of positive MRI indicators of TA B L E 3 AT(N) Framework for adults with Down syndrome Combined syndromal cognitive and biomarker categorization Part of Alzheimer’s AT(N) profiles Biomarker category continuum (Y/N) Stable cognition MCI Dementia A-/T−/(N)− Normal biomarkers No Stable cognition + MCI-DS + normal AD Dementia + normal AD normal AD biomarkers biomarkers biomarkers A+/T−/(N)− AD pathological change Yes Preclinical AD MCI-DS + AD Dementia + AD pathological change pathological change pathological change A+/T+/(N)− AD Yes Preclinical AD Prodromal AD AD + dementia A+/T+/(N)+ AD Yes Preclinical AD Prodromal AD AD + dementia A+/T−/(N)+ AD and concomitant Yes Preclinical AD* Prodromal AD* AD* + dementia suspected non- AD pathological change A−/T+/(N)− Non-AD pathological No Preclinical non-AD MCI not due to AD Non-AD dementia Change A−/T−/(N)+ Non-AD pathological No Preclinical non-AD MCI not due to AD Non-AD dementia Change A−/T+/(N)+ Non-AD pathological No Preclinical non-AD MCI not due to AD Non-AD dementia Change NOTE: AD* = AD and concomitant suspected non-AD pathological changes Abbreviations: AT(N), amyloid, tau, neurodegeneration; AD, Alzheimer’s disease; MCI, mild cognitive impairment. RAFII ET AL . 7 of 10 F I G U R E 1 Estimated prevalence of A/T(N) positivity across clinical diagnoses. We estimate that between ages 35 to 55 years, there will be 80% A+, 40% T+, and 10% (N)+ for cognitively stable adults with Down syndrome (DS); 80% A+, 60% T+, and 20% (N)+ for mild cognitive impairment (MCI)-DS; and 80% A+, 80% T+, and 60% (N)+ for the dementia in DS group. A+ = elevated brain amyloid, T+ = tau pathology present, (N) = neurodegeneration present atrophy (and hence neurodegeneration) versus increased levels of and to facilitate a better understanding of the biology ofAD in adults plasma NfL. Finally, these differences indicate that dichotomization with DS. This research framework is not intended for clinical use at this may potentially decrease sensitivity to changes in cognition. time. As longitudinal data become available, the utility of the AT(N) Further work on longitudinal AD biomarkers in DS should help clar- classification scheme will be compared with each individual’s clinical ify whether the AT(N) classification system can be applied to individ- status over time. We will test if there are differences between the uals with DS both for clinical trial stratification as well as for use as a biochemical and neuroimaging measures of AT(N) . We will confirm potential staging and prognostic tool in the clinic, representing a fun- prevalence of AD biomarker positivity across different ages and clini- damental tool for precision medicine. cal diagnoses. Additionally, we will assess how best we can operational- ize the biomarker binarization to ensure the external validity of the FUNDING INFORMATION results. Finally, we will evaluate the impact of cerebrovascular disease, Michael S. Rafii is supported by 1R61AG066543. including microhemorrhages related to cerebral amyloid angiopathy; neuroinflammation; and, as post mortem data accrue, other pathologies CONFLICTS OF INTEREST such as TDP-43 and Lewy bodies on A/T(N)’s accuracy in staging dis- The authors report no conflicts of interest that are relevant for this ease and predicting clinical status. A similar longitudinal AD biomarker manuscript. study (Horizon21) is ongoing in Europe with plans to harmonize some elements with ABC-DS going forward.44 REFERENCES 1. Knopman DS, Haeberlein SB, Carrillo MC, et al. The National Insti- tute on Aging and the Alzheimer’s Association Research Framework for Alzheimer’s disease: perspectives from the research roundtable. 6 CONCLUSIONS Alzheimers Dement. 2018;14(4):563-575. 2. Jack CR, Jr, Bennett DA, Blennow K, et al. AT(N): an unbiased descrip- Recent work on the AT(N) frameworkin the general population sug- tive classification scheme for Alzheimer disease biomarkers. Neurol- gests that individuals exhibiting abnormalities on all three biomarkers ogy. 2016;87(5):539-547. 3. Jack CR, Jr, Bennett DA, Blennow K, et al. NIA-AA research frame- are at the greatest risk of developing AD dementia. The AT(N) model work: toward a biological definition of Alzheimer’s disease. Alzheimers considers A, T, and (N) status relatively independent from one another Dement. 2018;14(4):535-562. with a known sequential order. The current framework is proposed as 4. Jack CR, Jr, Albert MS, Knopman DS, et al. Introduction to the rec- a starting point for use of A/T(N) classificationin the DS population. It ommendations from the National Institute on Aging-Alzheimer’s Asso- ciation workgroups on diagnostic guidelines for Alzheimer’s disease. is understood that this is not a final model and as new data emerge, the Alzheimers Dement. 2011;7(3):257-262. framework will be revised and updated accordingly in order to parallel 5. Ekman U, Ferreira D, Westman E. The AT(N) biomarker scheme and the state of current knowledge. As with the original AT(N) Framework, patterns of brain atrophy assessed in mild cognitive impairment. Sci- this staging system is intended to aid in the refinement of clinical trials entific Reports. 2018;8(1):8431. 8 of 10 RAFII ET AL . 6. Kern S, Zetterberg H, Kern J, et al. Prevalence of preclinical Alzheimer 26. Rafii MS, Aisen PS. Alzheimer’s disease clinical trials: moving toward disease: comparison of current classification systems. Neurology. successful prevention. CNS Drugs. 2019;33(2):99-106. 2018;90(19):e1682-e1691. 27. Ness S, Rafii M, Aisen P, Krams M, Silverman W, Manji H. Down’s syn- 7. Roe CM, Babulal GM, Stout SH, et al. Using the AT(N) framework to drome and Alzheimer’s disease: towards secondary prevention. Nat examine driving in preclinical AD. Geriatrics (Basel). 2018;3(2). pii: 23. Rev Drug Discov. 2012;11(9):655-656. 8. Food and Drug Administration. Draft Guidance for Industry. 28. Rafii MS. Pro: are we ready to translate Alzheimer’s disease modi- Alzheimer’s disease: developing drugs for the treatment of early stage fying therapies to people with Down syndrome? Alzheimers Res Ther. disease. 2013. http://www.fda.gov.libproxy2.usc.edu/downloads/ 2014;6(5-8):60. Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ 29. Rafii MS. Improving memory and cognition in individuals with Down UCM338287.pdf syndrome. CNS Drugs. 2016;30(7):567-573. 9. Food and Drug Administration. Early Alzheimer’s disease: 30. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen developing drugs for treatment guidance for industry. 2018. E. Mild cognitive impairment: clinical characterization and outcome https://www-fda-gov.libproxy2.usc.edu/downloads/Drugs/ (published correction appears in Arch Neurol 1999 Jun;56(6):760). GuidanceComplianceRegulatoryInformation/Guidances/ Arch Neurol. 1999;56(3):303-308. UCM596728.pdf 31. Ward A, Tardiff S, Dye C, Arrighi HM. Rate of conversion from pro- 10. Wolz R, Schwarz AJ, Gray KR, Yu P, Hill DL. Alzheimer’s Disease dromal Alzheimer’s disease to Alzheimer’s dementia: a systematic Neuroimaging Initiative. Enrichment of clinical trials in MCI due review of the literature. Dement Geriatr Cogn Dis Extra. 2013;3(1):320- to AD using markers of amyloid and neurodegeneration. Neurology. 332. 2016;87(12):1235-1241. 32. Sperling R, Mormino E, Johnson K. The evolution of preclini- 11. Dubois B, Feldman HH, Jacova C, et al. Advancing research diagnos- cal Alzheimer’s disease: implications for prevention trials. Neuron. tic criteria for Alzheimer’s disease: the IWG-2 criteria [published cor- 2014;84(3):608-622. rection appears in Lancet Neurol. 2014 Aug;13(8):757]. Lancet Neurol. 33. Krinsky-McHale SJ, Devenny DA, Kittler P, Silverman W. Selective 2014;13(6):614-629. attention deficits associated with mild cognitive impairment and early 12. Wisniewski KE, Wisniewski HM, Wen GY. Occurrence of neuropatho- stage Alzheimer’s disease in adults with Down syndrome. Am J Ment logical changes and dementia of Alzheimer’s disease in Down’s syn- Retard. 2008;113(5):369-386. drome. Ann Neural. 1985;17:278-282. 34. García-Alba J, Ramírez-Toraño F, Esteba-Castillo S, et al. Neuropsy- 13. Rumble B, Retallack R, Hilbich C, et al. Amyloid A4 and its precur- chological and neurophysiological characterization of mild cognitive sor in Down’s syndrome and Alzheimer’s disease. New Engl J Med. impairment and Alzheimer’s disease in Down syndrome. Neurobiol 1989;320:1446-1462. Aging. 2019;84:70-79. 14. Prasher VP, Farrer MJ, Kessling AM, Fisher EM, West RJ, Barber PC. 35. Jenkins EC, Ye L, Velinov M, et al. Mild cognitive impairment identified Molecular mapping of Alzheimer-type dementia in Down’s syndrome. in older individuals with Down syndrome by reduced telomere signal Ann Neurol. 1998;43:380-383. numbers and shorter telomeres measured in microns. Am J Med Genet 15. Doran E, Keator D, Head E, et al. Down syndrome, partial trisomy 21, B Neuropsychiatr Genet. 2012;159B(5):598-604. and absence of Alzheimer’s disease: the role of APP. J Alzheimers Dis. 36. Mapstone M, Gross TJ, Macciardi F, et al. Metabolic correlates of 2017;56(2):459-470. prevalent mild cognitive impairment and Alzheimer’s disease in adults 16. Mann DM, Alzheimer’s disease and Down’s syndrome. Histopathology. with Down syndrome. Alzheimers Dement (Amst). 2020;12(1):e12028. 1988;13(2):125-137. 37. Lawton MP, Brody EM. Assessment of older people: self-maintaining 17. Mann DM, Brown A, Prinja D, et al. An analysis of the morphol- and instrumental activities of daily living. Gerontologist. 1969;9(3 Part ogy of senile plaques in Down’s syndrome patients of different ages 1):179-186. using immunocytochemical and lectin histochemical techniques. Neu- 38. Schoufour JD, Mitnitski A, Rockwood K, Hilgenkamp TI, Evenhuis HM, ropathol Appl Neurobiol. 1989;15(4):317-329. Echteld MA. Predicting disabilities in daily functioning in older peo- 18. Antonarakis SE, Skotko BG, Rafii MS, et al. Down syndrome. Nat Rev ple with intellectual disabilities using a frailty index. Res Dev Disabil. Dis Primers. 2020;6(1):9. 2014;35(10):2267-2277. 19. Wiseman FK, Al-Janabi T, Hardy J, et al. A genetic cause of Alzheimer 39. Ballard C, Mobley W, Hardy J, Williams G, Corbett A. Dementia in disease: mechanistic insights from Down syndrome. Nat Rev Neurosci. Down’s syndrome. Lancet Neurol. 2016;15(6):622-636. 2015;16(9):564-574. 40. Hithersay R, Hamburg S, Knight B, Strydom A. Cognitive decline and 20. Teller JK, Russo C, DeBusk LM, et al. Presence of soluble amyloid beta- dementia in Down syndrome. Curr Opin Psychiatry. 2017;30(2):102- peptide precedes amyloid plaque formation in Down’s syndrome. Nat 107. Med. 1996;2(1):93-95. 41. Nakamura A, Kaneko N, Villemagne VL, et al. High performance 21. Park J, Oh Y, Chung KC. Two key genes closely implicated with the plasma amyloid-β biomarkers for Alzheimer’s disease. Nature. neuropathological characteristics in Down syndrome: dYRK1A and 2018;554(7691):249-254. RCAN1. BMB Rep. 2009;42(1):6-15. 42. Schindler SE, Bollinger JG, Ovod V, et al. High-precision plasma β- 22. Lautarescu BA, Holland AJ, Zaman SH. The early presentation of amyloid 42/40 predicts current and future brain amyloidosis. Neurol- dementia in people with down syndrome: a systematic review of lon- ogy. 2019;93(17):e1647-e1659. gitudinal studies. Neuropsychol Rev. 2017;27(1):31-45. 43. Handen BL. The search for biomarkers of Alzheimer’s disease in down 23. McCarron M, McCallion P, Reilly E, Mulryan N. A prospective 14-year syndrome. Am J Intellect Dev Disabil. 2020;125(2):97-99. longitudinal follow-up of dementia in persons with Down syndrome. J 44. Strydom A, Coppus A, Blesa R, et al. Alzheimer’s disease in Down Intellect Disabil Res. 2014;58:61-70. syndrome: an overlooked population for prevention trials. Alzheimers 24. McCarron M, McCallion P, Reilly E, Dunne P, Carroll R, Mulryan N. Dement (N Y). 2018;4:703-713. A prospective 20-year longitudinal follow-up of dementia in persons 45. Sheehan R, Sinai A, Bass N, et al. Dementia diagnostic criteria in Down with Down syndrome. J Intellect Disabil Res. 2017;61:843-852. syndrome. Int J Geriatr Psychiatry. 2015;30(8):857-863. 25. Sinai A, Mokrysz C, Bernal J, Bohnen I, Bonell S, Courtenay K. Predic- 46. Smith D, Chicoine B. Difficulties of diagnosing and managing demen- tors of age of diagnosis and survival of Alzheimer’s disease in Down tia in people with Down syndrome. Br J Psychiatry. 2018;213(5):668- syndrome. J Alzheimers Dis. 2018;61:717-728. 669. RAFII ET AL . 9 of 10 47. Prasher VP, Sachdeva N, Tarrant N. Diagnosing dementia in adults with ing of tau pathology and relationship to amyloid, longitudinal MRI, and Down’s syndrome. Neurodegener Dis Manag. 2015;5(3):249-256. cognitive change in down syndrome: results from the down syndrome 48. Dekker AD, Strydom A, Coppus AM, et al. Behavioural and psychologi- biomarker initiative (DSBI). J Alzheimers Dis. 2017;60(2):439-450. cal symptoms of dementia in Down syndrome: early indicators of clin- 67. Klunk WE, Price JC, Mathis CA, et al. Amyloid deposition begins in the ical Alzheimer’s disease? Cortex. 2015;73:36-61. striatum of presenilin-1 mutation carriers from two unrelated pedi- 49. Moran JA, Rafii MS, Keller SM, et al. The National Task Group on intel- grees. J Neurosci. 2007;27(23):6174-6184. lectual disabilities and dementia practices consensus recommenda- 68. Hartley SL, Handen BL, Devenny D, et al. Cognitive decline and brain tions for the evaluation and management of dementia in adults with amyloid-β accumulation across 3 years in adults with Down syndrome. intellectual disabilities. Mayo Clin Proc. 2013;88(8):831-840. Neurobiol Aging. 2017;58:68-76. 50. Holst G, Johansson M, Ahlström G. Signs in people with intellectual dis- 69. Schupf N, Patel B, Silverman W, et al. Elevated plasma amyloid beta- abilities: interviews with managers and staff on the identification pro- peptide 1-42 and onset of dementia in adults with Down syndrome. cess of dementia. Healthcare (Basel). 2018;6(3):103. Neurosci Lett. 2001;301(3):199-203. 51. Firth NC, Startin CM, Hithersay R, et al. Aging related cognitive 70. Mehta PD, Mehta SP, Fedor B, Patrick BA, Emmerling M, Dalton changes associated with Alzheimer’s disease in Down syndrome. Ann AJ. Plasma amyloid beta protein 1-42 levels are increased in old Clin Transl Neurol. 2018;5(6):741-751. Down syndrome but not in young Down syndrome. Neurosci Lett. 52. Startin CM, Hamburg S, Hithersay R, et al. Cognitive markers of 2003;342(3):155-158. preclinical and prodromal Alzheimer’s disease in Down syndrome. 71. Tapiola T, Soininen H, Pirttila T. CSF tau and Abeta42 levels in patients Alzheimers Dement. 2019;15(2):245-257. with Down’s syndrome. Neurology. 2001;56(7):979-980. 53. Sabbagh M, Edgin J. Clinical assessment of cognitive decline in adults 72. Coppus AM, Schuur M, Vergeer J, et al. Plasma beta amyloid and with down syndrome. Curr Alzheimer Res. 2016;13(1):30-34. the risk of Alzheimer’s disease in Down syndrome. Neurobiol Aging. 54. Hartley SL, Handen BL, Devenny DA, et al. Cognitive functioning in 2012;33(9):1988-1994. relation to brain amyloid-β in healthy adults with Down syndrome. 73. Schupf N, Zigman WB, Tang MX, et al. Change in plasma Abeta pep- Brain. 2014;137(Pt 9):2556-2563. tides and onset of dementia in adults with Down syndrome. Neurology. 55. Lessov-Schlaggar CN, Del Rosario OL, Morris JC, Ances BM, Schlaggar 2010;75(18):1639-1644. BL, Constantino JN. Adaptation of the clinical dementia rating scale for 74. Jones EL, Hanney M, Francis PT, Ballard CG. Amyloid beta concentra- adults with Down syndrome. J Neurodev Disord. 2019;11(1):39. tions in older people with Down syndrome and dementia. Neurosci Lett. 56. Sperling RA, Aisen PS, Beckett LA, et al. Toward defining the preclini- 2009;451(2):162-164. cal stages of Alzheimer’s disease: recommendations from the National 75. Head E, Doran E, Nistor M, et al. Plasma amyloid-beta as a function of Institute on Aging-Alzheimer’s Association workgroups on diagnostic age, level of intellectual disability, and presence of dementia in Down guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):280- syndrome. J Alzheimer’s Dis. 2011;23(3):399-409. 292. 76. Matsuoka Y, Andrews HF, Becker AG, et al. The relationship of plasma 57. Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of mild cog- Abeta levels to dementia in aging individuals with Down syndrome. nitive impairment due to Alzheimer’s disease: recommendations from Alzheimer Dis Assoc Disord. 2009;23(4):315-318. the National Institute on Aging-Alzheimer’s Association workgroups 77. Lee NC, Yang SY, Chieh JJ, et al. Blood beta-amyloid and tau in Down on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. syndrome: a comparison with Alzheimer’s disease. Front Aging Neurosci. 2011;7(3):270-279. 2016;8:316. 58. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of 78. Alhajraf F, Ness D, Hye A, Strydom A. Plasma amyloid and tau as dementia due to Alzheimer’s disease: recommendations from the dementia biomarkers in Down syndrome: systematic review and meta- National Institute on Aging-Alzheimer’s Association workgroups on analyses. Dev Neurobiol. 2019;79(7):684-698. diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 79. Tamaoka A, Sekijima Y, Matsuno S, Tokuda T, Shoji S, Ikeda SI. 2011;7(3):263-269. Amyloid beta protein species in cerebrospinal fluid and in brain 59. Lee NC, Chien YH, Hwu WL. A review of biomarkers for Alzheimer’s from patients with Down’s syndrome. Ann Neurol. 1999;46(6): disease in Down syndrome. Neurol Ther. 2017;6(Suppl 1):69-81. 933. 60. Handen BL, Cohen AD, Channamalappa U, et al. Imaging brain amyloid 80. Portelius E, Holtta M, Soininen H, et al. Altered cerebrospinal in nondemented young adults with Down syndrome using Pittsburgh fluid levels of amyloid beta and amyloid precursor-like protein compound B. Alzheimers Dement. 2012;8(6):496-501. 1 peptides in Down’s syndrome. Neuromol Med. 2014;16(2):510- 61. Rafii MS, Wishnek H, Brewer JB, et al. The down syndrome biomarker 516. initiative (DSBI) pilot: proof of concept for deep phenotyping of 81. Fortea J, Carmona-Iragui M, Benejam B, et al. Plasma and CSF Alzheimer’s disease biomarkers in Down syndrome. Front Behav Neu- biomarkers for the diagnosis of Alzheimer’s disease in adults rosci. 2015;9:239. with Down syndrome: a cross-sectional study. Lancet Neurol. 62. Lao PJ, Handen BL, Betthauser TJ, et al. Longitudinal changes in 2018;17(10):860-869. amyloid positron emission tomography and volumetric magnetic res- 82. Margallo-Lana ML, Moore PB, Kay DW, et al. Fifteen-year follow-up of onance imaging in the nondemented Down syndrome population. 92 hospitalized adults with Down’s syndrome: incidence of cognitive Alzheimers Dement (Amst). 2017;9:1-9. decline, its relationship to age and neuropathology. J Intellect Disabil 63. Abrahamson EE, Head E, Lott IT, et al. Neuropathological corre- Res. 2007;51(Pt. 6):463-477. lates of amyloid PET imaging in Down syndrome. Dev Neurobiol. 83. Kasai T, Tatebe H, Kondo M, et al. Increased levels of plasma total tau 2019;79(7):750-766. in adult Down syndrome. PLoS One. 2017;12(11):e0188802. 64. Fortea J, Vilaplana E, Carmona-Iragui M, etal. Clinical and biomarker 84. Startin CM, Ashton NJ, Hamburg S, et al. Plasma biomarkers for amy- changes of Alzheimer’s disease in adults with Down syndrome: a cross- loid, tau, and cytokines in Down syndrome and sporadic Alzheimer’s sectional study. Lancet (in Press). 2020;395 10242: disease. Alzheimers Res Ther. 2019;11(1):26. 65. Lao PJ, Handen BL, Betthauser TJ, et al. Alzheimer-like pattern of 85. Fang WQ, Hwu WL, Chien YH, et al. Composite scores of plasma tau hypometabolism emerges with elevated amyloid-β burden in Down and β-amyloids correlate with dementia in Down syndrome. ACS Chem syndrome. J Alzheimers Dis. 2018;61(2):631-644. Neurosci. 2020;11(2):191-196. 66. Rafii MS, Lukic AS, Andrews RD, et al, Down Syndrome Biomarker Ini- 86. Head E, Powell DK, Schmitt FA. Metabolic and vascular imaging tiative and the Alzheimer’s Disease Neuroimaging Initiative. PET imag- biomarkers in Down syndrome provide unique insights into brain 10 of 10 RAFII ET AL . aging and Alzheimer disease pathogenesis. Front Aging Neurosci. prodromal phase of Alzheimer’s disease. Am J Psychiatry. 2018;10:191. 2002;159(1):74-81. 87. Matthews DC, Lukic AS, Andrews RD, et al. Dissociation of Down 92. Strydom A, Heslegrave A, Startin CM, et al. Neurofilament light syndrome and Alzheimer’s disease effects with imaging. Alzheimers as a blood biomarker for neurodegeneration in Down syndrome. Dement (N Y). 2016;2(2):69-81. Alzheimers Res Ther. 2018;10(1):39. 88. Schapiro MB, Ball MJ, Grady CL, Haxby JV, Kaye JA, Rapoport 93. Rafii MS, Donohue MC, Matthews DC, et al. Plasma neurofilament SI. Dementia in Down’s syndrome: cerebral glucose utilization, light and Alzheimer’s disease biomarkers in down syndrome: results neuropsychological assessment, and neuropathology. Neurology. from the Down syndrome biomarker initiative (DSBI). J Alzheimers Dis. 1988;38(6):938-942. 2019;70(1):131-138. 89. Teipel SJ, Schapiro MB, Alexander GE, et al. Relation of corpus callo- sum and hippocampal size to age in nondemented adults with Down’s syndrome. Am J Psychiatry. 2003;160(10):1870-1878. 90. Aylward EH, Li Q, Honeycutt NA, et al. MRI volumes of the hippocam- How to cite this article: Rafii MS, Ances BM, Schupf N, et al. pus and amygdala in adults with Down’s syndrome with and without The AT(N) framework for Alzheimer’s disease in adults with dementia. Am J Psychiatry. 1999;156(4):564-568. Down syndrome. Alzheimer’s Dement. 2020;12:e12062. 91. Krasuski JS, Alexander GE, Horwitz B, Rapoport SI, Schapiro MB. Rela- tion of medial temporal lobe volumes to age and memory function in https://doi.org/10.1002/dad2.12062 nondemented adults with Down’s syndrome: implications for the