The n e w e ng l a n d j o u r na l of m e dic i n e

original article

Clinical and Biomarker Changes
in Dominantly Inherited Alzheimer’s Disease
Randall J. Bateman, M.D., Chengjie Xiong, Ph.D.,
Tammie L.S. Benzinger, M.D., Ph.D., Anne M. Fagan, Ph.D.,
Alison Goate, Ph.D., Nick C. Fox, M.D., Daniel S. Marcus, Ph.D.,
Nigel J. Cairns, Ph.D., Xianyun Xie, M.S., Tyler M. Blazey, B.S.,
David M. Holtzman, M.D., Anna Santacruz, B.S., Virginia Buckles, Ph.D.,
Angela Oliver, R.N., Krista Moulder, Ph.D., Paul S. Aisen, M.D.,
Bernardino Ghetti, M.D., William E. Klunk, M.D., Eric McDade, M.D.,
Ralph N. Martins, Ph.D., Colin L. Masters, M.D., Richard Mayeux, M.D.,
John M. Ringman, M.D., Martin N. Rossor, M.D., Peter R. Schofield, Ph.D., D.Sc.,
Reisa A. Sperling, M.D., Stephen Salloway, M.D., and John C. Morris, M.D.,
for the Dominantly Inherited Alzheimer Network
The order and magnitude of pathologic processes in Alzheimer’s disease are not well The authors’ affiliations are listed in the
understood, partly because the disease develops over many years. Autosomal domi- Appendix. Address reprint requests to
Dr. Bateman at the Washington Univer-
nant Alzheimer’s disease has a predictable age at onset and provides an opportunity sity School of Medicine, Department of
to determine the sequence and magnitude of pathologic changes that culminate in Neurology, 660 S. Euclid Ave., Box 8111,
symptomatic disease. St. Louis, MO 63110, or at batemanr@
This article was published on July 11, 2012,
In this prospective, longitudinal study, we analyzed data from 128 participants who at
underwent baseline clinical and cognitive assessments, brain imaging, and cerebro-
N Engl J Med 2012.
spinal fluid (CSF) and blood tests. We used the participant’s age at baseline assess- DOI: 10.1056/NEJMoa1202753
ment and the parent’s age at the onset of symptoms of Alzheimer’s disease to calcu- Copyright © 2012 Massachusetts Medical Society.

late the estimated years from expected symptom onset (age of the participant minus
parent’s age at symptom onset). We conducted cross-sectional analyses of baseline
data in relation to estimated years from expected symptom onset in order to deter-
mine the relative order and magnitude of pathophysiological changes.
Concentrations of amyloid-beta (Aβ)42 in the CSF appeared to decline 25 years before
expected symptom onset. Aβ deposition, as measured by positron-emission tomog-
raphy with the use of Pittsburgh compound B, was detected 15 years before expected
symptom onset. Increased concentrations of tau protein in the CSF and an increase
in brain atrophy were detected 15 years before expected symptom onset. Cerebral
hypometabolism and impaired episodic memory were observed 10 years before ex-
pected symptom onset. Global cognitive impairment, as measured by the Mini–Mental
State Examination and the Clinical Dementia Rating scale, was detected 5 years be-
fore expected symptom onset, and patients met diagnostic criteria for dementia at
an average of 3 years after expected symptom onset.
We found that autosomal dominant Alzheimer’s disease was associated with a se-
ries of pathophysiological changes over decades in CSF biochemical markers of
Alzheimer’s disease, brain amyloid deposition, and brain metabolism as well as
progressive cognitive impairment. Our results require confirmation with the use of
longitudinal data and may not apply to patients with sporadic Alzheimer’s disease.
(Funded by the National Institute on Aging and others; DIAN
number, NCT00869817.)
n engl j med 1
The New England Journal of Medicine
Downloaded from on July 11, 2012. For personal use only. No other uses without permission.
Copyright © 2012 Massachusetts Medical Society. All rights reserved.

7 mentary Appendix. imaging. longitudinal studies of Alzheimer’s mutation for autosomal dominant Alz­hei­mer’s disease biomarkers take many years to show the disease.6 Therefore. available with the full text of On the basis of the amyloid hypothesis. well-validated assessments between January 26. 2009. Clinical feedback was 2 n engl j med  nejm. The n e w e ng l a n d j o u r na l of m e dic i n e A  lzheimer’s disease is the most com. CDR 0. estimated to affect more than 5 million cognitive. Although autosomal cognitive change with the use of the Clinical De- dominant Alz­hei­mer’s disease accounts for a mentia Rating (CDR)16 scale. the order and timing of amy. cognitive. Clinicians who performed the assess- clinical onset of autosomal dominant Alz­hei­mer’s ments were not aware of the mutation status of disease is similar between generations14 and is participants. with an expected in. ticipant was a member of a pedigree with a known fest. increasing evidence13 impairment.5. and CDR 1 mild impairment. imaging. We hypothesized that autosomal domi- nant Alz­hei­mer’s disease and the more common CLINICAL ASSESSMENTS late-onset Alz­hei­mer’s disease12 have similar Participants underwent clinical assessments of pathophysiological features. All study procedures were approved by the cates that the targeting of amyloidosis in familial Washington University Human Research Protec- amyloid polyneuropathy improves clinical out. Participants at risk for carrying a mutation for ed to increase to a trillion dollars by 2050 unless autosomal dominant Alz­hei­mer’s disease were disease-modifying treatments are developed. were compared between mutation carriers and crease to 13 million by the year 2050. a loss of independence and causing a heavy per- sonal toll on the patient and the family. tion Office and the local institutional review boards comes. For personal use only. stood. All rights reserved. Copyright © 2012 Massachusetts Medical very mild cases of Alz­hei­mer’s DIAN participants are assessed at base- full pathologic cascade of events that lead to The New England Journal of Medicine Downloaded from nejm. and the biomarkers of Alzheimer’s disease processes are first data-cutoff point (April 28. and biochemical measures people in the United States. of the participating sites. The costs Me thods of care of patients with Alzheimer’s disease in 2010 were estimated at more than $172 billion in STUDY DESIGN the United States.8 this article at NEJM. Clinical. S1 in the Sup- tation carriers and noncarriers as a function of plementary Appendix). Each par- begin decades before the first symptoms mani. Alz­hei­mer’s disease and medical history. Furthermore. All authors vouch for the loidosis and other Alz­hei­mer’s disease processes accuracy of the data and the fidelity of the study to that lead to clinical dementia are not well under.15 We compared a wide family members were asked about the age of first range of pathophysiological markers between mu. ultimately leading to disease. The parental age at onset was deter- affected mostly by the mutation type and back. All participants provided amyloid-beta (Aβ) is currently the most common written informed consent or assent with proxy con- disease-modifying target. progressive cognitive decline (Fig. . 2012. an annual cost that is predict. The typical noncarriers in the first large international cohort clinical presentation is progressive loss of mem. in- alterations in Aβ processing and lead to Alz­hei­ cluding a neurologic evaluation (see the study mer’s disease with complete penetrance. The suggests that it overlaps with sporadic Alz­hei­mer’s DIAN assessments ascertained family history of disease. mined by a semistructured interview in which ground family on July 11. The age at protocol). line and in subsequent years with comprehensive mentia.1 enrolled in the Dominantly Inherited Alzheimer Alzheimer’s disease has been hypothesized to Network (DIAN) study at 1 of 10 sites. sent. the protocol (available at NEJM. 2011) went through needed to improve the design of clinical trials. the parental age at onset in order to evaluate the mon cause of dementia and is currently cascade of events that lead to dementia. Recent research indi. and biochemical as- treatment require large numbers of patients over sessments. of families with autosomal dominant Alz­hei­mer’s ory and cognitive function. Data from all 128 participants who extended periods owing to the slow progression of were enrolled and who had completed baseline cognitive symptoms. trials of disease-modifying clinical. with CDR 0 indicat- relatively small proportion (approximately 1%) of ing normal cognitive function. No other uses without permission.2-4 Thus. and par- PSEN1.9-11 However. and PSEN2) have been identified that cause ticipants underwent a physical examination. Mutations in one of three genes (APP. quality-control checks and were included in the develop more effective therapeutics. and offer the analysis (see the Methods section in the Supple- opportunity for prevention trials.

initial and ongoing quality Assay. (%)† two tests are reported here because of space lim- Symptomatic 43 (49) 1 (2) 0. For each ent’s age at onset was 45 years. cognitive. (%) nitive function. then coregistered with individual MRI images if the participant’s age was 35 years. Samples were shipped on dry ice to 9.92 NEUROPSYCHOLOGICAL TESTING Male sex — no. Aβx-40.tical analyses (see the Supplementary Appendix ed in the morning under fasting conditions by for details) were conducted with the use of the means of lumbar puncture and venipuncture.04 of neuropsychological tests. with scores rang.STATISTICAL ANALYSIS tary Appendix). with scores ranging from 0 (se- vere impairment) to 30 (no impairment).1±10. but results of only Cognitive status — no. For example. imaging.85 Participants underwent a comprehensive battery Education level — yr 13. scans from DIAN participants were processed through FreeSurfer (for details. see the Supplemen. Positive for apolipoprotein E 22 (25) 9 (22) 0. S1 in the Supplementary Appendix). An increased PIB interview with the use of all available historical SUVR indicates increased binding to fibrillar data ( on July 11.5±8.* No other research data. The parental age at the onset of clinical the use of a hand-drawn reference region en. kit “controls” within Neuroimaging Initiative (ADNI) protocol. Characteristic (N = 88) (N = 40) P Value Age — yr 39. and measurement consistency between pliance with the acquisition protocol by the ADNI plates of a common sample that was included in imaging core laboratories.19 All the expected range as defined by the manufac- images were reviewed for image quality and com.5 15. Table 1.The estimated years from expected symptom on- emission tomography (PET) with the use of fluo. the standardized mated years from expected symptom onset would uptake value ratio (SUVR) was calculated with be −10. . Dominantly Inherited Alzheimer’s Disease provided to participants if medically indicated.3 (SAS Institute). Statis- Cerebrospinal fluid (CSF) and blood were collect. memory.3 39.69 ε4 allele — no. The T1-weighted MRI each run. With each marker treated as a n engl j med  nejm.symptoms was determined by a semistructured compassing the brain stem.tions in the CSF of Aβ1-42. including a coefficient performed according to the Alz­hei­mer’s Disease of variation of 25% or less. Concentra- again after a 30-minute delay. 2012. Innogenetics).5 0. measures were compared as a function of esti- mated years from expected symptom onset be- BIOCHEMICAL ANALYSIS tween mutation carriers and noncarriers.9±2. and the par- for region-of-interest determination. Innogenetics). (%) 36 (41) 17 (42) 0.turer. version spectively. Copyright © 2012 Massachusetts Medical Society. were provided to research participants as part of Carriers Noncarriers the 3 The New England Journal of Medicine Downloaded from nejm.9 0. and a decreased FDG SUVR indicates Clinical. total tau. then the esti- FreeSurfer region of interest. Characteristics of the Study Participants. re. from the Logical Memory subtest of the Wechsler † Participants were defined as asymptomatic if they had Cognitive Dementia Memory Scale–Revised18 is a measure of episodic Rating scores of 0 (no cognitive decline) and as symptomatic if they had scores greater than 0. Images obtained through positron. including genetic status. No other uses without permission. amyloid. For personal use only. Story A * Plus–minus values are means ±SD. The Mini–Mental State Ex- 17 amination (MMSE) is a measure of general cog. Aβ1-42. both tests are widely used in research on Asymptomatic 45 (51) 39 (98) Alz­hei­mer’s disease. as were levels Volumetric magnetic resonance imaging (MRI) of plasma Aβ species (Aβ1-40. and biochemical decreased metabolism.29 itations. and was performed with the use of qualified 3-tesla Aβx-42) (INNO-BIA Plasma Aβ Forms Multiplex scanners at each site. phosphorylated at threonine 181 were measured by immunoassay (INNOTEST β-Amyloid1-42 and BRAIN IMAGING INNO-BIA AlzBio3.0±2.set were calculated as the age of the participant rodeoxyglucose (FDG) and Pittsburgh compound B at the time of the study assessment minus the (PIB) (FDG-PET and PIB-PET. and tau ing from 0 (no recall) to 25 (complete recall). All rights reserved.PROC MIXED procedure in SAS software. All values had to meet control and matching between site scanners were quality-control standards. respectively) were age of the parent at symptom onset. Participants recall as many details as they can from a short story containing 25 bits of infor- mation after it is read aloud by the examiner and the DIAN biomarker core laboratory.

16 2.5±1.3 Difference 0.7§ Aβ deposition in the precuneus (SUVR ratio)†† Noncarriers 0.5±4.)† Noncarriers 0 0 0 0 0 0 0 0 Carriers 0 0 0.1±2.02±0.94 1.24 Difference 0.5 The New England Journal of Medicine Downloaded from nejm.8 11.7 Carriers 29.01±0.9 −3.3 15. For personal use only.4 Difference 1.9 1.4±1.)¶ Noncarriers 29.42±0.6 29.2§ −8.36 1.7 −1.8 1.1‡ 2.g. of estimated years from expected symptom onset mate Student’s t-test results derived from the — that is.17§ 0.25±0.09§ −0.28 0.3±2.69 0.2±2.08§ −0.0** −4.67±0. Values R e sult s for individual participants were not displayed on graphs to protect the confidentiality of the mu.67 Carriers 0.7±1.61 1.15‡ 0.1 11.12§ −0.9 14.3 10.2§ −10.5§ −10.2±1.1 13.24 1.69 0.16±0.2±2.5±2.0±2. The n e w e ng l a n d j o u r na l of m e dic i n e continuous scale.01 1.2§ −6.6 15.07±0.69 0.7 15.7 Difference 0±2.0±1.16 0. Clinical.9 29.7±1.7§ MMSE score (no.70 0.38 1.54±0.69 0.90 Carriers 2. and biochemical measures.4 0.6 29.04 2.0§ −7.50 1. imaging.7 29.)‖ Noncarriers 14.5±2.3 5.21±0. Imaging. The mutation types reflected the dis- could deduce it from individual values of estimat.2 0.0 −2.4±3.1 0.0±2.4±1.1 1.71 on July 11. after adjustment for the correlation among family members.1§ 4. and apolipoprotein mutation carriers and noncarriers as a function E (APOE) status (positive or negative).* Variable Estimated Years from Expected Symptom Onset −25 −20 −15 −10 −5 0 5 10 CDR-SOB score (no.. with years from expected symptom onset. mutation the use of the standardized difference between status (carrier or noncarrier).2 0.12§ 0.92 1.5 24.3±3.72 1. Approxi.97 1. Cognitive.57±0.0‡ −5. and Biochemical Estimates in Mutation Carriers and Noncarriers.5 29. Table 2.15§ 0.21§ Glucose metabolism in the precuneus (SUVR ratio)‡‡ Noncarriers 2.1 19.2 0±1.5 29.83 1.13 −0.4 8. tribution of mutations described in the literature.1 28. a participant We analyzed 128 participants from the DIAN co- who did not know his or her mutation status hort (Table 1).15§ Total hippocampal volume (mm3) Noncarriers 8999 8874 8748 8622 8497 8371 8245 8120 Carriers 8767 8511 8255 7999 7743 7486 7230 6974 Difference −232±675 −363±548 −493±442** −623±370‡ −754±356§ −885±406§ −1015±500§ −1146±619§ 4 n engl j med  nejm.09§ −0. Copyright © 2012 Massachusetts Medical Society.9 12.6 22. Supplementary Appendix).11 −0.13§ 0.99 1.70±0. the predicted difference at each esti- model were used to determine whether marker mated year from expected symptom onset divid- values differed between mutation carriers and ed by the standard deviation for clinical.3§ 5.1§ Logical Memory score (no.5 11.06 2.7 1. Figure 2 to model each marker as a function of estimated was generated with the same final models.38±0.5 2. cogni- noncarriers at certain age points (Table S4 in the tive.5 −0.6 4.9 −1.34±0.4 29.70 0.6±1.39 Difference 0.51±0. No other uses without permission.05 1.17 0.68 0. a linear mixed model was used ed years from expected symptom onset).6 29.1 26.10±0.0±2.07±0.08 1.95 1. All rights reserved. 2012. STUDY PARTICIPANTS tation status of participants (e.6±4.76 0.0 0.1±2.2 2.0 5. .7 Carriers 16.

(Continued. Estimates are reported with 95% confidence intervals for the difference between mutation carriers and noncarriers.8§ 13.5 43. calculated as the age of the participant at assessment minus the age of the parent at symptom onset. No other uses without permission.9 Difference 5. of 3.4±10. 1A participants were mutation carriers. 0.6 41. P<0. carriers had decreasing of the Clinical Dementia Rating–Sum of Boxes MMSE scores at higher values of estimated years (CDR-SOB).5§ 16. 1B). 2012.) Variable Estimated Years from Expected Symptom Onset −25 −20 −15 −10 −5 0 5 10 Aβ42 in the CSF (mg/ml) Noncarriers 454 436 421 410 402 398 397 399 Carriers 532 433 352 289 245 218 210 219 Difference 78±149 −3±97 −69±82 −121±85‡ −157±88§ −180±90§ −187±107‡ −180±155** Tau in the CSF (pg/ml) Noncarriers 39 41 42 44 45 47 48 50 Carriers 35 55 76 97 117 138 159 179 Difference −4±52 14±41 34±33** 53±27§ 72±27§ 91±31§ 111±39§ 129±50§ Plasma Aβ42 (pg/ml) Noncarriers 37.9 42.1 31. Noncarriers had stable CDR-SOB significant differences in the presence of an APOE scores of 0 throughout the relative age range.01. and interactions between mutation status and estimated year from expected symptom onset as covariates regardless of the participant’s score on the Clinical Dementia Rating–Sum of Boxes (CDR-SOB) (Table S3 in the Supplementary Appendix). † Scores on the CDR-SOB range from 0 (cognitive normality) to 18 (maximal cognitive impairment). participants had a The DIAN parental age of symptom onset was 5 The New England Journal of Medicine Downloaded from nejm. ** P<0.2±6.001.0 34.0 46. significant cognitive impairment in mutation car- n engl j med  nejm.5§ * The timing of assessments was defined on the basis of the estimated years from expected symptom onset.8 30. tom onset. and 8 APP pedigrees. We found normality) to 18 (maximal cognitive impairment).8 7.7±5.3 30. In this cohort.7±6.5±7. CDR rating of mild dementia (CDR 1) at a mean related with the age of symptom onset for symp. ‡ P<0.0±8. ¶ Scores on the Mini–Mental State Examination (MMSE) range from 0 (severe impairment) to 30 (no impairment). estimated year from expected symptom onset (or higher-order term).5 Carriers 42.3±5. and SUVR standardized uptake value ratio. †† Deposition of amyloid-beta (Aβ) in the precuneus was measured by positron-emission tomography (PET) with the use of Pittsburgh com- pound B (PIB). with 40 PSEN1.2±6. A lower SUVR indicates lower metabolism. Significant differences in MMSE scores be- tween mutation carriers and noncarriers were de- CLINICAL and NEUROPSYCHOMETRIC FINDINGS tected at assessments performed 5 years before We measured clinical impairment with the use expected symptom onset.4 41. Dominantly Inherited Alzheimer’s Disease Table 2. The mean (±SD) age of at higher values of estimated years from expected parental onset of symptoms was 45.001). symptom on July 11. A score higher than 27 is considered normal.5 45. ε4 allele or sex between asymptomatic mutation whereas carriers had increasing CDR-SOB scores carriers and noncarriers.8 33.7 30. 3 PSEN2. All rights reserved. approximately 50% of the asymptomatic ers 5 years before expected symptom onset (Fig.8 41.3 years after the parent’s age of symp- tomatic offspring (Pearson correlation coefficient. Copyright © 2012 Massachusetts Medical Society.8 years. A higher SUVR indicates greater binding of PIB to fibrillar amyloid.4±10.0** 10.8§ 14. CSF denotes cerebrospinal fluid. There were no and Table 2). § P<0.20 As Significant differences in CDR-SOB scores were expected with an autosomal dominant inheritance detected between mutation carriers and noncarri- pattern. Estimates were obtained with use of a mixed model that treated mutation status (noncarrier or carrier). For personal use only.05.8±8.7 8. . ‖ Scores on the Logical Memory subtest of the Wechsler Memory Scale–Revised range from 0 (no recall) to 25 (complete recall). ‡‡ Glucose metabolism in the precuneus was measured by PET with the use of fluorodeoxyglucose.56.2 30.0‡ 11. with scores ranging from 0 (cognitive from expected symptom onset (Fig.

and deposition of amyloid-beta (Aβ) in the precuneus. In- symptom onset (Fig. as measured by PET with the use of Pitts- burgh compound B (Panel F). and Biochemical Changes in Autosomal Dominant Alzheimer’s Disease Mutation Carriers versus Noncarriers. as measured by positron-emission tomography (PET) with the use of fluorodeoxyglucose (Panel E). the Mini–Mental State Examination (scores range from 0 [severe impairment] to 30 [no impairment]) (Panel B). The clinical and cognitive measures of the Clinical Dementia Rating–Sum of Boxes (scores range from 0 [cognitive normality] to 18 [maximal cognitive impairment]) (Panel A).org on July 11. Plasma Aβ42 levels were elevated throughout the range of estimated years from expected symptom onset (Panel I). occurred approximately 10 years before expected symptom onset.000 1. For personal use only.8 1. creased atrophy of bilateral hippocampi was de- 6 n engl j med  nejm. The New England Journal of Medicine Downloaded from nejm. MRI measures of hippocampal volume (Panel D) showed increased brain atrophy approximately 15 years before expected symptom onset.0 SUVR SUVR 8. . Metabolic. 1C). in the de. Decreases in cerebral glucose metabolism. and the Logical Memory subtest of the Wechsler Memory Scale–Revised (scores range from 0 [no recall] to 25 [complete recall]) (Panel C) showed impaired ratings beginning approximately 5 to 10 years before expected symptom onset. and levels of Aβ42 (Panel H) decreased at least 15 years before expected symptom onset. The n e w e ng l a n d j o u r na l of m e dic i n e Noncarriers Carriers A Clinical Dementia Rating–Sum of Boxes B Mini–Mental State Examination C Logical Memory 8 20 6 30 15 4 25 Score Score Score 10 2 20 5 0 −2 15 0 −30 −20 −10 0 10 20 −30 −20 −10 0 10 20 −30 −20 −10 0 10 20 Estimated Yr from Symptom Onset Estimated Yr from Symptom Onset Estimated Yr from Symptom Onset D Hippocampal Volume E Glucose Metabolism in the Precuneus F Aβ Deposition in the Precuneus 10.4 1.2 1.000 2. 2012. All rights reserved. riers. Dashed lines represent 95% confidence intervals of the fitted curves. BRAIN ATROPHY layed-recall portion of the Logical Memory test21 MRI structural measures of hippocampal volume 10 years before expected symptom onset (Table 2). SUVR denotes standardized uptake value ratio.000 mm3 1.2 0. In the cerebrospinal fluid (CSF).8 7. levels of tau protein (Panel G) increased beginning 10 to 15 years before expected symptom onset.000 1.6 6.000 2. as compared with noncarriers.0 1.0 0. Cross-Sectional Analyses of Clinical. began approximately 15 to 20 years before expected symptom onset.4 1. Copyright © 2012 Massachusetts Medical Society. were compared between mutation carriers and Noncarriers remained stable in performance noncarriers with the use of an a priori hypothesis from 30 years before to 20 years after expected of increased atrophy in mutation carriers.4 2.6 0. Cognitive. No other uses without permission.2 9. According to Estimated Years from Expected Symptom Onset.4 −30 −20 −10 0 10 20 −30 −20 −10 0 10 20 −30 −20 −10 0 10 20 Estimated Yr from Symptom Onset Estimated Yr from Symptom Onset Estimated Yr from Symptom Onset G CSF Tau H CSF Aβ42 I Plasma Aβ42 250 700 55 200 600 50 500 45 150 pg/ml pg/ml pg/ml 400 40 100 300 35 50 200 30 0 100 25 −30 −20 −10 0 10 20 −30 −20 −10 0 10 20 −30 −20 −10 0 10 20 Estimated Yr from Symptom Onset Estimated Yr from Symptom Onset Estimated Yr from Symptom Onset Figure 1.

There lism. 1F and Table 2). 1G and Table 2). and tom onset ( 7 The New England Journal of Medicine Downloaded from nejm. n engl j med  nejm. before expected symptom onset. riers. as compared with A video showing the CSF decreased as a function of estimated years those in noncarriers.24 was chosen for analysis of −30 −20 −10 0 10 amyloid deposition. Cognitive.23. global cognitive impairment starting at 5 years carriers (Fig. cognitive. mentia Rating–Sum of Boxes [CDR-SOB]) occurring later. Copyright © 2012 Massachusetts Medical Society. Appendix. and see Video 1. Dominantly Inherited Alzheimer’s Disease tected in mutation carriers 15 years before expect- ed symptom onset (Table 2). then increased tau a priori hypothesis of increased regional amounts in the CSF (CSF tau). mutation carriers had significant amyloid deposi- tion in the precuneus 15 years before expected symptom onset (Fig.88. The amount of amyloid deposition in mutation carriers in. 1D). levels of tau in the CSF and in brain atrophy were crease by half in Aβ42 in the CSF and the increase detected approximately 15 years before expected in tau in the CSF were similar in magnitude to symptom onset.22 CSF tau 1 Standardized Difference CSF Aβ42 CDR–SOB CEREBRAL GLUCOSE METABOLISM FDG-PET measures of cerebral glucose use in the precuneus were compared with the use of an a 0 priori hypothesis of decreased metabolism in mutation carriers to determine regional metabolic Hippocampal −1 volume defects. The precuneus region. Aβ42 concentrations in the CSF in mutation (Fig. In mutation carriers. and BIOCHEMICAL MEASURES biochemical measures in the DIAN cohort (Fig. imaging. in autosomal dominant Alz­hei­mer’s disease were tom onset.26 Plasma Aβ42 levels were was detected at least 15 years before symptom onset.3 years before expected symptom onset. followed by cerebral hypometab- those typically observed in late-onset sporadic Alz­ olism and impaired episodic memory approximate- hei­mer’s disease. 1E and Table 2). there 2 Aβ deposition was an age-dependent decrease in hippocampal volumes in noncarriers (Fig. Biochemical Changes as a Function of Estimated Years from Expected Symptom Onset. ly 10 years before expected symptom onset and ed in mutation carriers. The order of differences suggests decreasing Aβ42 in the precuneus were compared with the use of an the CSF (CSF Aβ42). 2). 2012. The order and rate of pathophysiological changes pected symptom onset at least until clinical symp. No other uses without permission. 1I). . 1H). Aβ deposition as measured Aβ deposition over time is available from expected symptom onset and were pseudo. reaching low levels 10 years be. Comparison of Clinical. For personal use only. followed by hippocampal atrophy and hypometabo- of amyloid deposition in mutation carriers. The de. 3. Concentrations of Aβ42 in carriers appeared to decline. available at at NEJM. Increases in fore expected symptom onset ( on July 11. A significant decrease in cere- Estimated Yr from Expected Symptom Onset bral metabolism in the precuneus was detected in mutation carriers 10 years before expected symp. expected symptom onset (Table 2). estimated with the use of an analysis of the rela- tionship among clinical. COMBINED MODEL creased as a function of estimated years from ex. followed by fibrillar Aβ deposition. with cognitive and clinical changes (as measured by the Clinical De- was no detectable amyloid deposition in normal at approximately 20 years before expected NEJM. which is known to be an area of early deposition in both sporadic Alz­hei­mer’s disease and autosomal dominant Alz­ Glucose −2 metabolism hei­mer’s disease. As compared with noncarriers. As expected. levels of tau in the CSF were Beginning 25 years before expected symptom on- increased 15 years before expected symptom onset set. by PIB-PET (Fig. as compared with non. Mild dementia (CDR 1) occurred an average of 3. A β DEPOSITION The normalized differences between mutation carriers and noncarriers are shown versus estimated years from expected symptom onset and plotted PIB-PET measures of fibrillar Aβ deposition25 in with a fitted curve.4. All noncarriers had PIB-PET SUVR values of 95% confidence interval bands are shown in Figure S2 in the Supplementary less than 0. All rights reserved. Figure 2. Structural. Metabolic.

4 as Aβ misfolding and other modulating factors. indicating high amounts of amyloid.8 before dementia symptoms are manifest in spo- 0. Al- suggest that the diagnosis of clinical dementia is though the clinical and pathologic phenotypes of made late in the course of the biologic cascade of dominantly inherited Alz­hei­mer’s disease are autosomal dominant Alz­hei­mer’s disease. hei­mer’s disease. we have estimat. low (<15%) in this cohort and not significantly ment and eventually dementia. With the use of estimates of Although the findings of this study were largely years from expected symptom onset. which culminates in clinical impair. 2012. of the average of autosomal dominant Alzheimer’s disease mutation carriers and noncarriers 20 years before the “sporadic” form. 1. although many of our findings in autosomal dominant Alz­hei­mer’s disease are similar to findings in Discussion sporadic Alz­hei­mer’s disease. For personal use only. to high SUVR values sectional data. which do not represent individual (redder colors). No other uses without permission.2 Our results support the hy- Symptom Onset –20 –10 0 pothesis of a pathophysiological cascade and Figure 3. These findings different between carriers and The New England Journal of Medicine Downloaded from nejm. the estimated time of symptom onset. All rights reserved. disease mutations cause increased Aβ or Aβ42 ed the timing and order of changes in autosomal production. the prevalence these biologic changes. Changes begin in the brain at disease typically presents with early and pro- least two decades before the estimated onset of nounced PIB-PET signaling in the neostriatum. autosomal dominant Alz­hei­mer’s with the disease. iology between autosomal dominant Alz­hei­ Panel A compares the fibrillar Aβ deposition. Copyright © 2012 Massachusetts Medical Society. The scale ranges from low SUVR values because the current analyses are based on cross- (bluer colors).24 Previous cross-sec- tional studies in sporadic Alz­hei­mer’s disease have suggested a series of changes that lead to Estimated Yr from clinical disease. indicating low amounts of amyloid. results among the mutation gene types. which is fol. plementary Appendix) suggests no differences in lowed by brain amyloidosis. trends for increased Previous studies of autosomal dominant Alz­hei­ levels of Aβ42 in the CSF have not been reported mer’s disease have showed hippocampal atrophy. An increased SUVR indicates increased binding of PIB to fibrillar amyloid. a comparison with and magnitude of changes indicate that genetic PSEN2 and APP mutations (Table S2 in the Sup- mutations cause increased Aβ42. The estimated year from expected symptom the generalizability of the nature and sequence onset normalizes the stage of disease on the of brain changes in autosomal dominant Alz­hei­ 8 n engl j med  nejm. estimated time of onset of on July 11.0 loid deposition probably occurs years or decades 0. other findings suggest that amy- SUVR 1. shows additional Aβ deposition throughout the cortex and neostriatum at interpretations of the results are not certain. Panel C occur over a period of four decades. For example. cognitive impairment can of confounders such as vascular risk factors was be detected. Alz­hei­mer’s disease processes are likely independent of absolute age Carriers but rather depend on the start of processes such 1. Our findings suggest that once initiated. the order based on PSEN1 mutations. as compared with noncarriers (Panel B).6 Noncarriers radic Alz­hei­mer’s disease. a worldwide network of autosomal domi. After to the younger age of the cohort. The n e w e ng l a n d j o u r na l of m e dic i n e A B C basis of the parental age of onset.28 clinical symptoms. unlike sporadic Alz­hei­ dominant Alz­hei­mer’s disease in a large cohort mer’s disease.28 and biochemical abnor. tauopathy. . longitudinal changes.6 1. brain at. There was significant Aβ deposition A strength of this study is that it shows rela- in the caudate and cortex in mutation carriers more than 10 years before tive changes in Alz­hei­mer’s disease processes that expected symptom onset. similar to those of sporadic Alz­hei­mer’s disease.2 Furthermore. In addition. and decreased glucose metabolism. Owing rophy. dominant Alz­hei­mer’s disease. although this malities in the CSF.27 in sporadic Alz­hei­mer’s disease or autosomal fibrillar amyloidosis. there were some differences. as measured by PET with the mer’s disease and the much more common use of Pittsburgh compound B (PIB).30 Furthermore. Aβ Deposition in Autosomal Dominant Alzheimer’s Disease Years suggest the possibility of a common pathophys- before Expected Clinical Symptoms. However. because familial Alz­hei­mer’s nant Alz­hei­mer’s disease centers.29 With the establishment of trend was predicted in autosomal dominant Alz­ DIAN.

3. velopment and implementation of diagnostic and Treatment and prevention trials can incorporate predictive tests and the design of prevention tri. et 7. Green NS. and Martha Storandt vide better clinical outcomes than the treatment for input and revisions to an earlier draft of the manuscript.jsp 15. (http://clinicaltrials. disease: a review and proposal for the amyloidoses. Ardilla A. 12.).M.M.ema disease in a large kindred with an E280A 5. London ( on July 11. Alz­hei­mer’s dementia will eventually develop in Disclosure forms provided by the authors are available with persons with positive scans for amyloid deposi. ClinicalTrials. X. Joondalup.S. Edith Cowan References 1. E. delay cognitive and clinical impairment may ul- ic. Adamski-Werner SL.S. Braak E.461:916-22.B. Tangles and Science 2002.europa. Multimodal techniques for diagnosis and markers of the Alzheimer’s pathological pothesis of Alzheimer’s disease: progress prognosis of Alzheimer’s disease. al. the Departments of Psychiatry (W.277: to moderate Alzheimer’s disease (ApoE4 medicines/human/medicines/002294/ — all in Australia. University College London Institute of Neurology. University of New South Wales. Holtzman DM. ClinicalTrials. Morris JC. These findings suggest that the targeting of We thank the participants and their families for their altruism.) and Neurology (W. Acc Chem Res 2005.M.X.B. Pittsburgh. PSEN1 and clinical” Alzheimer’s disease.. and Harvard Medical School (R. 2. Schafer K.B. Perrin RJ. Melbourne.F. and Neuroscience Research Australia and the School of Medical Sciences. et al.M.). age categories. Hardy J. the Taub Institute for Research on Alzheimer’s Disease and the Aging Brain and the Department of Neurology. National estimates of the prevalence of et al.18: sation): Vyndaqel. tion. De Strooper B. the Mental Health Research Institute. Brigham and Women’s Hospital.. La Jolla (P. J. T.L. Butler Hospital.J.). Los Angeles. Los Angeles (J. Hebert L. 8.S.F. St. Summary of opinion (initial authori. 2012. Aisen PS.C. the DIAN research and support staff at each of the participating Aβ earlier in the course of the disease may pro. Neurobiol Aging 1997. Fagan AM. Native state kinetic stabilization as a Clinical dementia rating training and re- n engl j med  nejm.E. Louis. et al. Columbia University College of Physicians and Surgeons. Merlini G.C. Warren Alpert Medical School of Brown University. Price JL. Sydney (P. Brookmeyer R.. Nature &mid=WC0b01ac058001d127&murl=men APOE and other loci affect age-at-onset in NCT00574132).).G.33 heimer’s disease. Selkoe DJ. Hypothetical model of dynamic Copyright © 2012 Massachusetts Medical Society. Indianapolis (B.K. the Department of Pathology and Laboratory Medicine. New York (R.M.)..J. M.)..A. 2009 (http://clinicaltrials Y. Dominantly Inherited Alzheimer’s Disease mer’s disease remains to be determined for spo.9:119-28.B.L. strategy to ameliorate protein misfolding al. the full text of this article at NEJM. us/medicines/medicines Alzheimer’s disease families with PS2 6. this finding suggests that Supported by the National Institute on Aging and a private nonprofit foundation (see disclosure at NEJM. Am J Med Genet B Neuropsy- of Alzheimer’s disease (EXPEDITION).R. JAMA 1997. Yu X. 2012. sites for their contributions to this study. of mild to moderate dementia after substantial radic Alz­hei­mer’s disease. Daw EW. of Alzheimer-related lesions in different 10.). Cruchaga C. Jagust WJ.N. and problems on the road to therapeutics. K. Psychiatry (A. RI (S. Massachusetts General Hospital. Autosomal-dominant Alzheimer’s diseases: a focus on the transthyretin Alzheimer’s disease in the United Jack CR Jr. and Pathology and Immunology (N. anisms of amyloidosis.] et al. the Department of Neurosciences.. these pathophysiological changes to gauge the als. Chakraverty S. The amyloid hy..M. Indiana University School of Medicine.M.. Alz­ 21. A.297:353-6. JW.C.M.jsp?curl=pages/ presenilin-1 2002. All rights reserved. prevention of Alzheimer’s disease. et hei­mers Res Ther 2011. 2009.M. A. Alzheimer’s disease families. neuronal and synaptic loss has occurred. Morris JC. VIC (C. 11. our findings indicate that the of pathophysiological changes associated with Alz­hei­mer’s disease process begins more than Alz­hei­mer’s disease has implications for the de. the Department of Neurology. 14. University of California.R. Washington Uni- versity School of Medicine. Johnson SM. [Erratum.K. A.O.).7(2):e31039.7:61-73. and the Department of Neurol- ogy and the Memory and Aging Program. mutation. whereas no noncarriers had positive scans for timately test the amyloid hypothesis. Sekijima chiatr Genet 2005. 2007 smops/Positive/human_smop_000246. Secondary deposition will develop and be detectable in all prevention trials that are designed to prevent or persons with a mutation while still asymptomat. Wijsman EM. University of 9 The New England Journal of Medicine Downloaded from nejm.). . Bapineuzumab in patients with mild . Ann Neurol 9. .X. the Centre of Excellence for Alzheimer’s Disease Research and Care. 20 years before the clinical onset of dementia.45:358-68.) — all in Boston.A. 2011 (http://www.S. the Center for Alzheimer Research and Treatment and the Department of Neurology.E.132B:14-20. Bellotti V. Lopera F. plaques in nondemented aging and “pre.). WA (R. David Geffen School of Medicine. Ernesto C.N.. Knopman DS. D. Medicines Agency. et al.297:2209.G. Wiseman RL.31 For example. Effect of LY2062430 on the progression . Evans DA.M. Science 13. Molecular mech.38:911- Alzheimers Dement 2011. D. our data suggest that amyloid likelihood of future clinical success. PLoS One 4.32 The definition of the timing and magnitude In summary. Frequency of stages 349:583-96.S.M.jsp&jsenabled=true). University of Pittsburgh School of Medicine. Lancet Neurol 2010. V.). University of California. N Engl J Med 2003. PSEN2 increase risk for AD in late-onset 1999.).. just as the amyloid deposition.R.3:1. Mayo K. Braak Rare variants in APP. Biosta- tistics (C.). Radiology (T.S. If autosomal dominant Alz­ cholesterol hypothesis of heart disease was tested hei­mer’s disease is similar to late-onset Alz­ three decades ago. For personal use only. non-carrier). San Diego.. London: European Clinical features of early-onset Alzheimer 351-7. Appendix The authors’ affiliations are as follows: the Departments of Neurology (R.H. Bateman RJ.). No other uses without permission. Providence. Martínez A. Kelly 16.

ease. Ann Neurol 2004. Folstein SE. Wechsler D. Neurol 1997. Pratico D. Imaging brain amyloid in Alzheimer’s 1999. Tsou SK.5:828-34. Brain ry Scale–Revised: manual.71:85-92. Endo A. 29. tum of presenilin-1 mutation carriers percholesterolemia. McHugh PR. Metabolic with preclinical familial Alzheimer dis- hei­mers Disease Cooperative Study expe. Sheline YI. 2011 (http://www.55:306-19. Mutations). 32. Arch Neurol 2007. et 19. Larossa GN. Neurology 2006. ease. et al. Bartlett JW.27:6174-84. Neurology 1997. Mintun of Alzheimer’s disease: recommendations initiative. Price JC. Cerebro. diagnostic guidelines for Alzheimer’s dis- Dementia Mutation Database home page. Mueller SG. Sperling RA. Price JL. 27. Arch Neurol 2001. Minoshima S. Neurology 1997. Fagan AM. et al. TX: Psychological Corporation. Toward defining the preclinical stages The Alzheimer’s disease neuroimaging 26. Gómez-Isla T. Holtzman DM. cal Alzheimer disease. For personal use only. 25. 58:1395-402. Copyright © 2012 Massachusetts Medical Society. J Neurosci 35:259-66. 33. Berent S.42:85-94. Ko AI. et Tracking atrophy progression in familial the entorhinal cortex and CA1 in preclini- al. Neuron number in 21. mal aging and very mild Alzheimer’s dis. Engler H. McKeel DW. Roe CM. Barnes J.67:446-52. WMS-R: Wechsler Memo. Xu YC. Neurology 2008. Morris JC. Younkin SG. et presenilin 1 and presenilin 2 mutations for grading the cognitive state of patients al. spinal fluid tau/beta-amyloid(42) ratio as Alzheimer’s Association workgroups on 20. The New England Journal of Medicine Downloaded from nejm. All rights reserved. demented older adults. et al. No other uses without permission. Neuroimaging Clin N Am 2005. and cognitive variables and descriptive Lancet Neurol 2006. 2007.48:1508-10. Atherosclerosis 1980. Medial temporal atrophy on MRI in nor. data from Alzheimer Disease Centers. Aisen PS. mara MJ. Alzheimers Dement 2011. Thal LJ. Beckett LA. Dominantly Inherited Alzheimer’s Disease liability in multicenter studies: the Alz­ Frey KA. 1987.49:786-94. Alzheimer Disease & Frontotemporal a prediction of cognitive decline in non. Ridha BH. Alzheimer’s disease brain: evidence for 18. Klunk WE. Sudo H. rience. al. peutic effects of ML-236B in primary hy- 22. Ringman JM. Ann 30. 2012. Nordberg A. Thera- hei­mers Dis Assoc Disord 2006. Petersen RC. The impact of different “Mini-mental state:” a practical method 24. Weiner MW. Foster NL. Chui HC. et other phenotype-modifying MA. Klunk WE. Biochemical markers in persons 10 n engl j med  nejm. The Uniform Data Set (UDS): clinical Alzheimer’s disease: a serial MRI study.molgen. Jack CR Jr. . 23. tex in very early Alzheimer’s disease. Morris JC. disease with Pittsburgh Compound-B. from two unrelated pedigrees. Folstein MF. disease. Growdon WB.20:210-6. Yamamoto A. San Antonio. Mathis CA. Mintun MA. Wade MJ. [11C]PIB in a nondemented population: on amyloid deposition. from the National Institute on Aging– 15:869-77. Kuhl DE. Weintraub S. al. potential antecedent marker of Alzheimer changes and neuronal loss in the familial 12:189-98. et al. Copyright © 2012 Massachusetts Medical Society. McNa- 17. Giordani B. reduction in the posterior cingulate on July 11.7:280-92. Xiong C. neurofibrillary for the clinician. et al. et al. J Psychiatr Res 1975.122:1709-19. Morris JC. Amyloid deposition begins in the stria. Alz­ 28.