Research only
All of
ALZ.ORG Research Center Alzheimer's Association International Conference ISTAART Journal Grants TrialMatch Press Donate Contact

AAIC Press Release

<< See all AAIC Press Releases

Alzheimer's Association®  |  Media line: 312.335.4078  |  E-mail:  |
AAIC 2013 press room, July 13-18: 617.954.3414

Updated Model of Alzheimer's Disease Biomarker Sequence Supported and Refined with Clinical Evidence

Brain imaging and biomarker studies presented at AAIC 2013 show an increasingly detailed picture of Alzheimer's progression, starting with amyloid abnormalities and leading to memory and thinking symptoms.

BOSTON, July 16, 2013 – An updated model of the sequence of changes in the brains of people with Alzheimer's disease was reported today at the Alzheimer's Association International Conference® 2013 (AAIC® 2013) in Boston.

In the hypothetical model of Alzheimer's progression over time, now refined with several years of clinical evidence, biological markers of changes in beta-amyloid protein become abnormal first, then biomarkers of the degeneration and death of brain cells, followed by dementia symptoms.

Two other studies presented today support the updated Alzheimer's model by demonstrating that abnormal changes in beta-amyloid protein precede dementia symptoms by years or even decades and then begin to plateau by the time cognitive changes are observable.

"Having a clearer and more accurate portrayal of the sequence of biomarker changes in Alzheimer's disease is extremely valuable," said Maria Carrillo, Ph.D., Alzheimer's Association vice president of medical and scientific relations. "For example, it will enable us to better design treatment studies — enhancing the selection of study participants, targeting the proper time during the disease to give the treatment and measuring treatment effects."

"These three reports underscore the importance of Alzheimer's disease biomarker studies, and also begin to show us when those biomarkers might be potential targets for therapy," said Carrillo. "It is when biomarkers are at their most dynamic and changeable that they can best be measured and perhaps modified through treatment."

Research confirms hypothesis that amyloid abnormalities precede other Alzheimer's indicators; Updated biomarker model is presented

In 2010, Clifford Jack, Jr., M.D., of the Mayo Clinic College of Medicine and colleagues proposed a hypothetical model of the sequence of change in Alzheimer's disease biomarkers and related them to each other and to the emergence and progression of outward dementia symptoms. In that model, beta-amyloid markers become abnormal first, then biomarkers of neurodegeneration, followed by clinical symptoms.

At AAIC 2013, Jack and colleagues report that they have updated the model based on research evidence that has accumulated since the original publication that supports the original model's general outlines, yet also challenges some its assumptions. In the updated model:

"One of the most important criticisms of our original model was that it failed to account for abnormalities in tau protein in the brain that are a nearly universal feature of aging," said Jack. "Our model now distinguishes between biomarkers of Alzheimer's disease pathology and biomarkers of normal aging."

According to Jack, the updated change in biomarker ordering is as follows:

Amyloid abnormalities begin many years before clinical symptoms appear

The deposition of beta-amyloid in the brains of people with Alzheimer's begins years or even decades before cognitive changes become apparent, according to evidence from a clinical study reported at AAIC 2013 by Victor Villemagne, M.D., of the Florey Institute of Neuroscience and Mental Health at the University of Melbourne and colleagues.

Two-hundred-six (206) participants — 151 healthy controls, 36 with mild cognitive impairment (MCI) and 19 with Alzheimer's disease — in the Australian Imaging Biomarkers and Lifestyle (AIBL) Study of Ageing were evaluated at the beginning of the study and every 18 months for an average of 4.1 years.

Using data from neuropsychological examinations, MRI and PET scans, the scientists projected that, on average, it takes more than 10 years from being free of beta-amyloid to reach a level that is considered abnormal. From there they projected that it takes, on average, about 19 years to reach the levels usually found in people with Alzheimer's disease. At that point in the disease, beta amyloid deposition starts to slow, trending towards a plateau. Changes in the brain's memory center — the hippocampus — appear about five years before onset of full-blown dementia, and memory changes about three years before.

Significantly, the researchers found that amyloid deposition rates among the study participants were very similar to projected rates among patients with dominantly inherited Alzheimer's disease, a rare genetic form of the disease which usually strikes before age 60. "This suggests that similar basic mechanisms are at play, irrespective of dissimilar pathogenic pathways," Villemagne said.

"Beta-amyloid deposition in Alzheimer's disease is a slow and protracted process, likely to extend for more than two decades. The results of this and other longitudinal studies will allow us to better understand how changes in beta-amyloid in the brain relate to changes in cognition and other Alzheimer's biomarkers. This will facilitate the design and timing of therapeutic interventions aimed at modifying the course of this illness," Villemagne concluded.

Study confirms that abnormal beta-amyloid begins cascade of Alzheimer's pathology

According to new NIA/Alzheimer's Association diagnostic guidelines, Alzheimer's disease can be subdivided into three stages — asymptomatic Alzheimer's, MCI due to Alzheimer's and dementia due to Alzheimer's. The rates of neuropathological, cognitive and functional decline in each of these stages is not fully understood, and this limits clinical trial design, in particular in the predementia stages.

Daniela Bertens of VU University Medical Center, Amsterdam, Netherlands, and colleagues investigated 284 subjects with abnormal beta-amyloid from the Alzheimer's Disease Neuroimaging Initiative, using amyloid levels in cerebrospinal fluid (CSF) as a diagnostic marker and followed them for four years. The study included 44 subjects with asymptomatic Alzheimer's, 148 with Alzheimer's-related MCI and 92 with Alzheimer's dementia, plus a control group of cognitively normal subjects with normal amyloid levels. Participants were measured at regular intervals for biomarkers of amyloid, neuronal dysfunction and brain atrophy using CSF tests and brain imaging, as well as for cognition, using standard clinical scales and tests.

The researchers found that the rate of decline on beta-amyloid and tau decreased from the preclinical stage to the dementia stage. On the other hand, the rate of decline on FDG-PET, atrophy markers and cognition increased as Alzheimer's disease progressed.

"The observed rates of decline are in line with the amyloid cascade hypothesis," said Bertens. "Our findings give insight into the progression of Alzheimer's disease and therefore might be useful in the design of treatment studies. Some markers will be useful for the selection of study subjects, while other markers can be used to measure disease progression and treatment effects."

About AAIC

The Alzheimer's Association International Conference (AAIC) is the world's largest conference of its kind, bringing together researchers from around the world to report and discuss groundbreaking research and information on the cause, diagnosis, treatment and prevention of Alzheimer's disease and related disorders. As a part of the Alzheimer's Association's research program, AAIC serves as a catalyst for generating new knowledge about dementia and fostering a vital, collegial research community.

About the Alzheimer's Association

The Alzheimer's Association is the world's leading voluntary health organization in Alzheimer care, support and research. Our mission is to eliminate Alzheimer's disease through the advancement of research; to provide and enhance care and support for all affected; and to reduce the risk of dementia through the promotion of brain health. Our vision is a world without Alzheimer's. Visit or call 800.272.3900.

# # #

Proposal ID: 37450
Topic: Diagnosis and Prognosis
Subtopic: Biomarkers; Multi-modal comparisons
Presentation: Oral session, Tuesday, July 16, 2013, 2:15-3:45 p.m. ET

Update on hypothetical model of Alzheimer's disease biomarkers

Presenting author: Clifford Jack, Clifford Jack1, David Knopman1, William Jagust2, Ronald Petersen3, Michael Weiner4, Paul Aisen5, Leslie Shaw6, Prashanthi Vemuri1, Heather Wiste1, Stephen Weigand1, Timothy Lesnick1, Vernon Pankratz1, Michael Donohue7, John Trojanowski6 1Mayo Clinic, Rochester, Minnesota, United States; 2University of California, Berkeley, Berkeley, California, United States; 3Mayo Clinic Rochester, Rochester, Minnesota, United States; 4San Francisco Veterans Affairs Medical Centers; VA Medical Center and UCSF, San Francisco, California, United States; 5UCSD, La Jolla, California, United States; 6University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, United States; 7University of California, San Diego, La Jolla, California, United States.

Background: In 2010, the authors published a hypothetical model of the major biomarkers of Alzheimer's disease (AD) in which we described the temporal evolution of AD biomarkers in relation to each other and to the onset and progression of clinical symptoms.

Methods: We reviewed recent studies that empirically address testable features of our model and proposed modifications.

Results:Evidence that has accumulated since our model was first published supports its general temporal ordering framework where amyloid biomarkers become abnormal first then biomarkers of neurodegeneration, followed by clinical symptoms. Evidence has also appeared, however, that challenges some of our original assumptions which has allowed us to modify our model. In our revised model:
1) The horizontal axis is expressed as time, not clinical disease stage.
2) Cognitive outcome is expressed as a range of possible values, reflecting the fact that individual subjects respond to AD pathophysiology uniquely.
3) The revised model includes modifications of the specific ordering of some biomarkers.
4) All biomarkers are still configured as sigmoids, but the shapes of the sigmoid curves are no longer identical.
5) The biomarker curves are drawn closer together indicating less distinct temporal separation.

One of the most important criticisms of our original AD biomarker model was that it failed to account for the fact that medial temporal limbic tau pathology is a nearly universal feature of aging and typically appears at a younger age than Aß amyloid plaques. Thus the model was modified to incorporate different concepts: biomarkers of AD pathophysiology vs. histopathology that can be measured at autopsy.

Conclusions: We made several modifications to our original AD biomarker model based on data that has accumulated since it was first published. The model has also been modified to recognize that the two major proteinopathies underlying AD biomarker changes, Aß and tau, may be initiated independently in sporadic AD where an incident Aßopathy may accelerate antecedent medial temporal tauopathy.

Proposal ID: 38905
Topic: Diagnosis and Prognosis
Subtopic: Biomarkers; Longitudinal change over time
Presentation: Oral session, Tuesday, July 16, 2013, 2:15-3:45 p.m. ET

The natural history of the dynamics of Aß deposition, neurodegeneration and cognitive decline in sporadic Alzheimer's disease

Presenting author: Victor Villemagne, Victor L Villemagne, Austin Health; Samantha Burnham, CSIRO; Pierrick Bourgeat, CSIRO; Belinda Brown, Edith Cowan University; Kathryn A Ellis, St Georges Hospital; Olivier Salvado, CSIRO; Cassandra Szoeke, National Ageing Research Institute Inc. (NARI); Lance Macaulay, CSIRO; Ralph Martins, Edith Cowan University; Paul Maruff, CogState Ltd.; David Ames, National Ageing Research Institute Inc. (NARI); Christopher C Rowe, Austin Health; Colin L Masters, University of Melbourne

Background: We used longitudinal data to calculate the rates of Aß deposition, cerebral atrophy and cognitive decline in healthy controls (HC), mild cognitive impairment (MCI) and AD patients.

Methods: Two hundred participants (145 HC; 36 MCI; and 19 AD) were evaluated at enrolment and every 18 months for a mean follow up of 3.8 (CI 3.6-3.9) years. On each visit, participants underwent neuropsychological examination, MRI and a PiB PET scan. Aß burden was expressed as PiB standardized uptake value ratio (SUVR) using the cerebellar cortex as reference region. A SUVR of 1.5 was used to discriminate high from low Aß burdens. Rates of change for Aß deposition, MRI volumetrics and cognition were derived from the slope of the regression plots over three to five years and used in the analysis. Those participants with a positive rate of Aß deposition were used to calculate the trajectory of each parameter over time.

Results: At baseline significantly higher Aß burdens were observed in AD (2.3±0.4) and MCI (1.9±0.6) compared to HC (1.4±0.4). At follow up 163 (82 percent) of the 200 participants showed positive rates of Aß accumulation. Estimates of Aß deposition over time show it takes 19 (CI 17-23) years in an almost linear fashion — with increases averaging 0.043 (CI 0.037-0.049) SUVR/yr — to go from the threshold of PiB positivity (1.5 SUVR) to the levels observed in AD. Conversely, it takes 12 (CI 10-15 years) years to go from the levels observed in HC with low Aß (1.2±0.1 SUVR) to the threshold of PiB positivity. As AD progresses, the rate of Aß deposition slows, leaning towards a plateau. Our projections suggest a prolonged preclinical phase of AD where Aß deposition, hippocampal atrophy, and memory impairment become sequentially abnormal at 17 (CI 15-20), 4 (CI 4-5), and 3 (CI 3-5) years respectively, before the onset of dementia

Conclusions: Aß deposition is a slow and protracted process, likely to extend for more than two decades. Predicting the rate of evolution of preclinical changes and the onset of clinical phase of AD are essential for the design and timing of therapeutic interventions aimed at modifying the course of this illness.

Proposal ID: 39379
Topic: Diagnosis and Prognosis
Subtopic: Longitudinal change over time
Presentation: Oral session, Tuesday, July 16, 2013, 2:15-3:45 p.m. ET

Temporal evolution of biomarkers and cognitive markers from preclinical to clinical Alzheimer's Disease: an Alzheimer's Disease Neuroimaging Initiative (ADNI) study

Presenting author: Daniela Bertens, Daniela Bertens*1, Philip Scheltens2, Pieter Jelle Visser3 1VU Medical Center Amsterdam, Amsterdam, Netherlands; 2VU University Medical Center, Amsterdam, Netherlands; 3Alzheimer Centre VUMC, Maastricht, New York, Netherlands.

Background: According to new NIA-AA/IWG research criteria Alzheimer's disease (AD) can be subdivided in an asymptomatic stage, MCI stage and dementia stage. The rate of neuropathological, cognitive and functional decline in each of these stages is not fully understood. This limits trial design, in particular in the predementia stage. Our aim was to describe biomarker and cognitive changes from the asymptomatic stage of AD to dementia.

Methods: We selected cognitively normal, MCI and demented subjects with AD pathology at baseline (CSF abeta1-42<193pg/l). Studied markers were abeta1-42, tau, FDG uptake on PET, hippocampal, whole brain and ventricular volume on MRI, CDR-sum of boxes (CDR-sob), MMSE, ADAS-Cog, and composite scores of executive function (CSEF) and memory (CSM). To compare markers, raw scores were converted into z-scores relative to cognitively normal subjects with normal CSF abeta (control group). Group comparison of markers at baseline and slope of decline were estimated with mixed models. We tested whether scores differed significantly (p<0.05) at baseline and if slopes were different from 0 and differed between AD stages.

Results: We included 44 asymptomatic, 148 MCI and 92 demented subjects. Compared to the control group, AD-asymptomatic subjects had abnormal tau, ADAS Cog and CSEF scores. In the AD-MCI stage all markers were abnormal, except whole brain volume. In the AD-dementia stage all markers were abnormal. CSF abeta1-42 showed a trend towards decline in AD-asymptomatic subjects (annual z-score decline of -0.09). Tau levels increased over time in AD-asymptomatic (z-score 0.12) and AD-MCI subjects (z-score 0.07). Imaging markers declined in all stages and rate of decline increased with each stage with the largest effect on FDG-PET (decline of -0.16 for AD-asymptomatic -0.35 for AD-MCI and -0.7 for AD-dementia). Of the clinical measures only CDR-SOB declined in AD-asymptomatic. In AD-MCI and AD-dementia all cognitive scores declined and the rate of decline was higher in AD-dementia than in AD-MCI.

Conclusions: The observed rates of decline are in line with the amyloid cascade hypothesis. CSF abeta1-42 and tau decline first, abeta1-42 decline levels off in AD-MCI and tau in AD-dementia. Imaging markers decline before cognitive markers and for both markers rate of decline increases with clinical stage.

# # #


<< See all AAIC Press Releases

AAIC 2013 program

View the 2013 schedule and conference program.

Registered attendees have exclusive access to the 2013 AAIC mobile app with full speaker and session listings as well as abstract content. Attendees can email for download instructions.

Make history at AAIC 2013 • Boston, MA • July 13-18, 2013