Transition rates between amyloid and neurodegeneration biomarker states and to dementia: A population-based, longitudinal cohort study

Clifford R Jr. Jack, Terry M Therneau, Heather J. Wiste, Stephen D. Weigand, David S Knopman, Val Lowe, Michelle M Mielke, Prashanthi D Vemuri, Rosebud O Roberts, Mary Margaret Machulda, Matthew L. Senjem, Jeffrey L. Gunter, Walter A Rocca, Ronald Carl Petersen

Research output: Contribution to journalArticle

60 Citations (Scopus)

Abstract

Background: In a 2014 cross-sectional analysis, we showed that amyloid and neurodegeneration biomarker states in participants with no clinical impairment varied greatly with age, suggesting dynamic within-person processes. In this longitudinal study, we aimed to estimate rates of transition from a less to a more abnormal biomarker state by age in individuals without dementia, as well as to assess rates of transition to dementia from an abnormal state. Methods: Participants from the Mayo Clinic Study of Aging (Olmsted County, MN, USA) without dementia at baseline were included in this study, a subset of whom agreed to multimodality imaging. Amyloid PET (with 11C-Pittsburgh compound B) was used to classify individuals as amyloid positive (A+) or negative (A-). 18F-fluorodeoxyglucose (18F-FDG)-PET and MRI were used to classify individuals as neurodegeneration positive (N+) or negative (N-). We used all observations, including those from participants who did not have imaging results, to construct a multistate Markov model to estimate four different age-specific biomarker state transition rates: A-N- to A+N-; A-N- to A-N+ (suspected non-Alzheimer's pathology); A+N- to A+N+; and A-N+ to A+N+. We also estimated two age-specific rates to dementia: A+N+ to dementia and A-N+ to dementia. Using these state-to-state transition rates, we estimated biomarker state frequencies by age. Findings: At baseline (between Nov 29, 2004, to March 7, 2015), 4049 participants did not have dementia (3512 [87%] were clinically normal and 537 [13%] had mild cognitive impairment). 1541 individuals underwent imaging between March 28, 2006, to April 30, 2015. Transition rates were low at age 50 years and, with one exception, exponentially increased with age. At age 85 years compared with age 65 years, the rate was nearly 11-times higher (17·2 vs 1·6 per 100 person-years) for the transition from A-N- to A-N+, three-times higher (20·8 vs 6·1) for A+N- to A+N+, and five-times higher (13·2 vs 2·6) for A-N+ to A+N+. The rate of transition was also increased at age 85 years compared with age 65 years for A+N+ to dementia (7·0 vs 0·8) and for A-N+ to dementia (1·7 vs 0·6). The one exception to an exponential increase with age was the transition rate from A-N- to A+N-, which increased from 4·0 transitions per 100 person-years at age 65 years to 6·9 transitions per 100 person-years at age 75 and then plateaued beyond that age. Estimated biomarker frequencies by age from the multistate model were similar to cross-sectional biomarker frequencies. Interpretation: Our transition rates suggest that brain ageing is a nearly inevitable acceleration toward worse biomarker and clinical states. The one exception is the transition to amyloidosis without neurodegeneration, which is most dynamic from age 60 years to 70 years and then plateaus beyond that age. We found that simple transition rates can explain complex, highly interdependent biomarker state frequencies in our population. Funding: National Institute on Aging, Alexander Family Professorship of Alzheimer's Disease Research, the GHR Foundation.

Original languageEnglish (US)
JournalThe Lancet Neurology
DOIs
StateAccepted/In press - 2015

Fingerprint

Amyloid
Longitudinal Studies
Dementia
Cohort Studies
Biomarkers
Population
National Institute on Aging (U.S.)
Fluorodeoxyglucose F18
Amyloidosis
Alzheimer Disease
Cross-Sectional Studies
Pathology
Brain
Research

ASJC Scopus subject areas

  • Clinical Neurology

Cite this

@article{db59a8a7d8ca46e594c1bcc6fe5c942f,
title = "Transition rates between amyloid and neurodegeneration biomarker states and to dementia: A population-based, longitudinal cohort study",
abstract = "Background: In a 2014 cross-sectional analysis, we showed that amyloid and neurodegeneration biomarker states in participants with no clinical impairment varied greatly with age, suggesting dynamic within-person processes. In this longitudinal study, we aimed to estimate rates of transition from a less to a more abnormal biomarker state by age in individuals without dementia, as well as to assess rates of transition to dementia from an abnormal state. Methods: Participants from the Mayo Clinic Study of Aging (Olmsted County, MN, USA) without dementia at baseline were included in this study, a subset of whom agreed to multimodality imaging. Amyloid PET (with 11C-Pittsburgh compound B) was used to classify individuals as amyloid positive (A+) or negative (A-). 18F-fluorodeoxyglucose (18F-FDG)-PET and MRI were used to classify individuals as neurodegeneration positive (N+) or negative (N-). We used all observations, including those from participants who did not have imaging results, to construct a multistate Markov model to estimate four different age-specific biomarker state transition rates: A-N- to A+N-; A-N- to A-N+ (suspected non-Alzheimer's pathology); A+N- to A+N+; and A-N+ to A+N+. We also estimated two age-specific rates to dementia: A+N+ to dementia and A-N+ to dementia. Using these state-to-state transition rates, we estimated biomarker state frequencies by age. Findings: At baseline (between Nov 29, 2004, to March 7, 2015), 4049 participants did not have dementia (3512 [87{\%}] were clinically normal and 537 [13{\%}] had mild cognitive impairment). 1541 individuals underwent imaging between March 28, 2006, to April 30, 2015. Transition rates were low at age 50 years and, with one exception, exponentially increased with age. At age 85 years compared with age 65 years, the rate was nearly 11-times higher (17·2 vs 1·6 per 100 person-years) for the transition from A-N- to A-N+, three-times higher (20·8 vs 6·1) for A+N- to A+N+, and five-times higher (13·2 vs 2·6) for A-N+ to A+N+. The rate of transition was also increased at age 85 years compared with age 65 years for A+N+ to dementia (7·0 vs 0·8) and for A-N+ to dementia (1·7 vs 0·6). The one exception to an exponential increase with age was the transition rate from A-N- to A+N-, which increased from 4·0 transitions per 100 person-years at age 65 years to 6·9 transitions per 100 person-years at age 75 and then plateaued beyond that age. Estimated biomarker frequencies by age from the multistate model were similar to cross-sectional biomarker frequencies. Interpretation: Our transition rates suggest that brain ageing is a nearly inevitable acceleration toward worse biomarker and clinical states. The one exception is the transition to amyloidosis without neurodegeneration, which is most dynamic from age 60 years to 70 years and then plateaus beyond that age. We found that simple transition rates can explain complex, highly interdependent biomarker state frequencies in our population. Funding: National Institute on Aging, Alexander Family Professorship of Alzheimer's Disease Research, the GHR Foundation.",
author = "Jack, {Clifford R Jr.} and Therneau, {Terry M} and Wiste, {Heather J.} and Weigand, {Stephen D.} and Knopman, {David S} and Val Lowe and Mielke, {Michelle M} and Vemuri, {Prashanthi D} and Roberts, {Rosebud O} and Machulda, {Mary Margaret} and Senjem, {Matthew L.} and Gunter, {Jeffrey L.} and Rocca, {Walter A} and Petersen, {Ronald Carl}",
year = "2015",
doi = "10.1016/S1474-4422(15)00323-3",
language = "English (US)",
journal = "The Lancet Neurology",
issn = "1474-4422",
publisher = "Lancet Publishing Group",

}

TY - JOUR

T1 - Transition rates between amyloid and neurodegeneration biomarker states and to dementia

T2 - A population-based, longitudinal cohort study

AU - Jack, Clifford R Jr.

AU - Therneau, Terry M

AU - Wiste, Heather J.

AU - Weigand, Stephen D.

AU - Knopman, David S

AU - Lowe, Val

AU - Mielke, Michelle M

AU - Vemuri, Prashanthi D

AU - Roberts, Rosebud O

AU - Machulda, Mary Margaret

AU - Senjem, Matthew L.

AU - Gunter, Jeffrey L.

AU - Rocca, Walter A

AU - Petersen, Ronald Carl

PY - 2015

Y1 - 2015

N2 - Background: In a 2014 cross-sectional analysis, we showed that amyloid and neurodegeneration biomarker states in participants with no clinical impairment varied greatly with age, suggesting dynamic within-person processes. In this longitudinal study, we aimed to estimate rates of transition from a less to a more abnormal biomarker state by age in individuals without dementia, as well as to assess rates of transition to dementia from an abnormal state. Methods: Participants from the Mayo Clinic Study of Aging (Olmsted County, MN, USA) without dementia at baseline were included in this study, a subset of whom agreed to multimodality imaging. Amyloid PET (with 11C-Pittsburgh compound B) was used to classify individuals as amyloid positive (A+) or negative (A-). 18F-fluorodeoxyglucose (18F-FDG)-PET and MRI were used to classify individuals as neurodegeneration positive (N+) or negative (N-). We used all observations, including those from participants who did not have imaging results, to construct a multistate Markov model to estimate four different age-specific biomarker state transition rates: A-N- to A+N-; A-N- to A-N+ (suspected non-Alzheimer's pathology); A+N- to A+N+; and A-N+ to A+N+. We also estimated two age-specific rates to dementia: A+N+ to dementia and A-N+ to dementia. Using these state-to-state transition rates, we estimated biomarker state frequencies by age. Findings: At baseline (between Nov 29, 2004, to March 7, 2015), 4049 participants did not have dementia (3512 [87%] were clinically normal and 537 [13%] had mild cognitive impairment). 1541 individuals underwent imaging between March 28, 2006, to April 30, 2015. Transition rates were low at age 50 years and, with one exception, exponentially increased with age. At age 85 years compared with age 65 years, the rate was nearly 11-times higher (17·2 vs 1·6 per 100 person-years) for the transition from A-N- to A-N+, three-times higher (20·8 vs 6·1) for A+N- to A+N+, and five-times higher (13·2 vs 2·6) for A-N+ to A+N+. The rate of transition was also increased at age 85 years compared with age 65 years for A+N+ to dementia (7·0 vs 0·8) and for A-N+ to dementia (1·7 vs 0·6). The one exception to an exponential increase with age was the transition rate from A-N- to A+N-, which increased from 4·0 transitions per 100 person-years at age 65 years to 6·9 transitions per 100 person-years at age 75 and then plateaued beyond that age. Estimated biomarker frequencies by age from the multistate model were similar to cross-sectional biomarker frequencies. Interpretation: Our transition rates suggest that brain ageing is a nearly inevitable acceleration toward worse biomarker and clinical states. The one exception is the transition to amyloidosis without neurodegeneration, which is most dynamic from age 60 years to 70 years and then plateaus beyond that age. We found that simple transition rates can explain complex, highly interdependent biomarker state frequencies in our population. Funding: National Institute on Aging, Alexander Family Professorship of Alzheimer's Disease Research, the GHR Foundation.

AB - Background: In a 2014 cross-sectional analysis, we showed that amyloid and neurodegeneration biomarker states in participants with no clinical impairment varied greatly with age, suggesting dynamic within-person processes. In this longitudinal study, we aimed to estimate rates of transition from a less to a more abnormal biomarker state by age in individuals without dementia, as well as to assess rates of transition to dementia from an abnormal state. Methods: Participants from the Mayo Clinic Study of Aging (Olmsted County, MN, USA) without dementia at baseline were included in this study, a subset of whom agreed to multimodality imaging. Amyloid PET (with 11C-Pittsburgh compound B) was used to classify individuals as amyloid positive (A+) or negative (A-). 18F-fluorodeoxyglucose (18F-FDG)-PET and MRI were used to classify individuals as neurodegeneration positive (N+) or negative (N-). We used all observations, including those from participants who did not have imaging results, to construct a multistate Markov model to estimate four different age-specific biomarker state transition rates: A-N- to A+N-; A-N- to A-N+ (suspected non-Alzheimer's pathology); A+N- to A+N+; and A-N+ to A+N+. We also estimated two age-specific rates to dementia: A+N+ to dementia and A-N+ to dementia. Using these state-to-state transition rates, we estimated biomarker state frequencies by age. Findings: At baseline (between Nov 29, 2004, to March 7, 2015), 4049 participants did not have dementia (3512 [87%] were clinically normal and 537 [13%] had mild cognitive impairment). 1541 individuals underwent imaging between March 28, 2006, to April 30, 2015. Transition rates were low at age 50 years and, with one exception, exponentially increased with age. At age 85 years compared with age 65 years, the rate was nearly 11-times higher (17·2 vs 1·6 per 100 person-years) for the transition from A-N- to A-N+, three-times higher (20·8 vs 6·1) for A+N- to A+N+, and five-times higher (13·2 vs 2·6) for A-N+ to A+N+. The rate of transition was also increased at age 85 years compared with age 65 years for A+N+ to dementia (7·0 vs 0·8) and for A-N+ to dementia (1·7 vs 0·6). The one exception to an exponential increase with age was the transition rate from A-N- to A+N-, which increased from 4·0 transitions per 100 person-years at age 65 years to 6·9 transitions per 100 person-years at age 75 and then plateaued beyond that age. Estimated biomarker frequencies by age from the multistate model were similar to cross-sectional biomarker frequencies. Interpretation: Our transition rates suggest that brain ageing is a nearly inevitable acceleration toward worse biomarker and clinical states. The one exception is the transition to amyloidosis without neurodegeneration, which is most dynamic from age 60 years to 70 years and then plateaus beyond that age. We found that simple transition rates can explain complex, highly interdependent biomarker state frequencies in our population. Funding: National Institute on Aging, Alexander Family Professorship of Alzheimer's Disease Research, the GHR Foundation.

UR - http://www.scopus.com/inward/record.url?scp=84947318181&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947318181&partnerID=8YFLogxK

U2 - 10.1016/S1474-4422(15)00323-3

DO - 10.1016/S1474-4422(15)00323-3

M3 - Article

C2 - 26597325

AN - SCOPUS:84947318181

JO - The Lancet Neurology

JF - The Lancet Neurology

SN - 1474-4422

ER -