Complementary value of molecular, phenotypic, and functional aging biomarkers in dementia prediction

doi:10.1007/s11357-024-01376-w

Complementary value of molecular, phenotypic, and functional aging biomarkers in dementia prediction

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

DNA methylation age (MA), brain age (BA), and frailty index (FI) are putative aging biomarkers linked to dementia risk. We investigated their relationship and combined potential for prediction of cognitive impairment and future dementia risk using the ADNI database. Of several MA algorithms, DunedinPACE and GrimAge2, associated with memory, were combined in a composite MA alongside BA and a data-driven FI in predictive analyses. Pairwise correlations between age- and sex-adjusted measures for MA (aMA), aBA, and aFI were low. FI outperformed BA and MA in all diagnostic tasks. A model including age, sex, and aFI achieved an area under the curve (AUC) of 0.94 for differentiating cognitively normal controls (CN) from dementia patients in a held-out test set. When combined with clinical biomarkers (apolipoprotein E ε4 allele count, memory, executive function), a model including aBA and aFI predicted 5-year dementia risk among MCI patients with an out-of-sample AUC of 0.88. In the prognostic model, BA and FI offered complementary value (both βs 0.50). The tested MAs did not improve predictions. Results were consistent across FI algorithms, with data-driven health deficit selection yielding the best performance. FI had a stronger adverse effect on prognosis in males, while BA’s impact was greater in females. Our findings highlight the complementary value of BA and FI in dementia prediction. The results support a multidimensional view of dementia, including an intertwined relationship between the biomarkers, sex, and prognosis. The tested MA’s limited contribution suggests caution in their use for individual risk assessment of dementia.

Original language	English
Pages (from-to)	2099-2118
Number of pages	20
Journal	GeroScience
Volume	47
Issue number	2
DOIs	https://doi.org/10.1007/s11357-024-01376-w
State	Published - Apr 2025

Bibliographical note

Publisher Copyright:
© The Author(s) 2024.

Funding

Open access funding provided by University of Oslo (incl Oslo University Hospital). Data collection and sharing for ADNI is funded by the National Institute on Aging (National Institutes of Health Grant U19 AG024904). The grantee organization is the Northern California Institute for Research and Education. In the past, ADNI has also received funding from the National Institute of Biomedical Imaging and Bioengineering, the Canadian Institutes of Health Research, and private sector contributions through the Foundation for the National Institutes of Health (FNIH) including generous contributions from the following: AbbVie, Alzheimer\u2019s Association; Alzheimer\u2019s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research &Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics.

Funders	Funder number
Northern California Institute for Research and Education
National Institute of Biomedical Imaging and Bioengineering
Foundation for the National Institutes of Health
Universitetet i Oslo
DoD Alzheimer's Disease Neuroimaging Initiative
Canadian Institutes of Health Research
National Institute on Aging
National Institutes of Health (NIH)	U19 AG024904
National Institutes of Health (NIH)

Keywords

Biological age
Brain age
Deep learning
Dementia
Frailty index
Machine learning
Methylation age

ASJC Scopus subject areas

Aging
Veterinary (miscellaneous)
Complementary and alternative medicine
Geriatrics and Gerontology
Cardiology and Cardiovascular Medicine

Access to Document

10.1007/s11357-024-01376-w

Cite this

@article{7a563218a9ef40ad81d1ce58b03bf995,

title = "Complementary value of molecular, phenotypic, and functional aging biomarkers in dementia prediction",

abstract = "DNA methylation age (MA), brain age (BA), and frailty index (FI) are putative aging biomarkers linked to dementia risk. We investigated their relationship and combined potential for prediction of cognitive impairment and future dementia risk using the ADNI database. Of several MA algorithms, DunedinPACE and GrimAge2, associated with memory, were combined in a composite MA alongside BA and a data-driven FI in predictive analyses. Pairwise correlations between age- and sex-adjusted measures for MA (aMA), aBA, and aFI were low. FI outperformed BA and MA in all diagnostic tasks. A model including age, sex, and aFI achieved an area under the curve (AUC) of 0.94 for differentiating cognitively normal controls (CN) from dementia patients in a held-out test set. When combined with clinical biomarkers (apolipoprotein E ε4 allele count, memory, executive function), a model including aBA and aFI predicted 5-year dementia risk among MCI patients with an out-of-sample AUC of 0.88. In the prognostic model, BA and FI offered complementary value (both βs 0.50). The tested MAs did not improve predictions. Results were consistent across FI algorithms, with data-driven health deficit selection yielding the best performance. FI had a stronger adverse effect on prognosis in males, while BA{\textquoteright}s impact was greater in females. Our findings highlight the complementary value of BA and FI in dementia prediction. The results support a multidimensional view of dementia, including an intertwined relationship between the biomarkers, sex, and prognosis. The tested MA{\textquoteright}s limited contribution suggests caution in their use for individual risk assessment of dementia.",

keywords = "Biological age, Brain age, Deep learning, Dementia, Frailty index, Machine learning, Methylation age",

author = "Andreas Engvig and Kalleberg, \{Karl Trygve\} and Westlye, \{Lars T.\} and Leonardsen, \{Esten H{\o}yland\} and Raj, \{Balebail Ashok\} and Kristin Fargher and Amanda Smith and Lisa Raudin and Gloria Chaing and Norman Relkin and Smith, \{Karen Elizabeth\} and Hyungsub Shim and Ponto, \{Laura L.Boles\} and Schultz, \{Susan K.\} and Antero Sarrael and Raymundo Hernando and Nunzio Pomara and Dick Drost and Andrew Kertesz and John Rogers and Irina Rachinsky and Stephen Pasternak and Elizabether Finger and David Bachman and Kenneth Spicer and Jacobo Mintzer and Miller, \{Bruce L.\} and Rosen, \{Howard J.\} and Stephen Correia and Paul Malloy and Stephen Salloway and Geoffrey Tremont and Henry Querfurth and Ott, \{Brian R.\} and Franklin Watkins and Pradeep Garg and Williamson, \{Jeff D.\} and Sink, \{Kaycee M.\} and Schwartz, \{Eben S.\} and Kitzmiller, \{Tamar J.\} and Santulli, \{Robert B.\} and Karen Anderson and Karen Blank and Pearlson, \{Godfrey D.\} and Brown, \{Alice D.\} and Dzintra Celmins and Zimmerman, \{Earl A.\} and Anahita Adeli and Richard King and Greg Jicha",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2024.",

year = "2025",

month = apr,

doi = "10.1007/s11357-024-01376-w",

language = "English",

volume = "47",

pages = "2099--2118",

number = "2",

}

TY - JOUR

T1 - Complementary value of molecular, phenotypic, and functional aging biomarkers in dementia prediction

AU - Engvig, Andreas

AU - Kalleberg, Karl Trygve

AU - Westlye, Lars T.

AU - Leonardsen, Esten Høyland

AU - Raj, Balebail Ashok

AU - Fargher, Kristin

AU - Smith, Amanda

AU - Raudin, Lisa

AU - Chaing, Gloria

AU - Relkin, Norman

AU - Smith, Karen Elizabeth

AU - Shim, Hyungsub

AU - Ponto, Laura L.Boles

AU - Schultz, Susan K.

AU - Sarrael, Antero

AU - Hernando, Raymundo

AU - Pomara, Nunzio

AU - Drost, Dick

AU - Kertesz, Andrew

AU - Rogers, John

AU - Rachinsky, Irina

AU - Pasternak, Stephen

AU - Finger, Elizabether

AU - Bachman, David

AU - Spicer, Kenneth

AU - Mintzer, Jacobo

AU - Miller, Bruce L.

AU - Rosen, Howard J.

AU - Correia, Stephen

AU - Malloy, Paul

AU - Salloway, Stephen

AU - Tremont, Geoffrey

AU - Querfurth, Henry

AU - Ott, Brian R.

AU - Watkins, Franklin

AU - Garg, Pradeep

AU - Williamson, Jeff D.

AU - Sink, Kaycee M.

AU - Schwartz, Eben S.

AU - Kitzmiller, Tamar J.

AU - Santulli, Robert B.

AU - Anderson, Karen

AU - Blank, Karen

AU - Pearlson, Godfrey D.

AU - Brown, Alice D.

AU - Celmins, Dzintra

AU - Zimmerman, Earl A.

AU - Adeli, Anahita

AU - King, Richard

AU - Jicha, Greg

PY - 2025/4

Y1 - 2025/4

N2 - DNA methylation age (MA), brain age (BA), and frailty index (FI) are putative aging biomarkers linked to dementia risk. We investigated their relationship and combined potential for prediction of cognitive impairment and future dementia risk using the ADNI database. Of several MA algorithms, DunedinPACE and GrimAge2, associated with memory, were combined in a composite MA alongside BA and a data-driven FI in predictive analyses. Pairwise correlations between age- and sex-adjusted measures for MA (aMA), aBA, and aFI were low. FI outperformed BA and MA in all diagnostic tasks. A model including age, sex, and aFI achieved an area under the curve (AUC) of 0.94 for differentiating cognitively normal controls (CN) from dementia patients in a held-out test set. When combined with clinical biomarkers (apolipoprotein E ε4 allele count, memory, executive function), a model including aBA and aFI predicted 5-year dementia risk among MCI patients with an out-of-sample AUC of 0.88. In the prognostic model, BA and FI offered complementary value (both βs 0.50). The tested MAs did not improve predictions. Results were consistent across FI algorithms, with data-driven health deficit selection yielding the best performance. FI had a stronger adverse effect on prognosis in males, while BA’s impact was greater in females. Our findings highlight the complementary value of BA and FI in dementia prediction. The results support a multidimensional view of dementia, including an intertwined relationship between the biomarkers, sex, and prognosis. The tested MA’s limited contribution suggests caution in their use for individual risk assessment of dementia.

AB - DNA methylation age (MA), brain age (BA), and frailty index (FI) are putative aging biomarkers linked to dementia risk. We investigated their relationship and combined potential for prediction of cognitive impairment and future dementia risk using the ADNI database. Of several MA algorithms, DunedinPACE and GrimAge2, associated with memory, were combined in a composite MA alongside BA and a data-driven FI in predictive analyses. Pairwise correlations between age- and sex-adjusted measures for MA (aMA), aBA, and aFI were low. FI outperformed BA and MA in all diagnostic tasks. A model including age, sex, and aFI achieved an area under the curve (AUC) of 0.94 for differentiating cognitively normal controls (CN) from dementia patients in a held-out test set. When combined with clinical biomarkers (apolipoprotein E ε4 allele count, memory, executive function), a model including aBA and aFI predicted 5-year dementia risk among MCI patients with an out-of-sample AUC of 0.88. In the prognostic model, BA and FI offered complementary value (both βs 0.50). The tested MAs did not improve predictions. Results were consistent across FI algorithms, with data-driven health deficit selection yielding the best performance. FI had a stronger adverse effect on prognosis in males, while BA’s impact was greater in females. Our findings highlight the complementary value of BA and FI in dementia prediction. The results support a multidimensional view of dementia, including an intertwined relationship between the biomarkers, sex, and prognosis. The tested MA’s limited contribution suggests caution in their use for individual risk assessment of dementia.

KW - Biological age

KW - Brain age

KW - Deep learning

KW - Dementia

KW - Frailty index

KW - Machine learning

KW - Methylation age

UR - http://www.scopus.com/inward/record.url?scp=85207707838&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85207707838&partnerID=8YFLogxK

U2 - 10.1007/s11357-024-01376-w

DO - 10.1007/s11357-024-01376-w

M3 - Article

C2 - 39446224

AN - SCOPUS:85207707838

SN - 2509-2715

VL - 47

SP - 2099

EP - 2118

JO - GeroScience

JF - GeroScience

IS - 2

ER -

Complementary value of molecular, phenotypic, and functional aging biomarkers in dementia prediction

Abstract

Bibliographical note

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this