Identifying underlying patterns in Alzheimer's disease trajectory: a deep learning approach and Mendelian randomization analysis

Alzheimer's Disease Neuroimaging Initiatives

doi:10.1016/j.eclinm.2023.102247

Identifying underlying patterns in Alzheimer's disease trajectory: a deep learning approach and Mendelian randomization analysis

Alzheimer's Disease Neuroimaging Initiatives

Department of Information Systems

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

Abstract

Background: Alzheimer's disease (AD) is a heterogeneously progressive neurodegeneration disorder with varied rates of deterioration, either between subjects or within different stages of a certain subject. Estimating the course of AD at early stages has treatment implications. We aimed to analyze disease progression to identify distinct patterns in AD trajectory. Methods: We proposed a deep learning model to identify underlying patterns in the trajectory from cognitively normal (CN) to a state of mild cognitive impairment (MCI) to AD dementia, by jointly predicting time-to-conversion and clustering out distinct subgroups characterized by comprehensive features as well as varied progression rates. We designed and validated our model on the ADNI dataset (1370 participants). Prediction of time-to-conversion in AD trajectory was used to validate the expression of the identified patterns. Causality between patterns and time-to-conversion was further inferred using Mendelian randomization (MR) analysis. External validation was performed on the AIBL dataset (233 participants). Findings: The proposed model clustered out patterns characterized by significantly different biomarkers and varied progression rates. The discovered patterns also showed a strong prediction ability, as indicated by hazard ratio (CN→MCI, HR = 3.51, p < 0.001; MCI→AD, HR = 8.11, p < 0.001), C-Index (CN→MCI, 0.618; MCI→AD, 0.718), and AUC (CN→MCI, 3 years 0.802, 5 years 0.876; MCI→AD, 3 years 0.914, 5 years 0.957). In the external validation cohort, our model demonstrated competitive performance on conversion time prediction (CN→MCI, C-Index = 0.693; MCI→AD, C-Index = 0.752). Moreover, suggestive associations between CN→MCI/MCI→AD patterns with four/three SNPs were mediated and MR analysis indicated a causal link between MCI→AD patterns and time-to-conversion in the first three years. Interpretation: Our proposed model identifies biologically and clinically meaningful patterns from real-world data and provides promising performance on time-to-conversion prediction in AD trajectory, which could promote the understanding of disease progression, facilitate clinical trial design, and provide potential for decision-making. Funding: The National Key Research and Development Program of China, the Key R&D Program of Zhejiang, and the National Nature Science Foundation of China.

Original language	English
Article number	102247
Journal	EClinicalMedicine
Volume	64
DOIs	https://doi.org/10.1016/j.eclinm.2023.102247
State	Published - Oct 2023

Bibliographical note

Funding Information:
The National Key Research and Development Program of China, the Key R&D Program of Zhejiang, and the National Nature Science Foundation of China.This work was partially supported by the National Key Research and Development Program of China under Grant No 2022YFF1202400, the Key R&D Program of Zhejiang under Grant No. 2022C03134, and the National Nature Science Foundation of China under Grant No. 82272129. Data used in preparation of this article was obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf. The AIBL study is a consortium between the Commonwealth Scientific and Industrial Research Organisation (CSIRO), Edith Cowan University, The Florey Institute and Austin Health. Partial financial support was provided by the US Alzheimer's Association, the Alzheimer's Drug Discovery Foundation, an anonymous foundation, the Cooperative Research Centre for Mental Health, the CSIRO Science and Industry Endowment Fund, the Dementia Collaborative Research Centres, the Victorian Government Operational Infrastructure Support program, the Australian Alzheimer's Research Foundation, the National Health and MRC and the Yulgilbar Foundation.

Funding Information:
This work was partially supported by the National Key Research and Development Program of China under Grant No 2022YFF1202400 , the Key R&D Program of Zhejiang under Grant No. 2022C03134 , and the National Nature Science Foundation of China under Grant No. 82272129 . Data used in preparation of this article was obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database ( adni.loni.usc.edu ). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf . The AIBL study is a consortium between the Commonwealth Scientific and Industrial Research Organisation (CSIRO), Edith Cowan University , The Florey Institute and Austin Health. Partial financial support was provided by the US Alzheimer's Association, the Alzheimer's Drug Discovery Foundation , an anonymous foundation, the Cooperative Research Centre for Mental Health , the CSIRO Science and Industry Endowment Fund, the Dementia Collaborative Research Centres , the Victorian Government Operational Infrastructure Support program, the Australian Alzheimer's Research Foundation , the National Health and MRC and the Yulgilbar Foundation .

Publisher Copyright:
© 2023 The Author(s)

Keywords

Alzheimer's disease
Deep learning
Mendelian randomization
Progression pattern
Time-to-conversion prediction

ASJC Scopus subject areas

Medicine (all)

Access to Document

10.1016/j.eclinm.2023.102247

Cite this

@article{8a4bd0177382469791da828e688c5296,

title = "Identifying underlying patterns in Alzheimer's disease trajectory: a deep learning approach and Mendelian randomization analysis",

abstract = "Background: Alzheimer's disease (AD) is a heterogeneously progressive neurodegeneration disorder with varied rates of deterioration, either between subjects or within different stages of a certain subject. Estimating the course of AD at early stages has treatment implications. We aimed to analyze disease progression to identify distinct patterns in AD trajectory. Methods: We proposed a deep learning model to identify underlying patterns in the trajectory from cognitively normal (CN) to a state of mild cognitive impairment (MCI) to AD dementia, by jointly predicting time-to-conversion and clustering out distinct subgroups characterized by comprehensive features as well as varied progression rates. We designed and validated our model on the ADNI dataset (1370 participants). Prediction of time-to-conversion in AD trajectory was used to validate the expression of the identified patterns. Causality between patterns and time-to-conversion was further inferred using Mendelian randomization (MR) analysis. External validation was performed on the AIBL dataset (233 participants). Findings: The proposed model clustered out patterns characterized by significantly different biomarkers and varied progression rates. The discovered patterns also showed a strong prediction ability, as indicated by hazard ratio (CN→MCI, HR = 3.51, p < 0.001; MCI→AD, HR = 8.11, p < 0.001), C-Index (CN→MCI, 0.618; MCI→AD, 0.718), and AUC (CN→MCI, 3 years 0.802, 5 years 0.876; MCI→AD, 3 years 0.914, 5 years 0.957). In the external validation cohort, our model demonstrated competitive performance on conversion time prediction (CN→MCI, C-Index = 0.693; MCI→AD, C-Index = 0.752). Moreover, suggestive associations between CN→MCI/MCI→AD patterns with four/three SNPs were mediated and MR analysis indicated a causal link between MCI→AD patterns and time-to-conversion in the first three years. Interpretation: Our proposed model identifies biologically and clinically meaningful patterns from real-world data and provides promising performance on time-to-conversion prediction in AD trajectory, which could promote the understanding of disease progression, facilitate clinical trial design, and provide potential for decision-making. Funding: The National Key Research and Development Program of China, the Key R&D Program of Zhejiang, and the National Nature Science Foundation of China.",

keywords = "Alzheimer's disease, Deep learning, Mendelian randomization, Progression pattern, Time-to-conversion prediction",

author = "{Alzheimer's Disease Neuroimaging Initiatives} and Fan Yi and Yaoyun Zhang and Jing Yuan and Ziyue Liu and Feifei Zhai and Ankai Hao and Fei Wu and Judith Somekh and Mor Peleg and Zhu, {Yi Cheng} and Zhengxing Huang",

note = "Funding Information: The National Key Research and Development Program of China, the Key R&D Program of Zhejiang, and the National Nature Science Foundation of China.This work was partially supported by the National Key Research and Development Program of China under Grant No 2022YFF1202400, the Key R&D Program of Zhejiang under Grant No. 2022C03134, and the National Nature Science Foundation of China under Grant No. 82272129. Data used in preparation of this article was obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf. The AIBL study is a consortium between the Commonwealth Scientific and Industrial Research Organisation (CSIRO), Edith Cowan University, The Florey Institute and Austin Health. Partial financial support was provided by the US Alzheimer's Association, the Alzheimer's Drug Discovery Foundation, an anonymous foundation, the Cooperative Research Centre for Mental Health, the CSIRO Science and Industry Endowment Fund, the Dementia Collaborative Research Centres, the Victorian Government Operational Infrastructure Support program, the Australian Alzheimer's Research Foundation, the National Health and MRC and the Yulgilbar Foundation. Funding Information: This work was partially supported by the National Key Research and Development Program of China under Grant No 2022YFF1202400 , the Key R&D Program of Zhejiang under Grant No. 2022C03134 , and the National Nature Science Foundation of China under Grant No. 82272129 . Data used in preparation of this article was obtained from the Alzheimer{\textquoteright}s Disease Neuroimaging Initiative (ADNI) database ( adni.loni.usc.edu ). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf . The AIBL study is a consortium between the Commonwealth Scientific and Industrial Research Organisation (CSIRO), Edith Cowan University , The Florey Institute and Austin Health. Partial financial support was provided by the US Alzheimer's Association, the Alzheimer's Drug Discovery Foundation , an anonymous foundation, the Cooperative Research Centre for Mental Health , the CSIRO Science and Industry Endowment Fund, the Dementia Collaborative Research Centres , the Victorian Government Operational Infrastructure Support program, the Australian Alzheimer's Research Foundation , the National Health and MRC and the Yulgilbar Foundation . Publisher Copyright: {\textcopyright} 2023 The Author(s)",

year = "2023",

month = oct,

doi = "10.1016/j.eclinm.2023.102247",

language = "English",

volume = "64",

journal = "EClinicalMedicine",

issn = "2589-5370",

publisher = "Lancet Publishing Group",

}

TY - JOUR

T1 - Identifying underlying patterns in Alzheimer's disease trajectory

T2 - a deep learning approach and Mendelian randomization analysis

AU - Alzheimer's Disease Neuroimaging Initiatives

AU - Yi, Fan

AU - Zhang, Yaoyun

AU - Yuan, Jing

AU - Liu, Ziyue

AU - Zhai, Feifei

AU - Hao, Ankai

AU - Wu, Fei

AU - Somekh, Judith

AU - Peleg, Mor

AU - Zhu, Yi Cheng

AU - Huang, Zhengxing

N1 - Funding Information: The National Key Research and Development Program of China, the Key R&D Program of Zhejiang, and the National Nature Science Foundation of China.This work was partially supported by the National Key Research and Development Program of China under Grant No 2022YFF1202400, the Key R&D Program of Zhejiang under Grant No. 2022C03134, and the National Nature Science Foundation of China under Grant No. 82272129. Data used in preparation of this article was obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf. The AIBL study is a consortium between the Commonwealth Scientific and Industrial Research Organisation (CSIRO), Edith Cowan University, The Florey Institute and Austin Health. Partial financial support was provided by the US Alzheimer's Association, the Alzheimer's Drug Discovery Foundation, an anonymous foundation, the Cooperative Research Centre for Mental Health, the CSIRO Science and Industry Endowment Fund, the Dementia Collaborative Research Centres, the Victorian Government Operational Infrastructure Support program, the Australian Alzheimer's Research Foundation, the National Health and MRC and the Yulgilbar Foundation. Funding Information: This work was partially supported by the National Key Research and Development Program of China under Grant No 2022YFF1202400 , the Key R&D Program of Zhejiang under Grant No. 2022C03134 , and the National Nature Science Foundation of China under Grant No. 82272129 . Data used in preparation of this article was obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database ( adni.loni.usc.edu ). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf . The AIBL study is a consortium between the Commonwealth Scientific and Industrial Research Organisation (CSIRO), Edith Cowan University , The Florey Institute and Austin Health. Partial financial support was provided by the US Alzheimer's Association, the Alzheimer's Drug Discovery Foundation , an anonymous foundation, the Cooperative Research Centre for Mental Health , the CSIRO Science and Industry Endowment Fund, the Dementia Collaborative Research Centres , the Victorian Government Operational Infrastructure Support program, the Australian Alzheimer's Research Foundation , the National Health and MRC and the Yulgilbar Foundation . Publisher Copyright: © 2023 The Author(s)

PY - 2023/10

Y1 - 2023/10

N2 - Background: Alzheimer's disease (AD) is a heterogeneously progressive neurodegeneration disorder with varied rates of deterioration, either between subjects or within different stages of a certain subject. Estimating the course of AD at early stages has treatment implications. We aimed to analyze disease progression to identify distinct patterns in AD trajectory. Methods: We proposed a deep learning model to identify underlying patterns in the trajectory from cognitively normal (CN) to a state of mild cognitive impairment (MCI) to AD dementia, by jointly predicting time-to-conversion and clustering out distinct subgroups characterized by comprehensive features as well as varied progression rates. We designed and validated our model on the ADNI dataset (1370 participants). Prediction of time-to-conversion in AD trajectory was used to validate the expression of the identified patterns. Causality between patterns and time-to-conversion was further inferred using Mendelian randomization (MR) analysis. External validation was performed on the AIBL dataset (233 participants). Findings: The proposed model clustered out patterns characterized by significantly different biomarkers and varied progression rates. The discovered patterns also showed a strong prediction ability, as indicated by hazard ratio (CN→MCI, HR = 3.51, p < 0.001; MCI→AD, HR = 8.11, p < 0.001), C-Index (CN→MCI, 0.618; MCI→AD, 0.718), and AUC (CN→MCI, 3 years 0.802, 5 years 0.876; MCI→AD, 3 years 0.914, 5 years 0.957). In the external validation cohort, our model demonstrated competitive performance on conversion time prediction (CN→MCI, C-Index = 0.693; MCI→AD, C-Index = 0.752). Moreover, suggestive associations between CN→MCI/MCI→AD patterns with four/three SNPs were mediated and MR analysis indicated a causal link between MCI→AD patterns and time-to-conversion in the first three years. Interpretation: Our proposed model identifies biologically and clinically meaningful patterns from real-world data and provides promising performance on time-to-conversion prediction in AD trajectory, which could promote the understanding of disease progression, facilitate clinical trial design, and provide potential for decision-making. Funding: The National Key Research and Development Program of China, the Key R&D Program of Zhejiang, and the National Nature Science Foundation of China.

AB - Background: Alzheimer's disease (AD) is a heterogeneously progressive neurodegeneration disorder with varied rates of deterioration, either between subjects or within different stages of a certain subject. Estimating the course of AD at early stages has treatment implications. We aimed to analyze disease progression to identify distinct patterns in AD trajectory. Methods: We proposed a deep learning model to identify underlying patterns in the trajectory from cognitively normal (CN) to a state of mild cognitive impairment (MCI) to AD dementia, by jointly predicting time-to-conversion and clustering out distinct subgroups characterized by comprehensive features as well as varied progression rates. We designed and validated our model on the ADNI dataset (1370 participants). Prediction of time-to-conversion in AD trajectory was used to validate the expression of the identified patterns. Causality between patterns and time-to-conversion was further inferred using Mendelian randomization (MR) analysis. External validation was performed on the AIBL dataset (233 participants). Findings: The proposed model clustered out patterns characterized by significantly different biomarkers and varied progression rates. The discovered patterns also showed a strong prediction ability, as indicated by hazard ratio (CN→MCI, HR = 3.51, p < 0.001; MCI→AD, HR = 8.11, p < 0.001), C-Index (CN→MCI, 0.618; MCI→AD, 0.718), and AUC (CN→MCI, 3 years 0.802, 5 years 0.876; MCI→AD, 3 years 0.914, 5 years 0.957). In the external validation cohort, our model demonstrated competitive performance on conversion time prediction (CN→MCI, C-Index = 0.693; MCI→AD, C-Index = 0.752). Moreover, suggestive associations between CN→MCI/MCI→AD patterns with four/three SNPs were mediated and MR analysis indicated a causal link between MCI→AD patterns and time-to-conversion in the first three years. Interpretation: Our proposed model identifies biologically and clinically meaningful patterns from real-world data and provides promising performance on time-to-conversion prediction in AD trajectory, which could promote the understanding of disease progression, facilitate clinical trial design, and provide potential for decision-making. Funding: The National Key Research and Development Program of China, the Key R&D Program of Zhejiang, and the National Nature Science Foundation of China.

KW - Alzheimer's disease

KW - Deep learning

KW - Mendelian randomization

KW - Progression pattern

KW - Time-to-conversion prediction

UR - http://www.scopus.com/inward/record.url?scp=85172247834&partnerID=8YFLogxK

U2 - 10.1016/j.eclinm.2023.102247

DO - 10.1016/j.eclinm.2023.102247

M3 - Article

AN - SCOPUS:85172247834

SN - 2589-5370

VL - 64

JO - EClinicalMedicine

JF - EClinicalMedicine

M1 - 102247

ER -

Identifying underlying patterns in Alzheimer's disease trajectory: a deep learning approach and Mendelian randomization analysis

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this