Imaging markers for Alzheimer disease

Which vs how

Giovanni B. Frisoni, Martina Bocchetta, Gael Chételat, Gil D. Rabinovici, Mony J. De Leon, Jeffrey Kaye, Eric M. Reiman, Philip Scheltens, Frederik Barkhof, Sandra E. Black, David J. Brooks, Maria C. Carrillo, Nick C. Fox, Karl Herholz, Agneta Nordberg, Clifford R. Jack, William J. Jagust, Keith A. Johnson, Christopher C. Rowe, Reisa A. Sperling & 5 others William Thies, Lars Olof Wahlund, Michael W. Weiner, Patrizio Pasqualetti, Charles DeCarli

Research output: Contribution to journalArticle

139 Citations (Scopus)

Abstract

Revised diagnostic criteria for Alzheimer disease (AD) acknowledge a key role of imaging biomarkers for early diagnosis. Diagnostic accuracy depends on which marker (i.e., amyloid imaging, 18F-fluorodeoxyglucose [FDG]-PET, SPECT, MRI) as well as how it is measured ("metric": visual, manual, semiautomated, or automated segmentation/computation). We evaluated diagnostic accuracy of marker vs metric in separating AD from healthy and prognostic accuracy to predict progression in mild cognitive impairment. The outcome measure was positive (negative) likelihood ratio, LR1 (LR2), defined as the ratio between the probability of positive (negative) test outcome in patients and the probability of positive (negative) test outcome in healthy controls. Diagnostic LR1 of markers was between 4.4 and 9.4 and LR2 between 0.25 and 0.08, whereas prognostic LR1 and LR2 were between 1.7 and 7.5, and 0.50 and 0.11, respectively. Within metrics, LRs varied up to 100-fold: LR1 from approximately 1 to 100; LR2 from approximately 1.00 to 0.01. Markers accounted for 11% and 18% of diagnostic and prognostic variance of LR1 and 16% and 24% of LR2. Across all markers, metrics accounted for an equal or larger amount of variance than markers: 13% and 62% of diagnostic and prognostic variance of LR1, and 29% and 18% of LR2. Within markers, the largest proportion of diagnostic LR1 and LR2 variability was within 18F-FDG-PET and MRI metrics, respectively. Diagnostic and prognostic accuracy of imaging AD biomarkers is at least as dependent on how the biomarker is measured as on the biomarker itself. Standard operating procedures are key to biomarker use in the clinical routine and drug trials.

Original languageEnglish (US)
Pages (from-to)487-500
Number of pages14
JournalNeurology
Volume81
Issue number5
DOIs
StatePublished - Jul 30 2013

Fingerprint

Alzheimer Disease
Biomarkers
Fluorodeoxyglucose F18
Single-Photon Emission-Computed Tomography
Amyloid
Early Diagnosis
Outcome Assessment (Health Care)
Alzheimer's Disease
Imaging
Diagnostics
Pharmaceutical Preparations
Diagnostic Accuracy

ASJC Scopus subject areas

  • Clinical Neurology
  • Arts and Humanities (miscellaneous)

Cite this

Frisoni, G. B., Bocchetta, M., Chételat, G., Rabinovici, G. D., De Leon, M. J., Kaye, J., ... DeCarli, C. (2013). Imaging markers for Alzheimer disease: Which vs how. Neurology, 81(5), 487-500. https://doi.org/10.1212/WNL.0b013e31829d86e8

Imaging markers for Alzheimer disease : Which vs how. / Frisoni, Giovanni B.; Bocchetta, Martina; Chételat, Gael; Rabinovici, Gil D.; De Leon, Mony J.; Kaye, Jeffrey; Reiman, Eric M.; Scheltens, Philip; Barkhof, Frederik; Black, Sandra E.; Brooks, David J.; Carrillo, Maria C.; Fox, Nick C.; Herholz, Karl; Nordberg, Agneta; Jack, Clifford R.; Jagust, William J.; Johnson, Keith A.; Rowe, Christopher C.; Sperling, Reisa A.; Thies, William; Wahlund, Lars Olof; Weiner, Michael W.; Pasqualetti, Patrizio; DeCarli, Charles.

In: Neurology, Vol. 81, No. 5, 30.07.2013, p. 487-500.

Research output: Contribution to journalArticle

Frisoni, GB, Bocchetta, M, Chételat, G, Rabinovici, GD, De Leon, MJ, Kaye, J, Reiman, EM, Scheltens, P, Barkhof, F, Black, SE, Brooks, DJ, Carrillo, MC, Fox, NC, Herholz, K, Nordberg, A, Jack, CR, Jagust, WJ, Johnson, KA, Rowe, CC, Sperling, RA, Thies, W, Wahlund, LO, Weiner, MW, Pasqualetti, P & DeCarli, C 2013, 'Imaging markers for Alzheimer disease: Which vs how', Neurology, vol. 81, no. 5, pp. 487-500. https://doi.org/10.1212/WNL.0b013e31829d86e8
Frisoni GB, Bocchetta M, Chételat G, Rabinovici GD, De Leon MJ, Kaye J et al. Imaging markers for Alzheimer disease: Which vs how. Neurology. 2013 Jul 30;81(5):487-500. https://doi.org/10.1212/WNL.0b013e31829d86e8
Frisoni, Giovanni B. ; Bocchetta, Martina ; Chételat, Gael ; Rabinovici, Gil D. ; De Leon, Mony J. ; Kaye, Jeffrey ; Reiman, Eric M. ; Scheltens, Philip ; Barkhof, Frederik ; Black, Sandra E. ; Brooks, David J. ; Carrillo, Maria C. ; Fox, Nick C. ; Herholz, Karl ; Nordberg, Agneta ; Jack, Clifford R. ; Jagust, William J. ; Johnson, Keith A. ; Rowe, Christopher C. ; Sperling, Reisa A. ; Thies, William ; Wahlund, Lars Olof ; Weiner, Michael W. ; Pasqualetti, Patrizio ; DeCarli, Charles. / Imaging markers for Alzheimer disease : Which vs how. In: Neurology. 2013 ; Vol. 81, No. 5. pp. 487-500.
@article{b657b19418064bcba1afa06c77ee8b45,
title = "Imaging markers for Alzheimer disease: Which vs how",
abstract = "Revised diagnostic criteria for Alzheimer disease (AD) acknowledge a key role of imaging biomarkers for early diagnosis. Diagnostic accuracy depends on which marker (i.e., amyloid imaging, 18F-fluorodeoxyglucose [FDG]-PET, SPECT, MRI) as well as how it is measured ({"}metric{"}: visual, manual, semiautomated, or automated segmentation/computation). We evaluated diagnostic accuracy of marker vs metric in separating AD from healthy and prognostic accuracy to predict progression in mild cognitive impairment. The outcome measure was positive (negative) likelihood ratio, LR1 (LR2), defined as the ratio between the probability of positive (negative) test outcome in patients and the probability of positive (negative) test outcome in healthy controls. Diagnostic LR1 of markers was between 4.4 and 9.4 and LR2 between 0.25 and 0.08, whereas prognostic LR1 and LR2 were between 1.7 and 7.5, and 0.50 and 0.11, respectively. Within metrics, LRs varied up to 100-fold: LR1 from approximately 1 to 100; LR2 from approximately 1.00 to 0.01. Markers accounted for 11{\%} and 18{\%} of diagnostic and prognostic variance of LR1 and 16{\%} and 24{\%} of LR2. Across all markers, metrics accounted for an equal or larger amount of variance than markers: 13{\%} and 62{\%} of diagnostic and prognostic variance of LR1, and 29{\%} and 18{\%} of LR2. Within markers, the largest proportion of diagnostic LR1 and LR2 variability was within 18F-FDG-PET and MRI metrics, respectively. Diagnostic and prognostic accuracy of imaging AD biomarkers is at least as dependent on how the biomarker is measured as on the biomarker itself. Standard operating procedures are key to biomarker use in the clinical routine and drug trials.",
author = "Frisoni, {Giovanni B.} and Martina Bocchetta and Gael Ch{\'e}telat and Rabinovici, {Gil D.} and {De Leon}, {Mony J.} and Jeffrey Kaye and Reiman, {Eric M.} and Philip Scheltens and Frederik Barkhof and Black, {Sandra E.} and Brooks, {David J.} and Carrillo, {Maria C.} and Fox, {Nick C.} and Karl Herholz and Agneta Nordberg and Jack, {Clifford R.} and Jagust, {William J.} and Johnson, {Keith A.} and Rowe, {Christopher C.} and Sperling, {Reisa A.} and William Thies and Wahlund, {Lars Olof} and Weiner, {Michael W.} and Patrizio Pasqualetti and Charles DeCarli",
year = "2013",
month = "7",
day = "30",
doi = "10.1212/WNL.0b013e31829d86e8",
language = "English (US)",
volume = "81",
pages = "487--500",
journal = "Neurology",
issn = "0028-3878",
publisher = "Lippincott Williams and Wilkins",
number = "5",

}

TY - JOUR

T1 - Imaging markers for Alzheimer disease

T2 - Which vs how

AU - Frisoni, Giovanni B.

AU - Bocchetta, Martina

AU - Chételat, Gael

AU - Rabinovici, Gil D.

AU - De Leon, Mony J.

AU - Kaye, Jeffrey

AU - Reiman, Eric M.

AU - Scheltens, Philip

AU - Barkhof, Frederik

AU - Black, Sandra E.

AU - Brooks, David J.

AU - Carrillo, Maria C.

AU - Fox, Nick C.

AU - Herholz, Karl

AU - Nordberg, Agneta

AU - Jack, Clifford R.

AU - Jagust, William J.

AU - Johnson, Keith A.

AU - Rowe, Christopher C.

AU - Sperling, Reisa A.

AU - Thies, William

AU - Wahlund, Lars Olof

AU - Weiner, Michael W.

AU - Pasqualetti, Patrizio

AU - DeCarli, Charles

PY - 2013/7/30

Y1 - 2013/7/30

N2 - Revised diagnostic criteria for Alzheimer disease (AD) acknowledge a key role of imaging biomarkers for early diagnosis. Diagnostic accuracy depends on which marker (i.e., amyloid imaging, 18F-fluorodeoxyglucose [FDG]-PET, SPECT, MRI) as well as how it is measured ("metric": visual, manual, semiautomated, or automated segmentation/computation). We evaluated diagnostic accuracy of marker vs metric in separating AD from healthy and prognostic accuracy to predict progression in mild cognitive impairment. The outcome measure was positive (negative) likelihood ratio, LR1 (LR2), defined as the ratio between the probability of positive (negative) test outcome in patients and the probability of positive (negative) test outcome in healthy controls. Diagnostic LR1 of markers was between 4.4 and 9.4 and LR2 between 0.25 and 0.08, whereas prognostic LR1 and LR2 were between 1.7 and 7.5, and 0.50 and 0.11, respectively. Within metrics, LRs varied up to 100-fold: LR1 from approximately 1 to 100; LR2 from approximately 1.00 to 0.01. Markers accounted for 11% and 18% of diagnostic and prognostic variance of LR1 and 16% and 24% of LR2. Across all markers, metrics accounted for an equal or larger amount of variance than markers: 13% and 62% of diagnostic and prognostic variance of LR1, and 29% and 18% of LR2. Within markers, the largest proportion of diagnostic LR1 and LR2 variability was within 18F-FDG-PET and MRI metrics, respectively. Diagnostic and prognostic accuracy of imaging AD biomarkers is at least as dependent on how the biomarker is measured as on the biomarker itself. Standard operating procedures are key to biomarker use in the clinical routine and drug trials.

AB - Revised diagnostic criteria for Alzheimer disease (AD) acknowledge a key role of imaging biomarkers for early diagnosis. Diagnostic accuracy depends on which marker (i.e., amyloid imaging, 18F-fluorodeoxyglucose [FDG]-PET, SPECT, MRI) as well as how it is measured ("metric": visual, manual, semiautomated, or automated segmentation/computation). We evaluated diagnostic accuracy of marker vs metric in separating AD from healthy and prognostic accuracy to predict progression in mild cognitive impairment. The outcome measure was positive (negative) likelihood ratio, LR1 (LR2), defined as the ratio between the probability of positive (negative) test outcome in patients and the probability of positive (negative) test outcome in healthy controls. Diagnostic LR1 of markers was between 4.4 and 9.4 and LR2 between 0.25 and 0.08, whereas prognostic LR1 and LR2 were between 1.7 and 7.5, and 0.50 and 0.11, respectively. Within metrics, LRs varied up to 100-fold: LR1 from approximately 1 to 100; LR2 from approximately 1.00 to 0.01. Markers accounted for 11% and 18% of diagnostic and prognostic variance of LR1 and 16% and 24% of LR2. Across all markers, metrics accounted for an equal or larger amount of variance than markers: 13% and 62% of diagnostic and prognostic variance of LR1, and 29% and 18% of LR2. Within markers, the largest proportion of diagnostic LR1 and LR2 variability was within 18F-FDG-PET and MRI metrics, respectively. Diagnostic and prognostic accuracy of imaging AD biomarkers is at least as dependent on how the biomarker is measured as on the biomarker itself. Standard operating procedures are key to biomarker use in the clinical routine and drug trials.

UR - http://www.scopus.com/inward/record.url?scp=84881287638&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84881287638&partnerID=8YFLogxK

U2 - 10.1212/WNL.0b013e31829d86e8

DO - 10.1212/WNL.0b013e31829d86e8

M3 - Article

VL - 81

SP - 487

EP - 500

JO - Neurology

JF - Neurology

SN - 0028-3878

IS - 5

ER -