TY - JOUR
T1 - The cerebrospinal fluid "Alzheimer profile": Easily said, but what does it mean?
AU - Duits, Flora H.
AU - Teunissen, Charlotte E.
AU - Bouwman, Femke H.
AU - Visser, Pieter-Jelle
AU - Mattsson, Niklas
AU - Zetterberg, Henrik
AU - Blennow, Kaj
AU - Hansson, Oskar
AU - Minthon, Lennart
AU - Andreasen, Niels
AU - Marcusson, Jan
AU - Wallin, Anders
AU - Rikkert, Marcel Olde
AU - Tsolaki, Magda
AU - Parnetti, Lucilla
AU - Herukka, Sanna-Kaisa
AU - Hampel, Harald
AU - De Leon, Mony J.
AU - Schroeder, Johannes
AU - Aarsland, Dag
AU - Blankenstein, Marinus A.
AU - Scheltens, Philip
AU - van der Flier, Wiesje M.
PY - 2014/11
Y1 - 2014/11
N2 - Background: We aimed to identify the most useful definition of the "cerebrospinal fluid Alzheimer profile," based on amyloid-beta(1-42) (A beta(42)), total tau, and phosphorylated tau (p-tau), for diagnosis and prognosis of Alzheimer's disease (AD). Methods: We constructed eight Alzheimer profiles with previously published combinations, including regression formulas and simple ratios. We compared their diagnostic accuracy and ability to predict dementia due to AD in 1385 patients from the Amsterdam Dementia Cohort. Results were validated in an independent cohort (n = 1442). Results: Combinations outperformed individual biomarkers. Based on the sensitivity of the best performing regression formulas, cutoffs were chosen at 0.52 for the tau/A beta(42) ratio and 0.08 for the p-tau/A beta(42) ratio. Ratios performed similar to formulas (sensitivity, 91%-93%; specificity, 81%-84%). The same combinations best predicted cognitive decline in mild cognitive impairment patients. Validation confirmed these results, especially regarding the tau/A beta(42) ratio. Conclusions: A tau/A beta(42) ratio of >0.52 constitutes a robust cerebrospinal fluid Alzheimer profile. We recommend using this ratio to combine biomarkers.
AB - Background: We aimed to identify the most useful definition of the "cerebrospinal fluid Alzheimer profile," based on amyloid-beta(1-42) (A beta(42)), total tau, and phosphorylated tau (p-tau), for diagnosis and prognosis of Alzheimer's disease (AD). Methods: We constructed eight Alzheimer profiles with previously published combinations, including regression formulas and simple ratios. We compared their diagnostic accuracy and ability to predict dementia due to AD in 1385 patients from the Amsterdam Dementia Cohort. Results were validated in an independent cohort (n = 1442). Results: Combinations outperformed individual biomarkers. Based on the sensitivity of the best performing regression formulas, cutoffs were chosen at 0.52 for the tau/A beta(42) ratio and 0.08 for the p-tau/A beta(42) ratio. Ratios performed similar to formulas (sensitivity, 91%-93%; specificity, 81%-84%). The same combinations best predicted cognitive decline in mild cognitive impairment patients. Validation confirmed these results, especially regarding the tau/A beta(42) ratio. Conclusions: A tau/A beta(42) ratio of >0.52 constitutes a robust cerebrospinal fluid Alzheimer profile. We recommend using this ratio to combine biomarkers.
KW - Dementia
KW - Alzheimer's disease
KW - Mild cognitive impairment
KW - Cerebrospinal fluid
KW - Amyloid-beta(1-42)
KW - Tau
KW - Biomarkers
KW - Differential diagnosis
KW - Predictive value
U2 - 10.1016/j.jalz.2013.12.023
DO - 10.1016/j.jalz.2013.12.023
M3 - Article
C2 - 24721526
VL - 10
SP - 713
EP - 723
JO - Alzheimer's & Dementia
JF - Alzheimer's & Dementia
SN - 1552-5260
IS - 6
ER -