Premium
Plasma biomarkers predict amyloid pathology in cognitively unimpaired individuals
Author(s) -
den Braber Anouk,
Verberk Inge M.W.,
Tomassen Jori,
Stoops Erik,
Mauroo Kimberley,
Willemsen Gonneke,
Nivard Michel G.,
Van Berckel Bart N.M.,
Scheltens Philip,
Visser Pieter Jelle,
de Geus Eco J.C.,
Teunissen Charlotte E.
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.045470
Subject(s) - amyloid (mycology) , pathology , biomarker , medicine , standardized uptake value , amyloidosis , positron emission tomography , glial fibrillary acidic protein , oncology , nuclear medicine , chemistry , immunohistochemistry , biochemistry
Background Plasma biomarkers have shown promising for identification of patients with Alzheimer’s Disease (AD), but whether this holds for the earliest preclinical stages needs further study. We aimed to investigate the predictive value of plasma amyloid‐β 1‐42/1‐40 and glial fibrillary acidic protein (GFAP) for identifying amyloid pathology in cognitively unimpaired individuals. We tested this using plasma samples collected at time of CSF/PET amyloid assessment and, using plasma samples collected 10 years before CSF/PET amyloid assessment. Methods We selected 151 subjects with normal cognition from the EMIF‐AD PreclinAD study. Amyloid status was defined using visual read of dynamic [ 18 F]flutemetamol PET images or CSF amyloid‐β 1‐42/1‐40 (Euroimmun) <0.065. For a subsample of participants (n=42), plasma samples were available that had been collected 10 years before CSF/PET amyloid assessment as part of the Netherlands Twin Register‐Biobank Study. Simoa assays were applied to measure amyloid‐β 1‐42/1‐40 (ADx NeuroSciences‐AmsterdamUMC Amyblood) and GFAP (Quanterix) plasma levels. Pearson correlations were assessed between plasma, CSF and PET biomarkers. ROC curve analysis was used to determine the value of each plasma biomarker in predicting CSF/PET amyloid status. Results Amyloid‐positive individuals had lower plasma amyloid‐β 1‐42/1‐40 and higher GFAP levels (Table 1). In samples collected at time of CSF/PET amyloid assessment, significant correlations were found between plasma amyloid‐β 1‐42/1‐40 and CSF (r=.22,p=.048) and PET (r=‐.23,p=.035) amyloid, and between plasma and CSF GFAP (r=.33,p=.002). Both plasma markers were able to discriminate amyloid status with similar negative predictive value (.90), with amyloid‐β 1‐42/1‐40 showing higher sensitivity (.72) and GFAP higher specificity (.81). Combining the two did not improve results (Table 2). Plasma amyloid‐β 1‐42/1‐40 and GFAP in samples collected 10 years prior to CSF/PET amyloid assessment were also able to discriminate amyloid status (Table 3). In addition, correlations between these and biomarker levels from plasma collected at time of PET/CSF amyloid assessment were modest for amyloid‐β 1‐42/1‐40 (r=.36, p=.018), and high for GFAP (r=.82, p<0.001). Conclusion Our data suggest that amyloid‐β 1‐42/1‐40 and GFAP plasma markers can be used to identify AD pathology in cognitively unimpaired individuals at present and prospectively. In addition, the high negative predictive value indicates these markers’ potential as pre‐screening tools for secondary prevention trials.