Age but not diagnosis is the main predictor of plasma amyloid β-protein levels

Background: Plasma amyloid beta-protein Abeta42 levels are increased in patients with familial Alzheimer disease (AD) mutations, and high levels reportedly identify individuals at risk to develop AD. Objectives: To determine whether there are characteristic changes in plasma Abeta40 and Abeta42 levels in sporadic AD, and to examine the relationship of plasma Abeta measures with clinical, demographic, and genetic variables in a prospectively characterized outpatient clinic population. Patients: A total of 371 outpatients with sporadic AD (n = 146), mild cognitive impairment (n = 37), or Parkinson disease (n = 96) and nondemented control cases (n = 92). Methods: We collected plasma samples and determined Abeta40 and Abeta42 levels by sandwich enzyme-linked immunosorbent assay with the use of the capture antibody BNT77 (anti-Abeta11-28) and the detector anti-bodies horseradish peroxidase-linked BA27 (anti-Abeta40) and BC05 (anti-Abeta42). Results: Mean Abeta40 and Abeta42 levels increased significantly with age in each diagnostic group. When covaried for age, mean plasma levels of Abeta40 and Abeta42 did not differ significantly among the 4 diagnostic groups. Within the mild cognitive impairment and AD groups, Abeta40 and Abeta42 levels did not correlate with duration of memory impairment or with cognitive test scores. The Abeta measures were not influenced by family history of AD, apolipoprotein E genotype, or current medication use of cholinesterase inhibitors, vitamin E, statins, nonsteroidal anti-inflammatory drugs, or estrogen. Conclusions: Plasma Abeta measures increase with age, but, in contrast to reports on familial AD, plasma Abeta measures were neither sensitive nor specific for the clinical diagnosis of mild cognitive impairment or sporadic AD.