CSF beta-amyloid 1–42 – what are we measuring in Alzheimer's disease?

OBJECTIVE To characterize biological and technical factors which influence cerebrospinal fluid (CSF) Alzheimer's disease (AD) biomarker levels, including the presence of apolipoprotein E (APOE) ε4 allele, AD diagnosis, Aβ-binding proteins, sample processing, and preanalytical handling. METHODS CSF was collected from 140 subjects with normal cognition, mild cognitive impairment, AD, and non-AD dementia. CSF levels of beta-amyloid 1-42 (Aβ42), total Tau (t-Tau), and Tau phosphorylated at threonine 181 (p-Tau181) were analyzed following the standard and modified protocols. CSF levels of apoJ, apoE, albumin, and α-synuclein were measured in a subgroup (n = 69), and their effects on measured AD biomarker levels were also determined in vitro using human CSF samples. RESULTS CSF Aβ42 levels measured using the AD Neuro-imaging Initiative (ADNI) protocol (which we call suspended Aβ42 or susAβ) were lower than total measurable CSF Aβ42 in all groups, and on average represents 57% of the latter. Logistic regression analysis showed this proportion (% susAβ) to be directly correlated with CSF Aβ42 and apoJ levels, but inversely correlated with CSF t-Tau levels. Finally, we showed in vitro that increasing apoE and apoJ levels directly increased % susAβ. CONCLUSION CSF susAβ levels are influenced by biological and technical factors, and may represent a marker of Aβ susceptible to lipoprotein-mediated clearance. Clinical trials should include total measurable Aβ42 and susAβ to better inform outcomes.