Accuracy of cerebrospinal fluid Aβ1-42 measurements: evaluation of pre-analytical factors using a novel Elecsys immunosassay

Background: A decreased level of A beta 1-42 in cerebrospinal fluid (CSF) is characteristic of Alzheimer disease and often used to support clinical diagnosis. The measured concentration of CSF A beta 1-42, however, depends strongly on several pre-analytical and analytical confounding factors such as sample collection, material of testing tube, CSF handling and storage procedures (e.g. transfer to new tubes after centrifugation, freeze-thaw effects). As a consequence, substantial variations in the measured levels of this biomarker are observed even for the same sample. This study investigates whether the accuracy of quantitative analysis of CSF A beta 1-42 can be improved by pre-analytical treatment of CSF with agents that could potentially reduce a freeze-thaw and adhesion-related depletion of A beta 1-42 from CSF, including modulators of A beta aggregation and cryoprotecting or anti-adhesion agents. Methods: The concentration of CSF A beta 1-42 was assessed with a novel Elecsys immunoassay developed for quantification of A beta 1-42 in human CSF. Results: Low-molecular weight A beta oligomerization inhibitors, beta-sheet breaker peptides, or the mid domain 4G8 antibody do not improve the stability of CSF A beta 1-42 during a repeated freeze-thaw treatment. Cryoprotecting agents reduce a freeze-thaw dependent loss of A beta 1-42 only when spiked to CSF to final concentration of 300 mM or higher. Adhesion of A beta 1-42 can be prevented by pre-treating CSF with Tween or by using tubes with a siliconized surface. Conclusions: Between-center variability in measured level of CSF A beta 1-42 can be reduced only by standardized CSF collection into one specific tube that, without centrifugation, transfer or other types of pre-analytical processing, is directly analyzed after sample collection.