P1‐321: Biological standards to control variability in cerebrospinal fluid beta‐amyloid readings

Abstract

Listen

Translate article

CSF Aβ 42 is a promising biomarker of the Alzheimer's disease (AD) neurodegenerative process. However, measurement variability across laboratories may limit its utility as a diagnostic marker [Vanderstichele et al., Alzheimer & Dementia (2012;8:65-73)]. Laboratory control samples composed of pooled human CSF can be used to assess variability of measurements, and may further be a means of standardizing readings within and across laboratories. To investigate this, CSF samples from 104 AD cases and 300 normal controls were assayed by multiplex bead-based immunoassay (INNO-BIA AlzBio3, Innogenetics) at a single laboratory per the manufacturer's protocol using a total of 19 immunoassay plates. Large volume control CSF samples representing high, medium, and low CSF Aβ 42 levels were constructed by pooling samples from previously ascertained subjects. Control CSF samples were assayed in duplicate on each plate. CSF sample concentrations estimated by standard procedures were further calibrated to the CSF control samples using linear calibration curves fit to data from each plate. All standardization exercises were performed blinded to the case-control status of the samples. Analytic (synthetic Aβ) control sample concentrations did not vary consistently from plate to plate, while biological control CSF samples were often consistently relatively high or low. Intra-plate recalibration to the biologic control CSF samples reduced the overall variance of Aβ 42 levels by 20 percent. Graphical representation of the Aβ 42 levels before and after calibration suggest a substantive focusing of the data. By including a representative range of biologic control samples on each plate we were able to analytically remove the plate-to-plate drift from the sample readings and optimize the research utility of these samples. Extension to intra-laboratory calibration may help with standardization.