P1–166: Inter‐laboratory variation when using a unified test procedure for INNO‐BIA AlzBio3
Manu Leen Jef Vandijck, Ryozo Kuwano, Teresa Waligórska, Samantha De Smet, Tamao Tsukie, Lia Verdoodt, Willy Deleersnijder, John Q. Trojanowski, Takeshi Iwatsubo, Leslie Shaw · Alzheimer s & Dementia · 2013
Diagnosis of Alzheimer's disease and follow-up of treatment with disease-modifying drugs supported by CSF biomarkers amyloid-β 1–42 (Aβ 1–42), total tau (hTau), and phosphorylated tau (P-Tau 181P) is gaining acceptance. This increases the need for improved harmonization of the current CSF biomarker assays. A large between center analytical variability is observed, especially for Aβ 1–42. This is partially due to the intrinsic properties of the peptide and the absence of detailed laboratory Standard Operating Procedures (SOP) documenting all critical test parameters. Well-designed inter-laboratory evaluations aiming at the identification of the most important factors are needed. After comprehensive analysis of their laboratory SOPs and determination of the impact of each step of the test procedure, three different centers (A, B, and C) mutually agreed upon a detailed unified SOP for the INNO-BIA AlzBio3 (RUO; Innogenetics nv, Belgium) multiplexing xMAP ® assay. Using the same assay batch, a set of ten CSF pool samples and two buffer based control samples were analyzed using ready-to-use calibrators. A fresh aliquot of each sample was tested in quadruplicate in three independent assay runs (only 2 runs for center C). The goal of the study was to determine the with-in and inter-laboratory variability of the assay for hTau, Aβ 1–42, and P-Tau 181P after implementation of the unified SOP. The CSF analyte concentrations showed a strong correlation between all centers (R > 0.95; regression center versus overall center mean). For the CSF samples, the total inter-center variability over the 8 runs performed (%CV) ranged between 12.4 - 22.9% (hTau), 8.1 - 13.1% (Aβ 1–42), and 6.8 - 13.2% (P-Tau 181P). Across the 10 CSF samples, the mean within-laboratory variation ranged between 1.8% and 11% for all analytes. The variability observed for the control samples was even lower (max. 4.5%). Implementation of a unified SOP in the three laboratories resulted in an acceptable inter-laboratory variation of analyte concentrations for CSF - and control samples. Careful documentation of the critical parameters of the test procedure and rigorous adherence to a detailed instruction of use clearly leads to highly comparable CSF analyte concentrations between experienced centers.