Abstract Number: PB0539
Meeting: ISTH 2020 Congress
Background: Analysis of FVIII activity is important for diagnosis of FVIII deficiency, which is associated with bleeding. Conversely, raised levels of FVIII are a thrombotic risk. We conducted a multicentre evaluation of a new FVIII assay (Roche Diagnostics), intended for quantitative assessment of FVIII activity.
Aims: To evaluate the analytical performance of the FVIII assay on the cobas t 711 analyser.
Methods: Experiments were performed at three laboratories using 3.2% citrated, anonymised, ethically approved, residual or commercially purchased plasma samples. Five human plasma pools (HPP) and two controls were used for determination of assay repeatability (1 run per site; n=21 replicates per sample), intermediate precision and total reproducibility (5 aliquots/sample per day over 5 days); coefficients of variation (CV) and/or standard deviation (SD) were calculated. Lot-to-lot variability (3 reagent lots) and method comparison (versus Siemens Coagulation Factor VIII Deficient Plasma [in combination with Actin FS] reagent on Sysmex CS-5100 analyser) were evaluated by Passing-Bablok and Deming regression, respectively, and Pearson’s r calculated. Assay-specific reference range was determined at one site using fresh plasma samples from apparently healthy individuals aged ≥18 years, not receiving anticoagulant medication.
Results: Across sites, CVs for repeatability ranged 0.9-3.8% for samples with >1.0 IU/dL FVIII activity; SDs ranged 0.0160-0.046 for low activity samples (≤1.0 IU/dL). Among samples with mean activity 0.344-133 IU/dL, high intermediate precision (SD, 0.020 for sample with 0.344 IU/dL activity; CV, 1.8-4.7%) and good total reproducibility (CV, 2.0-13.3%) were observed. The FVIII assay showed excellent lot-to-lot variability (Pearson’s r, 0.999; Table 1) and good agreement with the comparator assay (Pearson’s r, 0.993-0.996; Table 2). Among apparently healthy individuals (n=199), 2.5th and 97.5th percentiles for FVIII activity were 82.2 IU/dL and 218 IU/dL, respectively.
Conclusions: The FVIII assay demonstrated robust analytical performance on the cobas t 711 analyser, supporting use in routine laboratory practice.
|Lot||A vs B||C vs A||B vs C|
|Absolute bias, IU/dL (activity range 0.2-2 IU/dL)||0.062||-0.037||-0.052|
|Absolute bias, IU/dL (activity range 1.5-10 IU/dL)||0.200||-0.040||0.340|
|Relative bias, % (activity range 50-70 IU/dL)||2.13||-1.47||-2.98|
[Table 1. Lot-to-lot comparison for the Roche FVIII assay]
|Absolute bias, IU/dL (activity range 1.5-10 IU/dL)||0.185||-0.390||0.470|
|Relative bias, % (activity range 50-70 IU/dL)||-7.89||-7.78||9.60|
[Table 2. Method comparison of the Roche FVIII assay versus the Siemens Factor VIII assay]
To cite this abstract in AMA style:Lowe A, Kitchen S, Jones R, Geisen U, Rozsnyai G, Jilma P, Quehenberger P. Multicentre Performance Evaluation and Reference Range Determination for a New Assay for the Quantitative Evaluation of Factor VIII in Citrated Plasma [abstract]. Res Pract Thromb Haemost. 2020; 4 (Suppl 1). https://abstracts.isth.org/abstract/multicentre-performance-evaluation-and-reference-range-determination-for-a-new-assay-for-the-quantitative-evaluation-of-factor-viii-in-citrated-plasma/. Accessed November 30, 2021.
« Back to ISTH 2020 Congress
ISTH Congress Abstracts - https://abstracts.isth.org/abstract/multicentre-performance-evaluation-and-reference-range-determination-for-a-new-assay-for-the-quantitative-evaluation-of-factor-viii-in-citrated-plasma/