Premium
Validation of factor VIII activity for monitoring standard and extended half‐life products and correlation to thrombin generation assays
Author(s) -
Augustsson Cecilia,
Norström Eva,
Lind Vivian,
Martin Myriam,
Astermark Jan,
Strandberg Karin
Publication year - 2021
Publication title -
haemophilia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.213
H-Index - 92
eISSN - 1365-2516
pISSN - 1351-8216
DOI - 10.1111/hae.14317
Subject(s) - thrombin generation , chromogenic , thrombin , medicine , multiplex , correlation coefficient , chromatography , urology , chemistry , bioinformatics , platelet , statistics , mathematics , biology
Monitoring replacement therapy with standard and extended half‐life (EHL) products is challenging, since one‐stage assay (OSA) and chromogenic substrate assay (CSA) results may differ significantly. Recent recommendations include local validation of each new product with recovery within 20–30%, depending on activity level. Aim To validate factor VIII (FVIII) activity for monitoring products in clinical use on Atellica Coag and to correlate it with thrombin generation. Methods Plasma samples spiked with Advate ® , Elocta ® , Adynovi ® , Nuwiq ® , NovoEight ® and Afstyla ® (0.05, 0.20, 0.50 and 0.80 IU/ml) were analysed using Atellica Coag 360 with CSA‐1 (Coatest SP) and CSA‐2 (FVIII chromogenic), and OSA (Actin FS). Thrombin generation was performed using two thrombin generation assays (TGA‐1 (Thrombinoscope) and TGA‐2 (Technothrombin). Results All products at levels above 0.05 IU/ml, except Adynovi, showed acceptable recovery using CSA‐1, whereas measurements using CSA‐2 gave more results outside the target level. All products, except Afstyla, showed acceptable recovery using OSA. Correlation between CSA‐1 and OSA was excellent ( r 2 =1.0) with biases of 6–32%, depending on FVIII product. A clear dose‐response was seen for all thrombin generation parameters and products using both methods, except at low levels for lag time using TGA‐1. With CSA‐1 as an independent variable, the correlations to thrombin peak (measured with TGA‐2) were good ( r 2 = .8–.9). Conclusion Our data revealed good correlation and acceptable bias between CSA and OSA using our sets of reagents, methods and analyser in spiked samples. Thrombin generation gave good correlation to CSA‐1 factor activity and is a possible complement to factor activity assays.