Determination of procoagulant activity in human normal immunoglobulin preparations for therapeutic use by FXIa chromogenic assay: Evaluation of test kit sensitivity, reference standard performance and product formulation effects on the FXIa assay.

Abstract

In 2010, the reporting of thrombotic adverse events for one subcutaneous and certain intravenous immunoglobulins (IGs) raised some concerns. In Europe, regulatory bodies rapidly revised compendial specifications for therapeutic IGs to ensure they do not exhibit thrombogenic (procoagulant) activity (PCA). At the global level, a working group (GWG) was launched with the aim of assessing PCA measurement methods and limits, considering results obtained by human IG manufacturers during in-process controls. The GWG created three dedicated subgroups to investigate the FXIa chromogenic assay, the non-activated partial thromboplastin time (NAPTT) test and the thrombin generation assay (TGA). The European Directorate for the Quality of Medicines & HealthCare (EDQM) was responsible for co-ordinating the subgroup in charge of evaluating the FXIa chromogenic assay in a study that assessed the sensitivity and robustness of two commercial chromogenic FXIa test kits. The impact of IG product formulation on FXIa recovery and the suitability of PCA-containing IG products as potential reference standards/controls were also assessed. IG materials representative of marketed products were provided to four laboratories for a study that was carried out in two steps: 1) two chromogenic FXIa test kit manufacturers assessed the performance and determined optimal test conditions by their respective methods, 2) two OMCLs studied both kits using an optimised study design. Regarding sensitivity, the study results identified suitable dose-response intervals and limits with both chromogenic FXIa test kits. This allowed the establishment of dilution ranges for optimal detection of FXIa/PCA in 5 % and 10 % IG products in the range of 1-6 mIU/mL. However, careful optimisation of the sample dilutions was required (notably to avoid potential matrix effects) and the choice of the mode of data acquisition (kinetic or end-point method) contributed to sensitivity in routine use. Importantly, the composition of IG products was of minor concern for FXIa determination with both test kits. Potential reference materials evaluated in the study behaved as expected and could be useful should a separate reference standard to the FXIa WHO IS be deemed necessary in future.