Introduction
Factor VIII activity (FVIII:C) is measured by one‐stage clotting assay (OSA) or chromogenic substrate assay (CSA). Significant differences in FVIII:C between OSA (FVIII:C1st) and CSA (FVIII:CChr) are described as assay discrepancy in nonsevere haemophilia A (HA). A large number of reagent combinations (APTT reagent and FVIII‐deficient plasma) are used for OSA, but the impact of variations in reagent combinations on assay discrepancy has not been fully characterized.
Aim
To clarify the variations in FVIII:C1st/FVIII:CChr ratios according to OSA reagent combination in HA subjects with/without assay discrepancy.
Methods
Thirty‐nine patients previously diagnosed with nonsevere HA were enrolled, and their FVIII genes were investigated and FVIII:C levels were assessed by a single CSA reagent and 11 OSA reagent combinations. Receiver operating characteristic (ROC) curve analysis was used to predict possible cut‐off values of the FVIII:C1st/FVIII:CChr ratio to define FVIII assay discrepancy for each reagent combination.
Results
Patients were categorized into nondiscrepant (n = 25), discrepant (n = 5) and unclassified (n = 9) groups according to their genotypes and information in the database. The FVIII:C1st/FVIII:CChr ratio in nondiscrepant HA varied widely, depending on the APTT reagents and FVIII‐deficient plasma used. The ratio in discrepant HA patients differed with respect to their genotype and the reagent combination used. ROC curve analyses revealed that cut‐off values to distinguish the assay discrepancy differed depending on the reagents used, but revealed two novel genotype variants, p.Cys573Gly and p.Gly582Arg, associated with FVIII assay discrepancy.
Conclusion
Our findings showed that the FVIII:C1st/FVIII:CChr ratio is dependent on the reagent combination used for OSA.