The Prevalence of Anti-Factor VIII Antibodies in the Hemophilia A Population of the Quebec Province of Canada: Results of Testing with the Bethesda/Nijmegen Assay and with Three ELISA Assays.

The development of anti-Factor VIII antibodies (FVIII Abs) is the most serious complication of the treatment of hemophilia A. The prevalence and incidence of FVIII Abs is usually assessed by the functional Bethesda/Nijmegen assay. We hypothesized that some FVIII Abs might not be detected by the Bethesda/Nijmegen assay but would be apparent in an ELISA assay. The FVIII Ab ELISA may be more sensitive or it may be detecting Abs against both functional and non-functional epitopes of FVIII.

In the Province of Quebec (population 7.8 × 10⁶), all subjects with hemophilia must be registered with one of the four designated hemophilia centers to be allowed to be treated with FVIII concentrates. Over the last year, blood specimens were obtained from all registered severe hemophilia A subjects to be tested for FVIII Abs. Citrated plasma specimens obtained more than 48 hours after the last FVIII treatment were tested with the Bethesda/Nijmegen assay and three described ELISAs (Haemophilia 2009;15:374-6) using as the coating antigen two different full-length recombinant FVIII (FLRFVIII) concentrates, Helixate^® FS and Advate^® respectively, and a B domain-deleted recombinant FVIII (BDDRFVIII) concentrate, Xyntha^®. Six normal plasmas were used as negative controls on each ELISA plate. Mean and standard deviation (SD) of absorbance were calculated for the total of all the plates used for each of the three coating antigens. Results were considered positive with Bethesda unit (BU) ≥ 0.4 /mL and ELISA absorbance ≥ 3 standard deviations (SD) of the mean of the normal plasmas.

At time of writing this abstract ≥ 80% of the target subjects have been tested and the remaining are being tested. Twelve out of 114 (10.5 %) are positive with the Bethesda/Nijmegen assay. Eleven out of these 12 are ELISA positive. Ten out of the 102 Bethesda negative subjects (9.8 %) are positive for the FLRFVIII ELISAs, all of them being negative for the BDDRFVIII ELISA. The titre of FVIII Abs measured by the Bethesda assay was highly correlated (R=0.93) with the titre measured with the three ELISAs.

Our observed prevalence (10.5%) of Bethesda positive subjects is comparable with values reported in similar unselected severe hemophilia A populations. There is no published literature with which to compare our observed prevalence of 9.8% of ELISA positive amongst Bethesda negative subjects. Bethesda negative plasmas that are positive for the FLRFVIII ELISAs and negative for the BDDRFVIII ELISA are presumed to have FVIII Abs directed against the B domain of the FLRFVIII concentrates. The clinical significance of this observation is presently unknown but is being investigated with pharmacokinetic studies.

No relevant conflicts of interest to declare.