ALBERT

Description: Abstract 3162 Poster Board III-99 Background and objectives Hemophilia A (HA) is a life-threatening hemorrhagic bleeding disorder caused by deficiency of Factor VIII (FVIII). Twenty to thirty percent of HA patients treated by FVIII develop inhibitors. The presence of these inhibitors in high titer (〉5 Bethesda Units (BU)/ml) requires modification of the treatment regimen: use of products that by-pass the action of FVIII for haemorrhagic accidents and treatment protocols by immune tolerance induction (ITI) to eradicate inhibitors. The aim of our study was to enumerate and characterize FVIII-specific memory B-cells in peripheral blood by using polyclonal activation of enriched B-cells and an ELISpot assay in hemophilia patients with inhibitors. Methods Two groups of patients were prospectively included. First group included six severe HA patients treated by ITI; three with inhibitors (mean 3.04 ± 0.67 BU/ml; mean 14.6±4.2 years old) while the three others had a past of inhibitors, successfully treated by ITI (

Description: Abstract 3493 Poster Board III-430 Background and Objectives Antibodies (Abs) directed against Factor VIII (FVIII) remain the main iatrogenic complication in haemophilia A (HA) patients. Anti-FVIII Abs neutralize procoagulant activity of FVIII. They are called inhibitors. Non-neutralizing antibodies (NNA) have also been described in hemophiliacs treated by FVIII concentrates. Their role and prevalence are still under debate. NNA could form immune complexes with FVIII and be responsible of an increased FVIII clearance inducing a shortened half-life. The aim of this retrospective study was to evaluate the prevalence of NNA in a large cohort of HA patients without inhibitors and to determine their epitope specificity against the two chains of FVIII or the B domain, using a x-MAP technology. Indeed, this approach has many advantages: it is fast, requires small volumes of plasma, and is suitable for multiplexing immuno-assay. Methods Samples of 187 patients (mean age 29 years, range [1-86]) with severe (n= 133), moderate (n= 31) or mild (n= 23) HA from three haemophilia centers were studied. All patients have been previously treated and were inhibitor-free at the time of the test (45 yrs] (n=33; 27%) was not statistically different. Conclusion This retrospective study, using the x-MAP technology, found a prevalence of 24% NNA in HA patients, comparable to other reports in the literature. Most of the NNA were directed against the HC of FVIII. In 9% of the cases, NNA recognized the B domain. Further prospective studies will allow to better explain the role of NNA, overall on the FVIII pharmacokinetics and particularly those binding to the B domain. Other prospective studies could highlight a difference according to the type of treatment. Lavigne-Lissalde G. et al. (2008). “Simultaneous detection and epitope mapping of anti-factor VIII antibodies.” Thromb Haemost99(6): 1090-6 Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 1120 Context: The development of alloantibodies (AlloAbs) directed against the factor VIII (FVIII) is actually the main iatrogenic complication in hemophilia A (HA). Immune tolerance induction (ITI) is the only validated treatment to eradicate inhibitors. ITI is based on the infusion of high doses of therapeutic FVIII. It has been recently suggested that epitope specificity during ITI could be related to the outcome of this therapy. Anti-FVIII immune response is polyclonal, complex, dynamic over time and preferentially directed against C2 and A2 domains of FVIII. Tools for fine epitope mapping are useful to identify new epitopes and follow the over time evolution of epitope specificity. The aim of our work was to identify new epitopes on FVIII A2 domain using computer-designed peptides mimicking discontinuous epitopes. We used a multiplex assay based on Luminex™ technology to select the most reactive peptide (1). Patients and methods: All the peptides were predicted by a bioinformatical tool named PEPOP using a new protocol of prediction. Starting from the tridimensional structure of the FVIII, PEPOP were able to work out the sequence of 40 peptides mimicking discontinuous epitopes distributed on the A2 domain surface. Seven negative control peptides were predicted in unaccessible solvant area. A previously published peptide mimicking a linear epitope (residues 484–508) were also synthetized and tested (2). The capacity of peptides to block the binding of the anti-A2 inhibitors on beads coated with the A2 domain obtained by thrombin cleavage of FVIII was assessed. This multiplexed inhibition assay were based on Luminex™ technology. We used a pool of 10 plasma with anti-A2 domain inhibitors as a source of antibodies. Results: The inhibition assays realized with pool of anti-A2 antibodies made it possible to identify 2 peptides mimicking discontinuous epitopes on the A2 domain. The inhibition rate of the two new identified peptides reaches to 50% and 26% at the maximal concentration. We also confirm that the previously published peptide is able to block the binding of anti-A2 inhibitors in our population. The localization of the discontinuous epitopes mimicked by peptides within the tridimensional structure of the A2 domain showed than these epitopes are only represented on a restricted area. Conclusion: We identified two peptides mimicking new discontinuous epitopes on the A2 domain and predicted by a new algorithm of PEPOP. According to the literature, immune response against A2 domain seems to be more restricted. Several residues involved in the new identified discontinuous epitopes are crucial for thrombin, protein S, FIXa and low-density lipoprotein receptor-related protein (LRP) interactions. Tools for fine epitope mapping are essential to identify epitopes on FVIII domains, with a particular concern for discontinuous epitopes. Further studies are needed to validate the concept that fine epitope mapping at the epitope level could be useful to predict ITI outcome. Disclosures: No relevant conflicts of interest to declare.

Description: Background and objectives: Acquired hemophilia (AH) is a severe life-threatening autoimmune disease due to the development of polyclonal autoantibodies (autoAbs) which neutralize the procoagulant activity of coagulation factor VIII (FVIII). The most common epitopes for autoAbs to FVIII are located in the A2, A3 and C2 domains of FVIII. We have recently developed a multiplex assay based on the Luminex technology that allows simultaneously the detection and epitope mapping of anti-FVIII antibodies (Lavigne-Lissalde et al, Thromb; Haemost. 2008). This technology is fast and requires only 100 μl of plasma and we showed that most autoAbs to FVIII are monospecific and preferentially bind to the light chain (LC) of FVIII. The aim of the present study was to evaluate the epitope specificity and IgG isotype/subclasses of autoAbs to FVIII detected in the post partum period. Methods: Acquired hemophilia was diagnosed in six women (median age: 28 years) included in a French cohort (SACHA) and who exhibited prolonged activated partial thromboplastin time, low FVIII levels (range = 0.5–21 IU/dL) and inhibitor titers between 1.5 and 40 Bethesda units (BU). The epitope mapping was studied in each patient by testing 5 plasma samples collected (at inclusion and after 1, 3, 6 and 12 months of follow up) using a multiplex assay as previously described (Lavigne-Lissalde et al, Thromb; Haemost. 2008). A panel of monoclonal antibodies that bind to LC, heavy chain (HC) and the C2 domain of FVIII was employed and the use of specific immunoconjugates also allowed to define the isotype (IgG, A, M) and subclass (IgG1 to IgG4) distributions of autoAbs to FVIII. Results were expressed in mean fluorescence intensity (MFI). Results: The autoAbs detected in the post partum in the six women studied recognized both and simultaneously HC and LC with predominant epitopes located in the LC and the C2 domain in 5 cases. At the inclusion, the autoAbs were mixtures of IgG1 and IgG4 in all patients, but the latter subclass was predominant in most cases. In addition, IgM Abs were also present in 3 women. No correlation between the titer of inhibitor (in BU) and the level of autoAbs binding (in MFI) could be evidenced. The evolution was good in all patients but one case for whom an increase in IgG4 binding to the FVIII C2 domain was demonstrated whereas inhibitor titer decreased (from 40 to 2.6 BU). Importantly, no modification in the epitope specificity between inclusion and 12 months of follow-up was found. Conclusion: The study shows that autoAbs to FVIII developed in the context of pregnancy are mainly IgG4 specific to the C2 domain. In addition, the epitope mapping of anti-FVIII antibodies by multiplex approach could contribute to a better understanding of the pathophysiology of acquired hemophilia.