Background

For persons with severe hemophilia A receiving exogenous factor VIII (FVIII) replacement therapy, antibodies that interfere with the coagulant function of FVIII (inhibitors) are the most severe complication of treatment. Although avoidance of immunologic “danger signals” during initial FVIII exposure has been proposed as a strategy to reduce the risk of inhibitors, its effectiveness is not proven or reproducible. We have previously reported that the active elimination of danger signals, using the potent anti-inflammatory and immunosuppressive corticosteroid dexamethasone (Dex), during initial exposure to FVIII decreased the anti-FVIII immune response in an immunologically humanized mouse model of severe hemophilia A. We sought to demonstrate the effectiveness of Dex in preventing anti-FVIII immune responses in a larger cohort of mice, and to determine the durability and antigen-specificity of the immunologic tolerance conferred by Dex.

Materials and Methods

Mice with a hemophilia A phenotype due to knockout of exon 17 of the F8gene and with a chimeric human/murine transgene for Major Histocompatibility Complex II allele DRB1*1501 on a C57Bl/S129 background were used for all experiments. (These mice have previously been demonstrated to have an approximately 30% incidence of anti-FVIII IgG antibodies after treatment with recombinant human FVIII). Recombinant human FVIII (Advate, approximately 0.1 mcg/unit) was given by tail vein injection (IV) at 6 units/dose. Dex was given intraperitoneally at 75 mcg/dose. Lipopolysaccharide (LPS) was given IV with FVIII at 2 mcg/dose. Plasma-derived human von Willebrand Factor (VWF, approximately 10 mcg/unit) was given IV at 2 units/dose. Blood samples were collected retro-orbitally or via cardiac puncture, and plasma was obtained by centrifugation with samples frozen at -80C until analysis. Anti-FVIII immunoglobulin G (IgG) and anti-VWF IgG were detected by ELISA, and FVIII inhibitory activity was measured using the Bethesda assay. Statistical comparisons were made using the chi-square or Fischer’s exact test, as appropriate.

Results

In the first experiment, mice were treated with either FVIII or FVIII and Dex (FVIII+Dex) daily x 5, and then sampled 5 weeks later. The incidence of anti-FVIII IgG was higher (21% vs. 6%, p = 0.084) in FVIII mice (n=34) than in FVIII+Dex mice (n=35). IgG negative FVIII mice were then given FVIII daily x 3 (FVIII/FVIII), while IgG negative FVIII+Dex mice were given either FVIII (FVIII+Dex/FVIII) or FVIII with LPS (FVIII+Dex/FVIII+LPS) daily x 3. Mice were sampled 3 weeks later. The incidence of anti-FVIII IgG was higher in FVIII/FVIII mice (n=23) than in FVIII+Dex/FVIII mice (n=15) (52% vs. 7%, p=0.005), but not higher than in FVIII+Dex/FVIII+LPS mice (n=10) (52% vs. 30%, p=0.28). While 8 of 23 FVIII/FVIII mice had a positive Bethesda assay, 0 of 15 FVIII+Dex/FVIII mice had inhibitors (35% vs 0%, p=0.01).

In the second experiment, mice (n=30) had FVIII+Dex daily x 5, and were all anti-FVIII IgG negative when sampled 4 and 14 weeks later. Mice then had either a re-exposure with FVIII daily x 3 (FVIII+Dex/FVIII) 16 weeks after first FVIII exposure, or an intermittent exposure to FVIII one day every 4 weeks followed by FVIII daily x 3 (FVIII+Dex/intFVIII) at 16 weeks. The incidence of anti-FVIII IgG was low (17% overall) and not statistically different between FVIII+Dex/FVIII (n=15) and FVIII+Dex/intFVIII (n=15) mice (7% vs. 27%, p=0.33). Bethesda assays showed a similar pattern. Mice in both groups were then given VWF weekly x 4, and sampled 1 week later: all mice developed anti-VWF IgG, except 1 mouse that had had FVIII+Dex/intFVIII.

Conclusions

When given during initial exposure to FVIII, Dex decreases the incidence of anti-FVIII IgG in an immunologically humanized mouse model of severe hemophilia A. The incidence of new anti-FVIII IgG is low after re-exposure to FVIII, either 6 weeks or 16 weeks after initial exposure. Importantly, Dex-treated mice were able to form antibodies to an unrelated antigen. These results suggest that Dex induces durable immunologic tolerance to FVIII specifically, and that anti-FVIII immune responses are not reduced simply by generalized immunosuppression. Further studies are warranted, both to determine the specific mechanisms by which Dex induces tolerance to FVIII and to investigate the feasibility and effectiveness of using Dex for this purpose in the clinical setting.

Disclosures

Moorehead:Baxter: Honoraria, Membership on an entity's Board of Directors or advisory committees; Bayer: Membership on an entity's Board of Directors or advisory committees; Pfizer: Honoraria. Reipert:Baxter Innovation GmbH: Employment. Steinitz:Baxter: Employment. Hough:Bayer: Research Funding. Lillicrap:Baxter: Research Funding; Bayer: Research Funding; CSL Behring: Research Funding; Biogen Idec: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.