Abstract 2202

Hemophilia B is the X-linked bleeding disorder caused by loss of coagulation factor IX (F.IX) function. Current treatment relies on infusion of plasma derived or recombinant F.IX protein. Approximately 2–4% of hemophilia B patients develop inhibitory antibodies to F.IX protein, and those with F.IX gene deletions are at risk for anaphylaxis. We recently described a murine mode for this pathogenic antibody response, namely C3H/HeJ hemophilia B (HB) mice with a F9 gene deletion (PNAS 107:7101). Anaphylactic reactions were the result of a strong Th2-driven antibody response, comprised of IgG1 and IgE. In new experiments, we again found that repeated exposure to F.IX protein from weekly intravenous injections of recombinant human F.IX (1 IU/mouse) resulted in inhibitors, with 53% of the treated mice experiencing fatal anaphylaxis following the third administration. Surviving animals were given three additional intravenous F.IX protein injections (the fifth and sixth injections were given with antihistamine) and resulted in an average inhibitor titer of 6.2 BU/ml corresponding to an IgG1 titer of 5 mg/ml. Previously, we generated C3H/HeJ HB mice expressing non-functional crm- hF.IX variants due a missense mutation analogous to the UNC-Chapel Hill strain of hemophilia B dog (HB-CH) or due to a stop codon at amino acid residue 338 (HB-338). These mice, in contrast to the gene deletion mutation, failed to develop CD8+ T cell responses to hF.IX, but still formed inhibitors (albeit at reduced titers) upon muscle-directed gene transfer with AAV vector (Mol. Ther 17:1733). Furthermore, HB mice transgenic for a crim+ hF.IX missense mutation (HB-180) were entirely unresponsive to functional hF.IX. When challenged by a total of 6 weekly injections of hF.IX protein, all three strains (HB-338, HB-CH, and HB-180) tolerated the hF.IX antigen without any indication of inhibitors/IgG formation or anaphylaxis, which was in sharp contrast to the gene deletion mice. These results indicate that anaphylaxis in the murine model is F.IX genotype dependent, as it is in humans. In other experiments, naïve gene deletion mice were successfully tolerized to hF.IX by hepatic gene transfer. AAV8 or AAV2 (Y444/500/730F) capsid vectors containing a livers-specific expression cassette were injected into the tail vein. AAV8 vector induced tolerance over a wide range of vector doses. For optimized AAV2, hF.IX levels of 8–14% of normal were achieved with a dose of 2×10^11 vg/mouse, which protected from inhibitor formation and anaphylaxis upon subsequent challenge with F.IX protein. Next, we attempted to reverse the inhibitor response that had formed after protein therapy in gene deletion mice. These were repeatedly treated with hF.IX protein resulting in an average inhibitor of 6 BU/ml. Shortly afterwards, AAV8 vector expressing hF.IX (1×10^11 vg/mouse, n=4) was injected via the tail vein for hepatic gene transfer. Strikingly, this resulted in a complete reversal of antibody titers and systemic hF.IX levels of >40% of normal and aPTTs in the normal range. There was no evidence for an amnestic response, and the animals could be challenged with intravenous hF.IX protein without anaphylaxis. These results demonstrate that inhibitor formation resulting from F.IX replacement therapy can be reversed by liver gene transfer. Moreover, gene therapy in pre-immune mice corrected coagulation and protected from subsequent anaphylactic reactions after gene transfer. We are currently testing whether the protocol is safe for even higher pre-existing inhibitor titers.

Disclosures:

Herzog: Genzyme Corp: Patents & Royalties.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution