Abstract 13

Hemophilia A and B result from deficiency in clotting factor VIII (FVIII) or IX (FIX), respectively. In a subset of patients, treatment by factor replacement therapy is limited by formation of inhibitory antibodies to the clotting factors, representing a serious complication that increases risks of morbidity and mortality. Immune responses to the therapeutic coagulation factors are also a concern in newly emerging gene therapies. Regulatory T cells (Treg) offer to be a novel alternative pathway toward immune tolerance. Treg has been identified as a major component of immune tolerance to coagulation factors in pre-clinical studies. Therefore, we hypothesize that ex vivo expanded autologous Treg can suppress inhibitor formation. Our study seeks to test this approach in hemophilic mice. Initially, we optimized in vitro expansion of murine BALB/c-derived Treg. Using flow sorting, GFP+ cells were purified (>98% purity) from spleens of BALB/c knock-in mice containing a GFP reporter linked to FoxP3 expression with an IRES sequence. Sorted cells were stimulated in culture using anti-CD28/anti-CD3 beads in the presence of high-level IL-2 (1000 U/ml). IL-2 was replenished every second day in culture. After ∼1 week, cells were freshly stimulated. At the end of 2 weeks, viability, purity, and FoxP3/GFP expression was confirmed. Greater than 30-fold expansion was repeatedly accomplished. Assumming a dose of 1×106 Treg/mouse, expansion is sufficiently robust to treat >30 mice starting with Treg from 2–3 donor mice. Ex vivo expanded Tregs were adoptively transferred to male hemophilia A mice (BALB/c F8e16 −/−), which were then treated with F.VIII (1 IU human B domain-deleted F.VIII, IV, once per week) for two months. Bethesda assays demonstrated that Treg transplant had effectively suppressed inhibitor formation. Inhibitor titers in control mice were 15–20 BU at 1 month and 30–40 BU at two months. In contrast, Treg treated mice (n=5 per group) formed at most low-titer inhibitors (2–3 BU for both time points). By 2 months, peripheral Treg frequencies had returned to near baseline. To further demonstrate presence of a Treg population capable of suppressing antibody formation against F.VIII, a secondary transfer of sorted CD4+CD25+ splenocytes was performed. Recipient hemophilia A mice were immunized against F.VIII in adjuvant. Compared to mice receiving control Treg, there was significant (P<0.05) suppression of inhibitor formation against F.VIII. In other experiments, Treg therapy was also able to significantly reduce inhibitor titers in hemophilia A mice with pre-existing F.VIII inhibitors. We chose hemophilia A mice (n=6) that developed on average 25 BU from F.VIII replacement therapy. Half of these received Treg transplant, and all mice received 8 more weeks of F.VIII treatment. Inhibitor titers in the control group increased to ∼100 BU. Treg therapy substantially reduced this response (to 15–20 BU, P<0.001). In order to evaluate Treg therapy for hemophilia B, BALB/c F9 −/− × FoxP3-GFP mice were generated. Treg were isolated from these mice as described above, expanded in vitro, and transferred to (BALB/c F9 −/−) hemophilia B mice. Two days later, to test their effectiveness in a gene therapy setting, these mice were treated by intramuscular injection of AAV1 vector expressing human F.IX. By 6 weeks after gene transfer, control mice had formed high-titer antibody against hF.IX (>20 mg IgG/ml plasma, ∼ 10 BU). In contrast, anti-hF.IX formation was undetectable in mice that had received Treg prior to vector administration (n=4/group). While perhaps not as potent as antigen-specific Treg, our data demonstrate the ability of highly purified and ex vivo expanded bulk Treg to control inhibitor formation and thus support their utility for tolerance induction in hemophilia. Their effectiveness may involve emergence of a more specific Treg population after repeated in vivo exposure to antigen. In gene therapy, Treg transplant may be a more desirable alternative to use of immune suppressive drugs. Providing additional immune regulation around the time of vector administration, i.e. when activation signals are provided to the immune system, could be sufficient to prevent immune rejection long-term while inducing antigen-experienced Treg for durable tolerance.


Herzog:Genzyme Corp.: Royalties, AAV-FIX technology, Royalties, AAV-FIX technology Patents & Royalties.

Author notes


Asterisk with author names denotes non-ASH members.