ALBERT

Description: Introduction: Adeno-associated virus (AAV)-based factor IX (FIX) gene therapy has the potential to provide clinical benefit in patients with hemophilia B. TAK-748 is a novel next-generation AAV vector for FIX gene therapy. The vector design includes the insertion of 3 hepatocyte-specific cis-regulatory elements to increase the strength of the liver-specific transthyretin promoter driving expression of a human FIX transgene. Aims: The objectives of this study were to investigate the TAK-748 dose-response relationship for FIX activity, and evaluate its efficacy, in FIX knockout (KO) mice and rhesus monkeys. Methods: Male FIX KO mice (N=12/group) received single intravenous doses of TAK-748 (7.4×1010, 1.5×1011, 7.4×1011, or 1.5×1012 vector genomes [vg]/kg) or buffer. Blood samples were taken on days 7, 14, 28, 42, and 56 for the analysis of plasma FIX activity. At the end of the observation period, the bleeding phenotype was assessed by a tail-tip bleeding assay. The viral transduction efficiency of TAK-748 in liver tissue was analyzed by quantitative real-time polymerase chain reaction. Safety assessments included monitoring for clinical signs, and histopathological analysis of selected organs (liver, spleen, kidney, and heart). Male rhesus monkeys (N=3/group) were administered single TAK-748 intravenous bolus injections (3.8×1011, 9.5×1011, or 1.9×1012 vg/kg). Blood samples were collected before dosing and weekly after dosing up to week 18. Plasma FIX activity, human (hu)FIX antigen, and anti-hu-Padua FIX neutralizing antibodies were analyzed. Results: No clinical signs or deaths were recorded in animals treated with TAK-748, and there were no TAK-748-related histopathological findings in the tissues collected from the mice. FIX activity levels in plasma from FIX KO mice treated with buffer were below the lower limit of quantification. Administration of TAK-748 resulted in a dose-dependent increase in mean plasma FIX activity, and supraphysiologic mean FIX levels up to 41.0 IU/mL (1.5×1012 vg/kg). In the tail-tip bleeding assay, blood loss was significantly reduced in the TAK-748 groups at dose levels above 7.4×1010 vg/kg (P

Description: Abstract 2188 25%-30% of patients with hemophilia A develop neutralizing antibodies following replacement therapy with factor VIII (FVIII). These patients can be treated with factor VIIa (FVIIa) which triggers the extrinsic pathway of coagulation and thereby bypasses the requirement for FVIII. We developed a new mouse model that is transgenic for human FVII and expresses specific immune tolerance to native human FVIIa. We aim to investigate the immunological impact of modified FVIIa product candidates and to characterize their immunogenicity by analyzing emerging FVIIa-specific T cell responses. The new mouse model offers a unique opportunity to study central and peripheral immune regulatory mechanisms and the generation of immune responses by pro-inflammatory antigen-specific effector T cells (Teff). We hypothesized that FVIIa-specific Teff having escaped clonal deletion are present in the periphery and may be actively suppressed by FVIIa-specific regulatory T cells (Tregs). To study this hypothesis, we immunized mice with recombinant FVIIa (rFVIIa) with or without LPS, a well-described “danger signal” being able to break immune tolerance by stimulating the innate immune system. Intravenous or subcutaneous administration of rFVIIa alone did not elicit antibody responses and thus immune tolerance to rFVIIa was not broken. However, co-administration of rFVIIa and LPS resulted in a specific antibody response that was not isotypically restricted. To further analyze the mechanisms behind this break of specific immune tolerance, we characterized rFVIIa-specific T cells by the expression of CD154, a marker of antigen-specific T cells. Cytokine production and CD154 expression were assessed upon re-stimulation with rFVIIa. In contrast to mice that were immunized with rFVIIa only, we found increased numbers of rFVIIa-specific T cells in rFVIIa-LPS-treated mice displaying a stable, highly pro-inflammatory (IL-2+/IFN-g+) memory phenotype. These data could suggest that rFVIIa-specific Teff that escaped clonal deletion during induction of central immune tolerance, are present in the periphery of human FVII-transgenic mice. This would imply that rFVIIa-specific Teff could be actively suppressed by Tregs. This suppression could be overcome by danger signals like LPS. We currently study the regulatory mechanisms that maintain tolerance upon administration of FVIIa without LPS. We are approaching this question by correlating the characteristics of FVIIa-specific Teff and Treg responses under both tolerant and non-tolerant conditions. Ultimately, we aim to understand which danger signals have to be provided to break immune tolerance and how tolerance is regulated. Understanding these regulatory mechanisms will enable us to develop new therapeutic strategies and prevent conditions that lead to the induction of antibodies against drug products in patients. Disclosures: Lenk: Baxter BioScience: Employment. Pasztorek:Baxter BioScience: Employment. Weiller:Baxter BioScience: Employment. Ahmad:Baxter BioScience: Employment. Schwarz:Baxter BioScience: Employment. Scheiflinger:Baxter BioScience: Employment. Reipert:Baxter Innovations GmbH: Employment. de la Rosa:Baxter BioScience: Employment.

Description: Abstract 2280 Development of neutralizing antibodies against FVIII is the major complication in the treatment of patients with hemophilia A. Although several genetic and environmental risk factors have been identified, it remains unclear why some patients develop antibodies while others do not. Understanding the underlying mechanisms that drive the decision of the immune system whether or not to make antibodies against FVIII would help to design novel therapeutics. We used a new humanized hemophilic mouse model that expresses the human MHC-class II molecule HLA-DRB1*1501 on the background of a complete knock out of all murine MHC-class II genes. Initial studies had indicated that only a fraction of these mice developed antibodies when intravenously (i.v.) treated with human FVIII. These findings which resemble the situation in patients with severe hemophilia A, evoked the question if the lack of antibody development in non-responder mice reflects the induction of specific immune tolerance after i.v. application of FVIII or represent non-responsiveness for other reasons. We addressed this question by choosing another application route (subcutaneous, s.c.) and by combining i.v. application with a concomitant activation of the innate immune system applying LPS, a well characterized ligand for toll-like receptor 4, together with FVIII. Both strategies resulted in the development of antibodies in all mice included in the study what suggested that non-responsiveness against i.v. FVIII does not reflect an inability to develop antibodies against FVIII. Next, we asked if i.v. FVIII does induce immune tolerance in non-responder mice. We pretreated mice with i.v. FVIII, selected non-responder mice and challenged them with s.c. FVIII. None of the mice developed antibodies what indicated that i.v. pretreatment had induced immune tolerance in non-responder mice. Currently, we test the hypothesis that immune tolerance after i.v. application is induced and maintained by FVIII-specific regulatory T cells. The differences in responder rates after i.v. and s.c. application of FVIII raised the question if there are differences in FVIII T-cell epitopes involved in the initial activation of FVIII-specific CD4+ T cells. We obtained spleen cells from mice treated with either i.v. or s.c. FVIII and generated CD4+ T-cell hybridoma libraries that were tested for peptide specificities. For this purpose we used a FVIII peptide library containing 15 mers with an offset of 3 amino acids. Our results indicate that the pattern of FVIII-specific T-cell epitopes involved in the activation of FVIII-specific CD4+ T cells after i.v. and s.c. application of FVIII is almost identical and represents a small set of FVIII peptides distributed over the A1, A2, B, A3 and C1 domains. Based on our results we conclude that the new HLA-DRB1*1501 hemophilic mouse model represents an interesting opportunity to uncover the mechanisms that drive the decision of the immune system whether or not to develop antibodies against FVIII. Disclosures: Steinitz: Baxter BioScience: Employment. Binder:Baxter BioScience: Employment. Lubich:Baxter BioScience: Employment. Ahmad:Baxter BioScience: Employment. Weiller:Baxter BioScience: Employment. de la Rosa:Baxter BioScience: Employment. Schwarz:Baxter BioScience: Employment. Scheiflinger:Baxter BioScience: Employment. Reipert:Baxter Innovations GmbH: Employment.

Description: Memory B cells are involved in long-term maintenance of antibody-dependent immunologic disorders. Therefore, it is essential to understand how the restimulation of FVIII-specific memory B cells in hemophilia A with FVIII inhibitors is regulated. We asked whether concurrent activation of the innate immune system by an agonist for toll-like receptor (TLR) 7 is able to facilitate the differentiation of FVIII-specific memory B cells in the absence of T-cell help. TLR7 recognizes single-stranded RNA as contained in RNA viruses such as influenza, Sendai, and Coxsackie B viruses. Our results indicate that highly purified murine memory B cells do not differentiate into FVIII-specific antibody-secreting cells in the presence of FVIII and the TLR7 agonist when cultured in the absence of CD4+ T cells. However, CD11c+ dendritic cells facilitate the T cell–independent differentiation of FVIII-specific memory B cells but only in the presence of FVIII and the TLR7 agonist. In contrast to T cell–dependent restimulation, the antibody response after T cell–independent restimulation of FVIII-specific memory B cells is skewed toward IgG2a, an antibody subclass that is efficient in activating the complement system and in inducing Fc-receptor–mediated effector functions, both are required for effective immune responses against pathogens.

Description: Replacement of the missing factor VIII (FVIII) is the current standard of care for patients with hemophilia A. However, the short half-life of FVIII makes frequent treatment necessary. Current efforts focus on the development of longer-acting FVIII concentrates by introducing chemical and genetic modifications to the protein. Any modification of the FVIII protein, however, risks increasing its immunogenic potential to induce neutralizing antibodies (FVIII inhibitors), and this is one of the major complications in current therapy. It would be highly desirable to identify candidates with a high risk for increased immunogenicity before entering clinical development to minimize the risk of exposing patients to such altered FVIII proteins. In the present study, we describe a transgenic mouse line that expresses a human F8 cDNA. This mouse is immunologically tolerant to therapeutic doses of native human FVIII but is able to mount an antibody response when challenged with a modified FVIII protein that possesses altered immunogenic properties. In this situation, immunologic tolerance breaks down and antibodies develop that recognize both the modified and the native human FVIII. The applicability of this new model for preclinical immunogenicity assessment of new FVIII molecules and its potential use for basic research are discussed.

Description: Today it is generally accepted that B cells require cognate interactions with CD4+ T cells to develop high-affinity antibodies against proteins. CD4+ T cells recognize peptides (epitopes) presented by MHC class II molecules that are expressed on antigen-presenting cells. Structural features of both the MHC class II molecule and the peptide determine the specificity of CD4+ T cells that can bind to the MHC class II–peptide complex. We used a new humanized hemophilic mouse model to identify FVIII peptides presented by HLA-DRB1*1501. This model carries a knockout of all murine MHC class II molecules and expresses a chimeric murine-human MHC class II complex that contains the peptide-binding sites of the human HLA-DRB1*1501. When mice were treated with human FVIII, the proportion of mice that developed antibodies depended on the application route of FVIII and the activation state of the innate immune system. We identified 8 FVIII peptide regions that contained CD4+ T-cell epitopes presented by HLA-DRB1*1501 to CD4+ T cells during immune responses against FVIII. CD4+ T-cell responses after intravenous and subcutaneous application of FVIII involved the same immunodominant FVIII epitopes. Interestingly, most of the 8 peptide regions contained promiscuous epitopes that bound to several different HLA-DR proteins in in vitro binding assays.

Description: Background: Pre-existing neutralizing antibodies (NAbs) are a major hurdle in adeno-associated virus 8 (AAV8) gene therapy and result in patients being excluded from clinical trials. Prevalence data published by Kruzik et al. suggest that up to 50% of patients have biologically relevant NAbs against AAV8, excluding them from current clinical trials. We developed a strategy to circumvent pre-existing anti-AAV8 immunity and specifically remove anti-AAV8 antibodies. Aim: Develop an AAV8-specific immune adsorption column (IAC) to specifically remove anti-AAV8 antibodies. Methods: An AAV8-specific IAC was developed. Protocols were optimized to couple AAV8 capsids to activated sepharose. All procedures complied with regulatory requirements to accelerate clinical development. Accompanying mouse studies tested potential concomitant immune suppressive regimens in Bl/6 mice using AAV8-FIX vectors. NAbs were assessed using validated NAb assays of confirmed biological relevance. Titers of AAV8 binding antibodies were assessed by ELISA and T-cell responses by IFN-g ELISpot. Results: Optimized coupling protocols revealed that AAV8 empty capsids can be linked in a stable way to resin. IAC was tested in vitro and in macaques: An IAC could be developed fitting into approved and marketed platforms for immune apheresis. By applying standard experimental procedures for apheresis column development, we showed an IAC can deplete AAV8 binding antibodies and NAbs with greater efficacy than a pan-Ig adsorber (Therasorb-IgOmni). We mimicked treatment of patients in vitro by applying different treatment cycles to plasma reservoirs and showed anti-AAV8 titers of 1:5260 could be depleted to undetectable levels in 4 to 5 treatment cycles. These data are supported by in vivo studies. The IAC is part of a strategy that uses concomitant immunosuppressants to block potential anti-AAV8 T-cell responses. Conclusion: IAC is an enabler for treatment of patients with pre-existing immunity against AAV8 and would also facilitate re-administration. IAC is intended to be applied in combination with Takeda's AAV8 based hemophilia programs. Disclosures Kruzik: Baxalta Innovations GmbH, a Takeda company: Employment. Raim:Baxalta Innovations GmbH, a Takeda company: Employment. Voelkel:Baxalta Innovations GmbH, a Takeda company: Employment. Weiller:Baxalta Innovations GmbH, a Takeda company: Employment. Hoellriegl:Baxalta Innovations GmbH, a Takeda company: Employment, Equity Ownership. Scheiflinger:Baxalta Innovations GmbH, a Takeda company: Employment, Equity Ownership. Rottensteiner:Baxalta Innovations GmbH, a Takeda company: Employment, Equity Ownership. Reipert:Baxalta Innovations GmbH, a Takeda company: Employment, Equity Ownership. de la Rosa:Baxalta Innovations GmbH, a Takeda company: Employment.

Description: Abstract 3323 Baxter is developing a PEGylated recombinant human factor VIII conjugate (BAX 855) based on the modification of full-length FVIII with polyethylene glycol (PEG). The FVIII molecule is derived from a CHO cell line using a plasma-protein-free method. Any chemical modification of FVIII might alter its immunogenic potential. Therefore, careful preclinical immunogenicity studies of PEGylated FVIII in comparison to unmodified FVIII are important. We evaluated the immunogenicity of BAX 855 in comparison to ADVATE, Baxter‘s unmodified recombinant full-length FVIII concentrate. For this purpose, we used different in vitro and in vivo models to s assess the potential impact of BAX 855 and ADVATE on both the innate and the adaptive immune system. The assessment of the potential impact of BAX 855 and ADVATE on the human innate immune system was based on their potential to induce the release of pro- inflammatory cytokines (IL-1β, IL-6, IL-8 and TNF-α) in an in vitro human whole blood assay and on its potential to activate the human complement system in human plasma in vitro. The assessment of the potential impact of BAX 855 and ADVATE on the adaptive immune system was based on their potential to induce antibodies in 3 different hemophilic mouse models and in cynomolgous monkeys. The hemophilic moue models included a hemophilic mouse model with a knockout of the murine FVIII that express a human F8 cDNA as a transgene, a hemophilic mouse model that expresses the human MHC-class II protein HLA-DRB1*1501 on the background of a knockout of the murine MHC-class II complex and a conventional hemophilic mouse model with a knockout of the murine FVIII gene. Hemophilic mice that express a human F8 cDNA as a transgene are immunologically tolerant to human FVIII and develop antibodies against human FVIII only if the immune tolerance breaks down. Using this model, we asked if BAX 855 is able to maintain immune tolerance to human FVIII. The hemophilic mouse models that expresses the human MHC-class II protein HLA-DRB1*1501 mimics the situation in an important fraction of hemophilia A patients with FVIII inhibitors. The human MHC-class II haplotype HLA-DRB1*1501 was previously shown to be associated with an increased risk for patients to develop FVIII inhibitors. Using this model, we asked if BAX 855 expresses an immunogenicity profile similar to ADVATE. BAX 855 and ADAVTE induced similar low levels of cytokine release (IL-1β, IL-6, IL-8 and TNF-α) that were similar to the cytokine release observed in the buffer control group after incubation with human whole blood in vitro. In addition, BAX 855 and ADVATE induced similar low levels of complement activation that were not different from the buffer control group after incubation with human plasma in vitro. Importantly, BAX 855 and ADVATE induced similar levels and incidences of antibodies against FVIII and against PEG-FVIII in all mouse models. In addition, immune tolerance to FVIII was maintained in hemophilic mice expressing the human F8 cDNA. Finally, no major differences in antibody titers were seen after treatment of cynomolgus monkeys with BAX 855 and ADVATE. We conclude that results obtained in comparative immunogenicity studies in vitro and in vivo demonstrate that BAX 855 and ADVATE have a similar immunogenicity profile. Therefore, we expect that BAX 855 will express a similar safety profile as ADVATE in patients. Disclosures: Horling: Baxter Innovations GmbH: Employment. Schwele:Baxter Innovations GmbH: Employment. Lubich:Baxter BioScience: Employment. Ahmad:Baxter Innovations GmbH: Employment. Weiller:Baxter BioScience: Employment. Spatzenegger:Baxter Innovation GmbH: Employment. Schwarz:Baxter Innovations GmbH: Employment. Reipert:Baxter Innovations GmbH: Employment.

Description: Abstract 3327 Baxter has developed a recombinant FVIIa (rFVIIa) product for the treatment of hemophilia patients with factor VIII or factor IX inhibitors. Before entering clinical development we assessed the immunogenic safety profile of the new product candidate BAX 817 using two novel mouse models that mimicked specific aspects of the situation in patients. Several comparative preclinical immunogenicity studies were conducted to assess the immunogenicity profile of Baxter`s rFVIIa BAX 817 in comparison to a licensed recombinant FVIIa product. Three different mouse models were used for this purpose. The first model expresses specific immune tolerance against human FVIIa and, therefore, mimics the situation in both hemophilia A and hemophilia B patients. Using this model, we asked if Baxter‘s rFVIIa is able to maintain immune tolerance to human FVIIa. The second model is a hemophilia A model that mimics the situation in an important fraction of hemophilia A patients with FVIII inhibitors. This model expresses a human MHC-class II haplotype (HLA-DRB1*1501) that was previously shown to be associated with an increased risk for the development of FVIII inhibitors. The third model represents a normal wildtype C57BL/6 mouse. All mice were treated with 8 weekly doses of either Baxter‘s rFVIIa BAX 817 or a licensed recombinant FVIIa product. Total anti-FVIIa antibodies were analyzed prior to the first dose as well as after the 4th and the 8th dose using ELISA assays. Baxter`s rFVIIa BAX 817 and a licensed recombinant FVIIa product induced similar titers of anti-FVIIa antibodies in C57BL/6 wildtype mice and in hemophilic HLA-DRB1*1501 mice. In addition, both Baxter`s rFVIIa BAX 817 and a licensed recombinant FVIIa product were able to maintain specific immune tolerance in a novel mouse model that is immunologically tolerant to human FVIIa. Based on the data obtained we conclude that both Baxter‘s rFVIIa BAX 817 and a licensed recombinant FVIIa product have a comparable immunogenic safety profile. Disclosures: Horling: Baxter Innovations GmbH: Employment. Lenk:Baxter BioScience: Employment. Ahmad:Baxter BioScience: Employment. Weiller:Baxter BioScience: Employment. Schuster:Baxter Innovation GmbH: Employment. Schwarz:Baxter Innovations GmbH: Employment. Reipert:Baxter Innovations GmbH: Employment.