ALBERT

Description: Background & Aim: T-cell bispecific antibodies (TCBs) binding to a target on tumor cells and CD3 on T cells induce potent T-cell mediated killing of cells carrying the target. In contrast to targets like e.g. CD38 or CD138, B-cell maturation antigen (BCMA) is suggested to be only expressed on plasma cells (PCs) and multiple myeloma (MM) PCs. Therefore, a BCMA-TCB should specifically act on these cell types. We report on a new class of BCMA-TCBs designed for effective and convenient therapy of MM. Molecular structure & its rationale (fig. 1A): The new class of BCMA-TCBs are asymmetric two-arm IgG-based human antibodies, bivalent to BCMA, monovalent to CD3 and comprising an engineered Fc. To achieve tumor specificity of the BCMA-TCBs and avoid unspecific T-cell activation, monovalent binding to CD3 was fixed at an affinity of KD= 70 nM, whereas various BCMA-TCBs with binding affinities to BCMA ranging from KD= 50 pM to 10 nM have been investigated (measurement of binding affinities by surface plasmon resonance and flow cytometry (FC)). For long elimination half-life, an Fc was introduced to enable once a week intravenous or subcutaneous administration. Fc is engineered to abolish binding to FcgR and C1q to minimize risk of infusion reactions without interfering with FcRn binding. The BCMA-TCBs can be well-manufactured and have no tendency to aggregation. In vitro profile: Potency of BCMA-TCBs to activate T cells and to induce killing of human MM cell lines was measured in a 24h co-culture assay with human PBMCs and MM cells in a ratio of 10:1. Results: (i) NCI-H929 cells expressing on cell surface up to 100 times more BCMA than primary patient MM cells were killed in presence of BCMA-TCBs with EC50 ranging from 5 to 50 pM, but not in presence of a control-TCB binding to CD3 only. (ii) As next, RPMI-8226 cells were used as target cells because surface BCMA expression was found to be only slightly higher than on primary patient MM cells (specific antigen binding capacity SABC as measured by FC between 1165 and 5461 per RPMI-8226 cell compared to 116 to 4479 per primary patient MM cell). Effective killing of RPMI-8226 MM cells was observed; the killing potency was higher respectively EC50 lower with the BCMA-TCBs having binding affinities below 1 nM (fig. 1B; EC50 from 50 pM to 1000 pM). (iii) Killing of U266 and L363 human MM cell lines was also observed with BCMA-TCBs. (iv) T-cell activation and increased T-cell function go in parallel with lysis of MM target cells as observed by an upregulation of CD69 and CD25 expression, release of granzyme B (〉20 ng/mL at 3 nM vs. 20 pg/mL for control) and proinflammatory cytokines e.g. IFN-g, TNF-a, IL-2, IL-6, and IL-10. (v) At 120h incubation, BCMA-TCBs induced concentration dependant CD4 and CD8 T-cell proliferation as observed by CFSE dilution. A small exploratory study in whole bone marrow (BM) aspirates from MM patients (n=3) suggested a concentration dependent killing of MM PCs induced by BCMA-TCBs in presence of autologous T cells, thus justifying for a much larger trial to investigate one of the BCMA-TCB of this new class to induce killing and T-cell activation/function in MM patient BM aspirates performed by two clinical groups (Abstract also submitted to ASH). In vivo profile: New BCMA-TCBs bind to BCMA as well as CD3 of cynomolgus monkeys (cyno). Single dose pharmacokinetics (PK) and pharmacodynamics of a BCMA-TCB was studied in cyno at 0.003, 0.03, 0.3 mg/kg intravenously. Dose linear PK was observed. Mean serum concentrations in the first 7 days after 0.03 mg/kg were ≈1 to 2 nM, BM aspirates collected at 96h showed also concentrations of ≈1 to 2 nM. Elimination between 24 h and 504 h was found to be first order with a half-life of approximately 6 to 8 days. These data suggest a convenient clinical dosing schedule, e.g. once a week. Peripheral blood T-cell redistribution was observed 24h after dosing. Reduction of PCs could be observed by FC while total B cells and other cell types were unaffected. Summary: Effective and specific killing of MM cells was demonstrated with the BCMA-TCBs. Killing goes in parallel with T-cell activation and increased function. TCB with binding affinity to BCMA below 1 nM have higher potency/lower EC50 than those with affinity above 1 nM. TCBs could be produced with high purity and were stable with no tendency to aggregation. In cynomolgus monkeys, a PK profile has been found, suitable for once a week administration. This new class of BCMA-TCB has promises for clinical development. Disclosures Vu: EngMab AG: Employment, Equity Ownership, Membership on an entity's Board of Directors or advisory committees. Moser:Roche: Employment. Delon:Roche: Employment. Latzko:Roche: Employment. Gianotti:Roche: Employment. Lüoend:Roche: Employment. Friang:Roche: Employment. Murr:Roche: Employment. Duerner:Roche: Employment. Weinzierl:Roche: Employment. Fauti:Roche: Employment. Bacac:Roche: Employment. Ast:Roche: Employment. Freimoser-Grundschober:Roche: Employment. Rodriguez Diaz:Roche: Employment. Zielonka:Roche: Employment. van Puijenbroek:Roche: Employment. Hosse:Roche: Employment. Bruenker:Roche: Employment. Mössner:Roche: Employment. Klein:Roche: Employment. Umaña:Roche: Employment. Strein:EngMab AG: Employment, Equity Ownership, Membership on an entity's Board of Directors or advisory committees; BB Biotech AG: Membership on an entity's Board of Directors or advisory committees; Novimmune SA: Membership on an entity's Board of Directors or advisory committees.

Description: GA101 is a novel humanized and glycoengineered CD20 antibody that was derived by humanization of the parental B-Ly1 mouse antibody and subsequent glycoengineering leading to the following characteristics: high affinity binding to the CD20 type II epitope; low complement dependent cytotoxicity (CDC) activity; high direct cell death induction; and high antibody-dependent cellular cytotoxicity (ADCC) potency. In vivo studies examined the consequences of GA101 on the growth of aggressive NHL xenograft models in SCID beige mice and the subsequent effects on survival. These studies demonstrated an outstanding dose-dependent anti-tumor activity of GA101. In particular, GA101 mediated superior anti-tumor efficacy in staged s.c. Z138 mantle cell lymphoma (MCL) and SU-DHL4 diffuse large B cell lymphoma (DLBCL) xenograft models in direct comparison to rituximab. Weekly dosing of GA101 at 10 mg/kg (Z138 model) or 30 mg/kg (SU-DHL4) was able to induce complete tumor remission and long-term survival (cures) in all animals. Consistent anti-tumor efficacy against established tumors was also observed in a number of additional NHL models including Raji Burkitt’s lymphoma or OCI-LY18 DLBCL models. In addition GA101 was evaluated in an orthotopic disseminated Z138 model. After i.v. injection of Z138 cells SCID beige mice developed large intraperitoneal lymphoid tumors in the ovary. Several experiments showed that GA101 mediated increased overall and median survival in this model in direct comparison to rituximab. All xenograft studies were performed in SCID beige mice that are incompetent for NK-mediated ADCC. Indeed, control experiments with non-glycoengineered GA101 in s.c. xenograft models suggest that the superiority of GA101 does not depend on enhanced interactions of the engineered Fc region with murine effector cells such as macrophages/monocytes or granulocytes. Thus, the anti-tumor effects of GA101 in s.c. xenograft models can be primarily attributed to non-effector cell mediated activities. Currently, experiments are ongoing to evaluate the efficacy of GA101 in combination with classical chemotherapy regimens and novel targeted agents. First results from these studies will be presented during the 2007 ASH Annual Meeting. In summary it can be assumed that the combination of the recognition of a type II epitope together with improved ADCC potency might translate into superior efficacy in the clinical treatment of CD20 positive malignant diseases.

Description: Introduction IL-2 therapy can lead to durable responses in cancer patients, but is associated with significant toxicity. None of the known IL-2-based immunocytokines has yet progressed to pivotal clinical trials due to various constraints in their design; in particular, the fusion of two wild-type IL-2 moieties to the antibody and retained FcgR binding of IgG-based immunocytokines. This design results in 1) high affinity binding with pM affinity to IL-2Raβγ on immune cells compromising tumor targeting and inducing rapid systemic clearance and short half-life; 2) high affinity for CD25 (IL-2Ra) expressed on pulmonary vascular endothelium contributing to pulmonary toxicity; and 3) preferential activation of Tregs over immune effectors. Here we describe a novel class of monomeric tumor-targeted immunocytokines where a single, engineered IL-2 variant (IL2v) with abolished CD25 binding is fused to the C-terminus of an antibody with a heterodimeric Fc-part. FcγR and C1q binding is completely abolished by a novel Fc mutation. For tumor targeting, human(-ized) high affinity antibodies against CEA (GA504, CEA-IL2v) or FAP (GA501, FAP-IL2v) were selected. CEA-IL2v recognizes a membrane proximal epitope of human carcinoembryonic antigen (CEA) and binds preferentially to membrane-bound CEA, but not shed CEA. Methods CEA- and FAP-IL2v were produced as recombinant proteins and their activity tested oneffector cells by assessing the activation of P-STAT5, cell proliferation, sensitivity to Fas-induced apoptosis, expression of activation markers and cytokine release upon treatment. Safety, pharmacokinetics (PK), pharmacodynamics (PD) and anti-tumor efficacy were analyzed in SCID and fully immunocompetent C57Bl/6 mice as single agent and in combination with trastuzumab and cetuximab. Tumor targeting was investigated in the orthotopic syngeneic Renca renal cell cancer tumor model in Balb/c mice by SPECT imaging. Results FAP- and CEA-IL2v completely lack binding to CD25, but retain IL-Rβγ binding, and show pM binding affinity to respective antigens, FAP on fibroblasts and CEA on tumor cells. As consequence of abolished binding to CD25 these molecules do not preferentially activate Tregs. The treatment of effector cells with IL2v reduces their sensitivity for Fas-mediated apoptosis (also known as activation induced cell death) as compared to wild-type IL-2 based immunocytokine. IL-2Rβγ bioactivity was retained and FAP- and CEA-IL2v activate NK, CD4+ and CD8+ T cells as shown by induction of activation markers, cell proliferation and cytokine release. Furthermore, CEA-IL2v and FAP-IL2v enhanced the cytotoxic activity of NK cells when combined with ADCC-competent antibodies. Mechanism of action studies in fully immunocompetent mice showed that the molecules strongly expand and activate NK, CD8+ T cells and gd T cells (up to 100-fold) and skew the CD4:CD8 ratio strongly towards CD8+ T cells in the peripheral blood, lymphoid tissues, and in the tumor. In C57Bl/6 mice, CEA- and FAP-IL2v demonstrate improved safety despite a higher exposure and circulatory half-life than the analogous IL-2 based immunocytokine. MicroSPECT/CT imaging with radioactively labeled FAP-IL2v revealed good FAP-mediated tumor targeting in the orthotopic syngeneic Renca model with low normal tissue uptake and low accumulation in lymphoid tissues, contrary to analogous IL-2 based immunocytokine that showed preferential targeting to lymphoid tissue. Studies in tumor-bearing mice showed dose-dependent anti-tumor efficacy of FAP-IL2v and CEA-IL2v in syngeneic models. Additional studies in xenograft models in SCID mice transgenic for human CD16A showed that CEA-IL2v strongly enhances the antitumor efficacy and/or survival mediated by ADCC-competentantibodies, including trastuzumab and cetuximab. Conclusion CEA- and FAP-IL2v demonstrate superior safety, PK and tumor targeting, while lacking preferential induction of Tregs due to abolished CD25 binding, monovalency and high-affinity tumor-targeting as compared to classical IL-2-based immunocytokines. They retain capacity to activate and expand NK and CD8+ effector T cells through IL-2Rβγ in the periphery and the tumor microenvironment. These data support their further nonclinical and clinical investigation for immunotherapy of cancer. Clinical trials with CEA-IL2v are foreseen in 2014. Disclosures: Klein: Roche Glycart AG: Employment. Waldhauer:Roche: Employment. Nicolini:Roche: Employment. Dunn:Roche: Employment. Freimoser-Grundschober:Roche: Employment. Danny:Roche: Research Funding. Boerman:Roche: Research Funding. Nayak:Roche: Employment. Herter:Roche: Employment. Van Puijenbroek:Roche: Employment. Ast:Roche: Employment. Hofer:Roche: Employment. Hosse:Roche: Employment. Lang:Roche: Employment. Neumann:Roche: Employment. Kettenberger:Roche: Employment. Neubauer:Roche: Employment. Gorr:Roche: Employment. Tuerck:Roche: Employment. Evers:Roche: Employment. Gerdes:Roche: Employment. Levitsky:Roche: Employment. Bacac:Roche: Employment. Moessner:Roche: Employment. Umana:Roche: Employment, Equity Ownership.

Description: Abstract 3925 GA101 is Type II, glycoengineered CD20 monoclonal antibody currently in PhII/III clinical trials. We have previously shown that GA101 mediates superior in vitro and in vivo activity compared to the Type I CD20 antibody rituximab. By epitope mapping and crystallography we have shown that GA101 recognizes CD20 in a unique way that is different from Type I CD20 antibodies and have proposed that this may be the basis for the Type II character of GA101. Here we compare for the first time GA101 with rituximab, the standard of care in various clinical settings in NHL and B-CLL in combination with chemotherapy, as well as with the Type I CD20 antibody ofatumumab, which was recently approved for treatment of B-CLL patients refractory to fludarabine and alemtuzumab. The following assays were used to compare the three anti-CD20 antibodies: i) Binding to NHL cell lines Z138 (MCL, ca. 60.000 CD20 binding sites per cell) and SU-DHL4 (DLBCL, ca. 1 Mio CD20 binding sites per cell) assessed by FACS, ii) Cell death induction, detected by AxV/PI staining and FACS, on a panel of NHL cell lines, iii) Antibody dependent cellular cytotoxicity mediated by PBMNCs as effector and Z138, SU-DHL4 as target cells (ADCC, LDH release assay); iv) Complement dependent cytotoxicity with Z138, SU-DHL4 as target cells (CDC, LDH release assay) and v) B-cell depletion (assessed by FACS) in whole blood from healthy donors. Dose-dependent anti-tumoral activity was assessed in a s.c. SU-DHL4 NHL xenograft model in Scid beige mice. Survival experiments in a disseminated Z138 MCL model are ongoing and an update on the results will be included as part of the poster presentation. Ofatumumab (“Arzerra”) was purchased from a local pharmacy, GA101 and rituximab were obtained from Hoffmann La Roche AG, Basel. First, binding studies confirmed that GA101 shows half-maximal binding to NHL cells relative to rituximab and ofatumumab, a known property of Type II CD20 antibodies. EC50 values of binding were comparable indicating that GA101, rituximab and ofatumumab have apparent binding affinities in the low nanomolar range on NHL cells independent of the level of CD20 expression. Second, the three CD20 antibodies were compared for their induction of direct cell death as measured by AxV/PI staining. Overall, GA101 mediated superior direct cell death induction compared to rituximab and ofatumumab utilizing a panel of NHL cell lines of different origins. Immune effector-related mechanisms of action were subsequently compared by ADCC and CDC assays. GA101, a glycoengineered antibody with enhanced affinity for FcgRIIIa, was found to exhibit up to 100-fold higher ADCC potency than rituximab and ofatumumab on Z138 and SU-DHL4 cells. CDC, as expected for a Type II CD20 antibody was ca. 10 to 1,000 less potent compared to the Type I antibodies rituximab and ofatumumab. In order to integrate the different mechanisms of action (direct cell death, ADCC, CDC), autologous ex vivo B-cell depletion assays with whole blood from healthy donors containing natural immune effector cells, human complement and physiological concentrations of human immunoglobulins were performed. These studies showed that GA101 was more potent in terms of EC50 values and more efficacious in terms of absolute B-cell depletion when compared to rituximab and ofatumumab. Finally, the dose-dependent effects of the three CD20 antibodies was studied on the growth of s.c. SU-DHL4 DLBCL xenografts in SCID beige mice. GA101 induced a dose-dependent anti-tumoral effect including complete tumor remission and was superior to the Type I antibodies rituximab and ofatumumab at saturating antibody doses. In summary, the preclinical data presented herein demonstrate that the Type II, glycoengineered CD20 antibody GA101 is differentiated from the Type I CD20 antibodies rituximab and ofatumumab by its superior overall activity supporting its further clinical investigation. Of note, in contrast to previous publications, in this series of assays no superior preclinical activity of ofatumumab was observed when compared to rituximab. Disclosures: Herter: Roche: Employment, Patents & Royalties. Waldhauer:Roche: Employment. Otz:Roche: Employment. Herting:Roche: Employment, Patents & Royalties. Lang:Roche: Employment. Nicolini:Roche: Employment. Römmele:Roche: Employment. Friess:Roche: Employment, Patents & Royalties. Van Puijenbroek:Roche: Employment. Bacac:Roche: Employment. Weidner:Roche: Employment, Equity Ownership. Gerdes:Roche: Employment, Equity Ownership, Patents & Royalties. Umana:Roche: Employment, Equity Ownership, Patents & Royalties. Klein:Roche: Employment, Equity Ownership, Patents & Royalties.

Description: CD20 is an important target for the treatment of B-cell malignancies, including non-Hodgkin lymphoma as well as autoimmune disorders. B-cell depletion therapy using monoclonal antibodies against CD20, such as rituximab, has revolutionized the treatment of these disorders, greatly improving overall survival in patients. Here, we report the development of GA101 as the first Fc-engineered, type II humanized IgG1 antibody against CD20. Relative to rituximab, GA101 has increased direct and immune effector cell-mediated cytotoxicity and exhibits superior activity in cellular assays and whole blood B-cell depletion assays. In human lymphoma xenograft models, GA101 exhibits superior antitumor activity, resulting in the induction of complete tumor remission and increased overall survival. In nonhuman primates, GA101 demonstrates superior B cell–depleting activity in lymphoid tissue, including in lymph nodes and spleen. Taken together, these results provide compelling evidence for the development of GA101 as a promising new therapy for the treatment of B-cell disorders.