"Redirecting an anti-IL-1β antibody to bind a new, unrelated and computationally predicted epitope on hIL-17A"

Antibody engineering technology is at the forefront of therapeutic antibody development. The primary goal for engineering a therapeutic antibody is the generation of an antibody with a desired specificity, affinity, function, and developability profile. Mature antibodies are considered antigen specific, which may preclude their use as a starting point for antibody engineering. Here, we explore the plasticity of mature antibodies by engineering novel specificity and function to a pre-selected antibody template. Using a small, focused library, we engineered AAL160, an anti-IL-1β antibody, to bind the unrelated antigen IL-17A, with the introduction of seven mutations. The final redesigned antibody, 11.003, retains favorable biophysical properties, binds IL-17A with sub-nanomolar affinity, inhibits IL-17A binding to its cognate receptor and is functional in a cell-based assay. The epitope of the engineered antibody can be computationally predicted based on the sequence of the template antibody, as is confirmed by the crystal structure of the 11.003/IL-17A complex. The structures of the 11.003/IL-17A and the AAL160/IL-1β complexes highlight the contribution of germline residues to the paratopes of both the template and re-designed antibody. This case study suggests that the inherent plasticity of antibodies allows for re-engineering of mature antibodies to new targets, while maintaining desirable developability profiles.

704 The computationally designed human antibody, AU-007, mediates human immune activation by endogenous IL-2, while uniquely breaking the IL-2 auto-inhibitory loop and preventing Treg expansion

Background

IL-2 binds two forms of IL-2 receptor: a high affinity trimeric receptor composed of CD25, CD122, and CD132, and a low affinity dimeric receptor composed of CD122 and CD132. Binding to the dimeric receptors, expressed on effector cells, causes expansion of the effector arm of the immune system including CD8 T-cells, NK-cells and NKT-cells. Binding to the trimeric receptor, expressed on Tregs as well as on pulmonary and vascular epithelium, results in expansion of Treg cells and Vascular Leak Syndrome, both are undesired outcomes of high-dose recombinant IL-2 (Aldesleukin), approved for treatment of Melanoma and Renal-Cell-Carcinoma.

Methods

Flow-cytometry analysis of immune-cell populations of C57BL/6 mice and hPBMCs. Tumor-Growth-Index of murine cancer models.

Results

AU-007, is a computationally designed human antibody that bind the CD25-binding portion on IL-2, preventing binding of IL-2 to the trimeric receptor, but not the dimeric receptor. This leads to immune effector activation while also preventing the Treg expansion via the autoinhibitory loop caused by endogenous IL-2 secreted from activated T effector cells (figure 1). AU-007 binds human IL-2 with picomolar affinity and has excellent biophysical properties with low potential for anti-drug immunogenicity (figure 2). Administration of an AU-007/low dose hIL-2 complex to non-tumor bearing C57BL/6 mice promoted proliferation of effector cells with no effect on Tregs (figure 3). Additionally, an AU-007/low dose hIL-2 complex was highly effective in inhibiting tumor progression in a syngeneic B16F10 melanoma model (figure 4). pSTAT5 analysis of hPBMCs incubated with AU-007 and hIL-2 demonstrated activation of the effector cells and inhibition of Tregs expansion (figure 5). hPBMCs activated with anti-CD3/anti-CD28 and treated with either AU-007 or an isotype control antibody but without exogenous IL-2, showed expansion of effector cells. However, while the isotype control antibody expanded also Tregs , AU-007 inhibited Tregs proliferation, indicating that AU-007 captures endogenous IL-2 and prevents the Treg expansion autoinhibitory loop caused by endogenous IL-2 secreted from activated T effector cells (figure 6).Additionally, following establishment of the IL-2 auto-secretion feedback loop in mice genetically engineered to express hIL-2 instead of murine IL-2, AU-007 treatment significantly inhibited MC38 colorectal-tumor growth for twelve days, in a manner comparable to treatment with anti-PD1 (figure 7).

Conclusions

AU-007 is a human antibody that blocks the CD25-binding epitope on IL-2. It redirects endogenous IL-2 to promote effector cell expansion while simultaneously blocking the Treg expansion autoinhibitory loop, indicating its unique therapeutic profile and high potential as a novel cancer treatment. AU-007 is expected to enter clinical testing in 2021.

Abstract 704 Figure 1

Schematic representation of IL-2 mechanism of action and its dual role in controlling immune response. IL-2 structure consists of three binding epitope sites that interact with different forms of IL-2-R complexes with different affinities (Left Panel). IL-2R complexes expressed on different cell populations and their different affinities to IL-2 allow immunosuppression under conditions of low local concentrations of IL-2 and immune stimulation when IL-2 local concentration rises (middle panel). Au-007 utilize autocrine human IL-2 MOA to promote immune stimulation. Targeting IL-2 to different cell populations can be used to modulate the immune response toward towards immune activation. An anti-human IL-2 antibody tumor clearance while reducing IL-2's undesired interactions with endothelial CD25 expressing cells preventing IL-2 induced pulmonary edema and vascular leaking.

Abstract 704 Figure 2

Au-007 bind human IL-2 with high affinity and inhibits the binding to CD25 while preserving the binding to CD122. Affinity and binding site are demonstrated using Surface Plasmon Resonance. Au-007 was captured on CM5 chip and soluble hIL-2 was injected, forming a complex. Subsequently, soluble CD25 was injected followed by injection of soluble CD122 (A). SPR trace of complex formation of Ab/IL-2/IL-2R arrows indicate where Au-007 (17.069), hIL-2, CD25 and CD122 were injected (B). SPR trace and calculated binding kinetics of chip bound Au-007 with hIL-2 serving as analyte (C). Biophysical profile of Au-007. Au-007 was subjected to five freeze thaw cycles, agitation for 3 days and incubation at 40°C for 1 week. Post treatment Au-007 integrity and indicated biophysical properties were measured (D).

Abstract 704 Figure 3

Au-007 demonstrated in-vivo potent immune stimulating effects in a dose depended manner, with no observed effect on Tregs. C57BL/6 healthy mice were administered daily with Au-007/hIL-2 complex for four days. On day five splenocytes were isolated and immune cells populations were analyzed using flow cytometry. (A) Dosing regimen outline. (B–E) Mean values of immune cells calculated as a percentage from parent population of each experimental group (n=6 per group)

Abstract 704 Figure 4

Au-007 inhibits tumor growth in an I/O resistant tumor model with a tolerable profile. C57BL/6 healthy mice were inoculated with B16F10 melanoma tumor cells (day 0), at day 5 mice were randomized to experimental groups (n=10 per group) and administered daily, with single injection per day of Ab/hIL-2 mix (20 ug/1 ug respectively) or with PBS for four days. From the end of schedule administration at day 8 until experiment endpoint, mice were monitored daily for tumor volume (A) and for mean percent of body weight change for each experimental group (B).

Abstract 704 Figure 5

AU-007 inhibits the effect of IL-2 on Tregs while preserving its effect on Teffs and NKs. (A and B) Phosphorylated STAT5 levels of human immune cell subsets responding to various concentrations of hIL-2 with and without 200 nM AU-007. Total naïve hPBMC culture were incubated with increasing doses of hIL-2 or with increasing doses of hIL-2 + 200nM AU-007 for 15 min. Immune cells subpopulations were analyzed by flow cytometry, gating was defined based on FMOs. (C–F) Phosphorylated STAT5 levels of human immune cell subsets responding to titrated AU-007 or isotype control. Total naïve hPBMC culture were incubated with hIL-2 and with increasing doses of indicated antibody for 15 min. Data presented is an average of 3 biological repeats from 3 human PBMC donors.

Abstract 704 Figure 6

Au-007 can rely on endogenous IL-2 to break auto-inhibitory loop in human PBMCs. Total hPBMCs were stimulated for 24h with anti-CD3/anti-CD28 (stimulation only, green) or stimulated with anti-CD3/anti-CD28 in the presence of: 200 nM of Au-007 mAb (red) or with 200 nM of isotype control mAb (blue). No exogenous IL-2 was added. Immune cells subpopulations were analyzed by flow cytometry. Percentage of immune cell sub-populations demonstrate exclusive inhibition of Tregs (A–E). Au-007 downregulate the suppressive markers of CD4+ regulatory Tregs from panel A, as defined by significant reduction in MFI of CD25 and FoxP3 (F and G). Gating was defined based on FMOs. Data presented is an average of 3 biological repeats from 3 human PBMC donors.

Abstract 704 Figure 7

Au-007 captures endogenous hIL-2 and inhibits tumor growth in colorectal cancer model (MC38). Genetically modified C57BL/6 mice, engineered to express human IL-2 in the background of complete knock-out of mouse IL-2, were inoculated with MC38 colorectal tumor cells. All animals treated with Au-007 showed significant inhibition in tumor growth with no observed significant adverse effects. (A) Administration outline: PBS (black), anti-mouse-PD-1 antibody (yellow), Au-007 pre-complexed with low dose IL-2 (blue) and Au-007 alone every three days followed with a single immune kick start with IL-2 (green, IL-2 single dose is marked in red). (B) Tumor growth progression of the four groups treated. (C) Percent of body weight changes per treatment.

Characterization of BDG8: An antibody-human IL-2 complex that selectively activates the effector arm of the immune system

e15004

Background: Interleukin 2 is a 15.4 kDa type I cytokine of the four helix bundle structure. IL-2 signaling has two opposite effects, it can enhance immune response by activation of effector cells and regulate immune response by activation proliferation of regulatory T cells (Tregs). IL-2 mediates its effect by binding to two forms of IL-2 receptor: i) trimeric receptors made of IL-2Rα (CD25), IL-2Rβ (CD122), and a common IL-2Rγ (γc, CD132) chains or ii) a dimeric receptor of only the IL-2Rβ and IL-2Rγ subunits. In Tregs, activation of the trimeric receptor is associated with FoxP3 mediated transcription leading to the production of suppression factors. Tregs express higher levels of the αβγ trimeric IL-2 receptor than effector cells. However binding of IL-2 to the βγ dimer is associated with activation of effector cells expressing relatively high levels of the βγ dimer and express low levels of trimer. High dose IL-2 therapy is used in melanoma and metastatic renal cell carcinoma with response rates 10%-15%. While efficacious, this approach has several disadvantages: 1) IL-2 dependent adverse effects such as Vascular Leak Syndrome (VLS) excluding many patients from being considered 2) The short half-life of IL-2 requires frequent administration, resulting in repeated spikes in the level of circulating IL-2. 3) Administered IL-2 is not selective and enhances a non-desired activation of Tregs. Methods: Antibodies were computational designed to bind to a specific epitope on Il-2 allowing for the binding to the CD122/CD132 complex and exclude the CD122/CD132/CD25 complex. Results: Here we show that we specifically designed and tested a humanized antibody (BD8) that: i) binds tightly the human IL-2 (hIL-2) and ii) the antibody- hIL-2 complex preferentially to bind the IL-2Rβ sub-unit and blocks the IL-2Rα subunit. We show administering the BD8 Ab-IL-2 complex to C57BL/6 mice has a profound effect on activation of MP CD8+, NK and NKT effector cells with minimal effect on Tregs. In mice the antibody-hIL-2 complex serum half -life is much longer then the half-life of hIL-2. In the B16-F10 Melanoma model, BD8-hIL2 resulted in robust effect compared to hIL-2 alone. Conclusions: BD8 is a computationally designed antibody specifically binds to IL-2 inducing the activation of effector cells without stimulating Tregs and demonstrates efficacy in animal cancer models.