Characterization of a Trispecific PD-L1 Blocking Antibody That Exhibits EGFR-Conditional 4-1BB Agonist Activity

Abstract

Immune checkpoint blockade has changed the treatment paradigm for advanced solid tumors, but the overall response rates are still limited. The combination of checkpoint blockade with anti-4-1BB antibodies to stimulate tumor-infiltrating T cells has shown anti-tumor activity in human trials. However, the further clinical development of these antibodies has been hampered by significant off-tumor toxicities. Here, we generated an anti-4-1BB/EGFR/PD-L1 trispecific antibody consisting of a triple-targeting tandem trimerbody (TT) fused to an engineered silent Fc region. This antibody (IgTT-4E1-S) was designed to combine the blockade of the PD-L1/PD-1 axis with conditional 4-1BB costimulation specifically confined to the tumor microenvironment (TME). The antibody demonstrated simultaneous binding to purified EGFR, PD-L1, and 4-1BB in solution, effective blockade of the PD-L1/PD1 interaction, and potent 4-1BB-mediated costimulation, but only in the presence of EGFR-expressing cells. These results demonstrate the feasibility of IgTT-4E1-S specifically blocking the PD-L1/PD-1 axis and inducing EGFR-conditional 4-1BB agonist activity.

Keywords: 4-1BB costimulation; cancer immunotherapy; epithelial growth factor receptor; immune checkpoint blockade; trispecific antibody.

Figure 1

Gene construct, molecular model, and binding characteristics of the IgTT-4E1 antibody. (a) Gene layout of the IgTT-4E1 antibody bearing a signal peptide from oncostatin M (white box), one anti-4-1BB scFv (magenta boxes), one anti-EGFR V_HH (green box), one anti-PD-L1 V_HH (blue box), three collagen-derived trimerization (TIE) domains (yellow boxes) flanked by peptide linkers, and the Fc-encoding element (gray box). N-terminal FLAG-Strep and C-terminal Myc-His tags (orange boxes) were appended for purification and immunodetection purposes. (b) Schematic diagram showing the three-dimensional model of the anti-4-1BB/EGFR/PD-L1 tandem trimerbody (TT). (c) Molecular diagram of the IgTT-4E1 antibody and its three-dimensional modelizations, in front and top views. (d) Human EGFR-Fc (EGFR) was immobilized onto four different biosensors; three biosensors were incubated in 20 nM IgTT-4E1 for 20 min (red trace), while the fourth was kept in HBS as a negative control (gray trace). Then, 50 nM human PD-L1-Fc (PD-L1) was added to two of the IgTT-4E1-loaded biosensors (blue trace) and the control biosensor for 20 min. Also, 50 nM human 4-1BB-hFc (4-1BB) was then added to one PD-L1- and IgTT-4E1-treated biosensor (black trace) and the control biosensor. The cumulative signal increases on the biosensor receiving IgTT-4E1, PD-L1, and 4-1BB demonstrate that all three antigens can be bound by the IgTT-4E1 antibody simultaneously. (e) The binding to human EGFR, PD-L1, and 4-1BB on the cell surface of 3T3, 3T3^EGFR, CHO, CHO^EGFR, CHO^PD-L1, Jurkat, and Jurkat^4-1BB cells by rituximab (rtx), cetuximab (ctx), atezolizumab (atz), urelumab (url), and IgTT-4E1 at 6.67 nM was measured by FACS. Cells incubated with PE-conjugated GAH antibody (GAH-PE) are shown as non-filled histogram. The y-axis shows the relative cell number, and the x-axis represents the intensity of fluorescence expressed on a logarithmic scale.

Figure 2

Effect of IgTT-4E1-S on PD-L1/PD-1 blockade, 4-1BB costimulation, and IFNγ secretion. (a) ADCC reporter bioassay response to rituximab (rtx), IgTT-4E1, and IgTT-4E1-S, using ADCC bioassay effector Jurkat^NFAT-CD16 cells co-cultured with 3T3 or 3T3^EGFR target cells at a 6:1 E/T ratio for 6 h at 37 °C. After incubation, Bio-Glo™ Luciferase Assay Reagent was added for luminescence determination. Data shown represent the mean ± standard deviation of triplicates. Data are presented as the mean ± SD (n = 3). Significance was determined by two-way ANOVA with Tukey’s multiple comparison test. (b) PD-L1/PD-1 blockade bioassay assesses the inhibitory activities of IgTT-4E1-S. Y-axis represents reporter gene fold induction. Cetuximab (ctx) was used as a negative control, and atezolizumab (atz) as a positive control. Fold induction relative to negative control (ctx)-incubated cells. Results are expressed as mean ± SD (n = 3). Significance was measured by one-way ANOVA with Dunnett’s multiple comparison test. (c) Costimulation of Jurkat^NFkB-4-1BB cells co-cultured with CHO cells, CHO^EGFR, or CHO^PD-L1 cells in the presence of 10-fold increasing concentrations of IgTT-4E1-S, urelumab (url), and rituximab (rtx) antibodies for 6 h at 37 °C. After incubation, luminescence was determined. Data are presented as fold induction relative to the values obtained from unstimulated Jurkat^NFkB-4-1BB cells. Representative dose–concentration curves are presented and expressed as mean ± SD (n = 3). Significance was determined by two-way ANOVA with Tukey’s multiple comparison test. (d) Flow cytometry analysis of EGFR and PD-L1 expression in MDA-MB-231 cancer cells. Cells incubated with PE-conjugated and APC-conjugated isotypes are shown as gray-filled histogram. (e) Flow cytometry analysis of PD-1 and 4-1BB expression in PBMCs preactivated with anti-CD3 mAb for 24 h. PBMCs incubated with FITC-conjugated and PE-conjugated isotypes are shown as gray-filled histogram. (f) Anti-CD3 preactivated PBMCs were co-cultured with MDA-MB231 target cells at effector/target (E/T) ratios of 1:1 and 2:1. Control human polyclonal IgG (control hIgG), urelumab (url), atezolizumab (atz), and IgTT-4E1-S were added at 6.67 nM. The IFNγ concentrations in supernatant after 48 h were analyzed and expressed as mean ± SD (n = 3). Significance was determined by two-way ANOVA with Tukey’s multiple comparison test.