A bispecific anti-PD-1 and PD-L1 antibody induces PD-1 cleavage and provides enhanced anti-tumor activity

Abstract

Combinatorial strategies, such as targeting different immune checkpoint receptors, hold promise to increase the breadth and duration of the response to cancer therapy. Here we describe the preclinical evaluation of CTX-8371, a protein construct which combines PD-1 and PD-L1 targeting in one bispecific, tetravalent antibody. CTX-8371 matched or surpassed the activity of anti-PD-1 and PD-L1 benchmark antibodies in several in vitro T cell activation assays and outperformed clinically approved benchmarks in the subcutaneous MC38 colon and the B16F10 lung metastasis mouse tumor models. Investigation into the mechanism of action revealed that CTX-8371 co-engagement of PD-1 and PD-L1 induced the proteolytic cleavage and loss of cell surface PD-1, which is a novel and non-redundant mechanism that adds to the PD-1/PD-L1 signaling axis blockade. The combination of CTX-8371 and an agonistic anti-CD137 antibody further increased the anti-tumor efficacy with long-lasting curative therapeutic effect. In summary, CTX-8371 is a novel checkpoint inhibitor that might provide greater clinical benefit compared to current anti-PD-1 and PD-L1 antibodies, especially when combined with agents with orthogonal mechanisms of action, such as agonistic anti-CD137 antibodies.

Keywords: Cancer immunotherapy; immuno-oncology.

Figure 1.

Discovery and binding profile of CTX-8371. (a) One-way MLR assay where monocyte derived DCs were used as stimulators and donor mismatched CD4⁺T cells as responders. The activity of StitchMab™ bispecifics was compared to that of their unstitched combinations. IFN-γ was determined by MSD (mean ± SEM, n = 3) *, p < 0.05, one-way ANOVA. (b) CTX-8371 domain organization. (c) Sensorgrams of CTX-8371 binding to recombinant human PD-1 (left) and PD-L1 (right). The different curves correspond to determinations obtained with increasing concentrations of analyte (PD-1 or PD-L1, light to darker shades of blue). Red profiles represent the mathematical interpolation of the experimental data which was used to derive the kinetic parameters. (d) Binding of CTX-8371 to primary activated CD3⁺ T cells from each species by FACS (mean ± SEM, n = 3).

Figure 2.

CTX-8371 enhances T cell function in primary human cells. (a) Jurkat-PD-1/NFAT-Luciferase reporter assay (mean ± SD; n = 3); (b) SEA polyclonal PBMC activation; (c) Antigen specific killing of tumor targets; (d) IFN-γ levels in the MLR (mean± SEM, n = 3). In this experiment, CD14⁺ monocytes were used as stimulators and mismatched CD3⁺ cells as responders.

Figure 3.

CTX-8371 anti-tumor activity in vivo. (a) Average tumor volume over time (mean ± SEM, n = 8). (b) Individual tumor growth curves. (c) %TGI on Day 27 (mean ± SEM, n = 7). ****, p < 0.0001, *, p < 0.05, one-way ANOVA, Tukey’s multiple comparisons test. (d) Experimental design for lung metastasis prevention study. (e, g) Lungs, 15 (e) or 14 (g) Days after tumor cell injection and treatment. (f) Number of metastatic nodules in individual lungs (mean ± SEM, n = 7), ***, p = 0.001, one-way ANOVA and Tukey’s multiple comparisons test. The lung nodules in control mice were too numerous to count (TNTC) and were excluded from the statistical analysis. (h) Number of lung metastatic nodules in individual mice (mean ± SEM, n = 9). **, p < 0.01, one-way ANOVA.

Figure 4.

CTX-8371 promotes the loss of PD-1 in T cells. (a) PBMCs were stimulated with anti-CD3/CD28 for 72 h and treated overnight with antibodies. Endogenous levels of total PD-1 protein were detected by WB using a commercial rabbit monoclonal antibody directed against the intracellular domain of PD-1. (b) WB of PBMCs from (a), treated with antibodies overnight ± increasing concentrations of batimastat. (c) PD-1 surface expression and levels of PD-1 ECD in the supernatant of PBMCs stimulated as in (a) and treated overnight with antibodies (mean ± SEM, n = 2). *, p < 0.05, ***, p < 0.005, unpaired t test. (d) PDCD1 fold change compared to naive PBMCs of PBMCs activated with anti-CD3/CD28 for 72 h and activated PBMCs treated with antibodies overnight in the absence of anti-CD3/CD28. (e, f) PBMCs were stimulated for 72 h with anti-CD3/CD28 in the presence of IgG1 or CTX-8371. PD-1 surface expression (e) and PD-1 ECD (f) were determined every 24 h (mean ± SEM, n = 2). ****, p < 0.0001 for PD-1 MFI, ***, p < 0.005 for PD-1 ECD, two-way ANOVA. (g) 4 day-stimulated PBMCs were incubated overnight with antibodies, washed and restimulated with anti-CD3/anti-CD28 for 5 more days. The 5-day kinetics of PD-1 loss by WB after overnight exposure to CTX-8371. Pembro, pembrolizumab.

Figure 5.

Loss of PD-1 by CTX-8371 occurs in trans and requires proteolytic activity. (a) Trans or cis Jurkat systems were treated with antibodies and increasing concentrations of batimastat. PD-1 expression by FACS is plotted as a function of batimastat concentration. Dotted lines represent the cut-off PD-1 expression on Jurkat-PD-L1 (left) or Jurkat parental cells (right). (b) WB of PD-1 expression in lysates of mixed Jurkat cells from (a). (c) PD-1 ECD levels in the mixed reactions (mean ± SD, n = 3). ****, p < 0.0001 between CTX-8371 and IgG1 or α-PD1+α-PD-L1 at all concentrations, one-way ANOVA and Tukey’s multiple comparisons test. (d) Representative images of trans Jurkat cells by IncuCyte.

Figure 6.

CTX-8371 causes CD8 T cell infiltration in tumors and global loss of PD-1 on T cells. mice received a single dose of CTX-8371. Spleen, tdLns, PBMCs, and tumor cell suspensions were analyzed by FACS. (A) CD8⁺ T cells percentages out of CD45⁺ immune cells on day 7. (B) Absolute numbers of CD45⁺ (upper) and CD8⁺T cells (lower) per gram of tumor tissue (mean ± SEM, n = 4), ****, p < 0.0001, ***, p < 0.005, **, p < 0.001, *, p < 0.05. Two-way ANOVA and Tukey’s multiple comparisons test. (C) Representative dot plots of PD-1-expressing T cells on day 3 after dosing with CTX-8371 or IgG1. (D) PD-1⁺ T cell frequencies in all tumors (mean ± SEM, n = 6). *, p < 0.05, **, p < 0.001, ***, p < 0.005, ****, p < 0.0001 unpaired t test. (E) CTX-8371 reduces PD-1 on CD4⁺ and CD8⁺ T lymphocytes in the periphery on day 6. (mean ± SEM, n = 6), ****, p < 0.0001, ***, p < 0.005, **, p < 0.001, *, p < 0.05 unpaired t test.

Figure 7.

CTX-8371-mediated loss of PD-1 on peripheral T lymphocytes of cynomolgus monkeys. (a, c) Representative dot plots of monkey PD-1⁺ CD4⁺ and CD8⁺ T cells at pre-treatment and on days 2 and 9 after CTX-8371 dosing. (b, d) peripheral blood PD-1⁺ CD4⁺ and CD8⁺ T cell frequencies and PD-1 MFI in all monkeys (mean ± SEM, n = 3). **, p < 0.001, *, p < 0.05, one-way ANOVA and Dunnett’s multiple comparisons test.

Figure 8.

Anti-tumor efficacy of CTX-8371 in combination with anti-CD137 antibodies. (a) Representative contour plots for each dose and time point showing CTL frequencies, gated on CD8⁺ T cells. (b) CD137⁺Granzyme B⁺CD8⁺ T cell frequencies in the 10 mg/kg and control (mean ± SEM, n = 4). ****, p < 0.0001, **, p < 0.001, *, p < 0.05. One-way ANOVA and Dunnett’s multiple comparisons test. (c) Individual tumor growth curves (n = 7) (d) Average tumor volume (mm³) for each treatment group. (e) MC38-hPD-L1 tumor growth plot of re-challenged tumor-free and naïve control mice (mean ± SEM, n = 4–10).