An Antibody Targeting ICOS Increases Intratumoral Cytotoxic to Regulatory T-cell Ratio and Induces Tumor Regression

KY1044 hIgG1 triggered ADCC and had costimulatory properties. A, ADCC cell reporter assay with KY1044. KY1044 hIgG1 or the isotype control was incubated with FcgRIIIA Jurkat reporter cells and ICOS⁺CHO cells as target cells. B, KY1044 induced ADCC in a human primary NK cell assay. KY1044 or isotype control was coincubated with dye-loaded ICOS⁺ CEM target cells and NK cells for 4 hours. ADCC activity was quantified by measuring Specific Dye Release upon target killing and normalized to percentage of maximum lysis compared with the positive control, a digitonin-based lysis buffer (mean of triplicates ± SEM). For A and B, the vertical dotted line indicates 100 ng/mL. Plate-bound (C) and cross-linked soluble (D) KY1044 induced the release of IFNγ from MJ cells. For C and D, the vertical dotted line indicates 10 μg/mL. The concentration of IFNγ was assessed in the supernatant of MJ cells cultured for 72 hours. E and F, Levels of IFNγ induced in human primary T cells activated with anti-CD3/CD28 (to induce ICOS) and cultured with either 5 μg/mL of plate-bound KY1044 or the isotype control IgG1 (E; n = 10) or with either 15 μg/mL of soluble KY1044 or soluble isotype control with or without the addition of Fc cross-linking anti-human F(ab')₂ Fragments (F; n = 10). G, Cytokine production by KY1044 requires anti-CD3. Levels of TNFα in T-cell culture with 5 μg/mL of KY1044 or isotype control IgG1 (plate bound; n = 8) in the presence or not of anti-CD3 (Wilcoxon statistic test for E, F, and G). H, RNA-sequencing analysis comparing T-cell stimulation with either KY1044 + anti-CD3 or control (anti-CD3 only). (i) Genes reported to be downstream of ICOS signaling showed a pattern of upregulation following KY1044 costimulation, with significant P values obtained for Ifng, Il10 and Il4. (ii) Metascape pathway enrichment analysis of the differentially expressed genes. Enriched gene sets were ranked on the basis of the significance of their P values. (iii) Heatmap showing unsupervised clustering of samples based on the expression levels of the top gene ontology terms from the Metascape enrichment analysis. IC, isotype control.

Effect of KY1044 mIgG2 in the A20 and J558 tumor models. A, Spider plots of a representative experiment showing individual A20 tumor volumes (n = 10 per group). Tumor-bearing mice were dosed intraperitoneally with 200 μg of KY1044 mIgG2a or 200 μL saline starting from day 8 following tumor cell implantation. B, Spider plots showing individual J558 tumor volumes (n = 7 per group). Tumor-bearing mice were dosed intraperitoneally with 60 μg of KY1044 mIgG2a or 200 μL saline starting from day 9 following tumor cell implantation. The dosing days are indicated by vertical dotted lines. The survival curves depicting the control and KY1044 mIgG2a groups are shown. Statistics were calculated using log-rank (Mantel–Cox) test. Data are representative of at least two experiments.

KY1044 mIgG2a improved the efficacy of anti–PD-L1. A, Survival curves of mice harboring CT26.WT tumors and treated with control, KY1044 mIgG2a monotherapy at 60 μg/dose, anti–PD-L1 mIgG2a monotherapy at 200 μg/dose, and the combination of KY1044 mIgG2a and anti–PD-L1 mIgG2a at 60 and 200 μg/dose, respectively (left). The dosing days are indicated by vertical dotted lines. Statistics were calculated using log-rank (Mantel–Cox) test. The data are representative of six independent experiments. Tumor-bearing mice that cleared CT26.WT tumors were randomly allocated to two groups and rechallenged with either CT26.WT or EMT6 cells (right). B, Effect of CD4⁺ and CD8⁺ T-cell depletion on antitumor efficacy. Mice implanted with CT26.WT cells (n = 10 per group) were depleted of CD8⁺ and/or CD4⁺ T cells and treated with saline or with KY1044 mIgG2a and anti–PD-L1 mIgG2a combination as described in A. Data are representative of two independent experiments.

KY1044 depleted ICOS^high T cells in vivo. A, In vivo experimental protocol of the pharmacodynamic study in the CT26.WT tumor–bearing mice. Mice were dosed twice with saline or KY1044 mIgG2a (ranging from 0.3–10 mg/kg) on days 13 and 15, and the tumor and spleen were harvested on day 16 and the immune cells analyzed by FACS. Effect of KY1044 on Treg depletion (plotted as a percentage of live cells) and on the CD8⁺ T_Eff to Treg cell ratio in the TME (B) and in the spleen (C) of treated mice (n = 7/8 mice per groups). D, Relative expression of ICOS and frequency of T-cell subsets in the blood of NHP (the number of datapoints are indicated in brackets). The percentages indicate the frequency of each cell types within CD3/CD4 T cells in NHP PBMCs. geoMFI, geometric mean fluorescence intensity. E–G, Graphs showing the changes (vs. baseline pretreatment) of total CD4 T_M cells and follicular Th cells in the blood of NHP over time following a single-dose of KY1044 human IgG1 at 0, 10, 30, and 100 mg/kg (n = 3 NHP per groups).

KY1044 mIgG2a activated intratumoral T cells in vivo. A, Effect of KY1044 mIgG2a on intratumoral CD8⁺ T-cell expression of CD69 and CD44 (n = 7/8 mice per groups). Measurement of CD69 and CD69/CD44 double-positive cells was performed 24 hours after a second dose of KY1044 mIgG2a (see Fig. 5A). B, Graphs showing the flow cytometry analysis of intracellular IFNγ and/or TNFα expression in intratumoral CD8⁺ and CD4⁺ T cells from CT26.WT tumor collected 7 days after a single intraperitoneal injection of saline or KY1044 mIgG2a at 3 mg/kg. Groups of mice (n = 4 mice/group; Tukey multiple comparisons test).