A Bispecific Antibody to Link a TRAIL-Based Antitumor Approach to Immunotherapy

Abstract

T-cell-based immunotherapy strategies have profoundly improved the clinical management of several solid tumors and hematological malignancies. A recently developed and promising immunotherapy approach is to redirect polyclonal MHC-unrestricted T lymphocytes toward cancer cells by bispecific antibodies (bsAbs) that engage the CD3 complex and a tumor-associated antigen (TAA). The TNF-related apoptosis-inducing ligand receptor 2 (TRAIL-R2) is an attractive immunotherapy target, frequently expressed by neoplastic cells, that we decided to exploit as a TAA. We found that a TRAIL-R2xCD3 bsAb efficiently activates T cells and specifically redirect their cytotoxicity against cancer cells of different origins in vitro, thereby demonstrating its potential as a pan-carcinoma reagent. Moreover, to mimic in vivo conditions, we assessed its ability to retarget T-cell activity in an ex vivo model of ovarian cancer patients' ascitic fluids containing both effector and target cells-albeit with a suboptimal effector-to-target ratio-with remarkable results.

Keywords: T-cell retargeting; TRAIL-R2; bispecific antibody; immunotherapy; malignant ascites.

Figure 1

ScDb-redirected T-cell killing of melanoma cell lines. (A) Growth inhibition assays performed using nine different batches of PBLs (each derived from a different healthy donor) in the presence or absence of scDb. TRAIL-R2-expressing Me64, Me41, and Me15 and TRAIL-R2-negative MDA-MB-468 were used as target cells treated with 0.5 μg/mL of scDb and non-activated PBLs at an E:T ratio of 5:1 for 48 h (left panel) or 96 h (right panel). Each point represents a single experiment. Bars represent the mean of all nine experiments; error bars, SD. (B) Percentage of apoptosis (early + late apoptosis, annexin V⁺) of two sTRAIL-sensitive melanoma cell lines (top, Me15 and Me41) and two sTRAIL-resistant melanoma cell lines (bottom, Me71 and Me79) co-cultured for 72 h with 0.1 μg/mL scDb, non-activated PBLs, or both at E:T = 5:1. (n = 3; error bars, SD). (C) Representative plot of data shown in (B). (D) Tumor cells were treated for 4 and 16 h with pre-activated PBLs in the presence or absence of TRAIL-R2xCD3 scDb. Irrelevant scDb Mec14xCD3 was used as control. The graphs show the percentage of direct cell lysis as the mean ± SD of three experiments. (E) Immunofluorescence was performed using Me15 cells loaded with calcein-AM (green) and grown for 4 h with pre-activated PBLs stained using a PKH26 Red Fluorescent Cell Linker Kit in the absence (left panels) or presence (right panels) of 0.5 μg/mL TRAIL-R2xCD3 scDb. 0.5 μg/mL irrelevant scDb Mec14xUCHT1 was used as control (central panels). The pictures were taken at the end of the incubation time. Statistical analysis in (A,B,D) by one-way ANOVA followed by Tukey's post-test. ^*p < 0.05, ^**p < 0.01, ^***p < 0.001; ns: not significant.

Figure 2

Specificity of scDb-mediated cytotoxicity. (A) Growth inhibition assays were performed treating Me15, Me41, and Me64 cells with 0.5 μg/mL irrelevant scDb Mec14xCD3 or scDb TRAIL-R2xCD3 and non activated PBLs at an E:T ratio of 5:1 for 48 and 96 h. Mean values and SD of triplicate determinations are shown. (B) Competition between a commercial anti-TRAIL-R2 mAb and the scDb was evaluated by FACS. Before the addition of the scDb, TRAIL-R2⁺ Me41 cells were incubated for 1 h with the anti-human TRAIL-R2 mAb (left panel) or with PBS (middle panel). Binding of the scDb is revealed by an anti-6xHis tag rabbit mAb followed by an Alexa Fluor 488-labeled anti-rabbit Ab. The binding of the competing anti-TRAIL-R2 mAb was revealed by a secondary Alexa Fluor 488-labeled antimouse Ab (right panel). Empty peaks: negative control; gray peaks: scDb or mAb. (C) Growth inhibition assays were performed treating Me15, Me41, and Me64 cells with 0.5 μg/mL scDb and non-activated PBLs at an E:T ratio of 5:1 for 48 h. Pre-incubation with 10 μg/mL anti-TRAIL-R2 mAb was used as scDb-competing mAb to test scDb specificity. The graphs show the mean ± SD of three experiments. Each point represents a single experiment. Bars represent the mean ± SD. Statistical analysis in (A,C) by one-way ANOVA followed by Tukey's post-test. ^***p < 0.001; n.s.: not significant.

Figure 3

Tumor growth inhibition assay on a panel of TRAIL-R2-positive or negative cells. ScDb-dependent PBL-mediated growth inhibition was performed on several cancer cell lines using scDb concentrations of 0, 0.05, 0.1, 0.5, and 1 μg/mL and an E:T ratio of 5:1 for 48 h (black line) or 96 h (gray line). Non-activated PBLs were used as effectors. The graph represents the mean ± SD of three experiments. In each graph the expression level of TRAIL-R2 as mean fluorescence intensity (MFI) is reported.

Figure 4

Effects of scDb on T-cell activation. (A) Frequency of CD25⁺, CD137⁺, and CD69⁺ T lymphocytes in CD8⁺ (top) and CD4⁺ (bottom) subsets after treatment with increasing doses of scDb in the presence of TRAIL-R2⁺ Me15 cells at E:T=5:1. (B) Effects of scDb on T-cell activation when TRAIL-R2 negative BT-474 cells were treated. Frequency of CD25⁺, CD137⁺, and CD69⁺ T lymphocytes in CD8⁺ subsets after treatment with 0.5 μg/mL scDb in the presence of TRAIL-R2⁻ BT-474 cells at E:T=5:1. Pre-incubation with 10 μg/mL anti-CD3 was used as positive control. The graphs show the mean ± SD of three experiments. (C) Representative plot of (B). (D) Frequency of CD25⁺, CD137⁺, and CD69⁺ cells in CD8⁺ T lymphocytes (after gating on CD45⁺CD3⁺ cells) after 16 h of treatment with 0.5 μg/mL scDb in the presence or absence of the melanoma target cell line (Me15). Pre-incubation with 10 μg/mL anti-TRAIL-R2 mAb and anti-CD3 was used as scDb-competing Ab and positive control, respectively. The graphs show the mean ± SD of three experiments. Statistical analysis in (B,D) by one-way ANOVA followed by Tukey's post-test. ^***p < 0.001; ns: not significant.

Figure 5

ScDb-mediated CD4⁺ and CD8⁺ T-cell proliferation. (A) Proliferation (CFSE dilution assay) of CD8⁺ and CD4⁺ T cells after overnight treatment with increasing doses of scDb in the presence or absence of TRAIL-R2⁺ melanoma target cells. (B) Representative plot of data shown in (A). (C) Proliferation assay of CD8⁺ T cells at day +4 after treatment with 0.1 μg/mL scDb in the presence or absence of TRAIL-R2 negative BT-474 cells. Stimulation with PHA (5 μg/mL) was used as positive control. Graph represents mean ± SD of three different experiments. (D) Representative plot of (C). Statistical analysis in (C) by one-way ANOVA followed by Tukey's post-test. ns: not significant.

Figure 6

ScDb triggering of PBL lytic machinery. (A) Increased frequency among CD3⁺/CD8⁺ and CD3⁺/CD4⁺ T lymphocytes of CD107a⁺ or IFN-γ⁺ cells after treatment with 0.1 μg/mL scDb in the presence or absence of a melanoma target cell line (Me15) at E:T 5:1. The graphs represent the mean ± SD of three experiments. (B) Surface expression of CD107a and intracellular staining for IFN-γ by flow cytometry in CD3⁺/CD8⁺ (left) and CD3⁺/CD4⁺ (right) T-cell subsets after treatment with 0.1 μg/mL scDb in the presence or absence of the Me15 melanoma target cell line. (C) Frequency among CD3⁺/CD8⁺ (left graphs) and CD3⁺/CD4⁺ (right graphs) T lymphocytes of perforin, granzyme B and Ki-67 expression, after treatment with 0.1 μg/mL scDb in the presence or absence of a melanoma target cell line (Me15) at E:T 5:1. The graphs represent the mean ± SD of three experiments. Statistical analysis in (A,C) by one-way ANOVA followed by Tukey's post-test. ^*p < 0.05, ^**p < 0.01, ^***p < 0.001; ns: not significant.

Figure 7

Treatment of ex vivo cells derived from ovarian cancer patients. (A) Primary ovarian carcinoma cytotoxicity assay after 4 and 16 h of treatment with TRAIL-R2xCD3 or Mec14xCD3 scDbs and pre-activated healthy donors' PBLs. 09ST and 10ST: short-term ovarian cancer cell lines established from biopsies; 13A, 15A, and 16A: cells isolated from ascitic fluid of ovarian cancer patients. The graphs show the percentage of direct cell lysis as the mean ± SD of three wells for treatment. Statistical analysis by one-way ANOVA followed by Tukey's post-test. ^***p < 0.001; ns: not significant. (B) Representative staining for CD45 and EpCAM (left) and TRAIL-R2 expression in EpCAM⁺ tumor cells (right) in an ascitic fluid from an ovarian cancer patient. (C) Flow cytometry expression of the surface activation markers CD107a and CD137 on CD8⁺ (left) and CD4⁺ (right) T lymphocytes of ascites samples from three ovarian cancer patients, treated or not with scDb (0.5 μg/mL, 18 h). (D) Annexin-V/propidium iodide apoptosis assay of tumor cells (EpCAM⁺/CD45⁻) treated or not with 0.5 μg/mL scDb for 48 h. Statistical analysis by Student T-test. ^*p < 0.05. (E) Annexin-V/propidium iodide apoptosis assay on CD8⁺ T cells treated as in (D).