Epcoritamab induces potent anti-tumor activity against malignant B-cells from patients with DLBCL, FL and MCL, irrespective of prior CD20 monoclonal antibody treatment

Abstract

Epcoritamab (DuoBody-CD3xCD20, GEN3013) is a novel bispecific IgG1 antibody redirecting T-cells toward CD20⁺ tumor cells. Here, we assessed the preclinical efficacy of epcoritamab against primary tumor cells present in the lymph node biopsies from newly diagnosed (ND) and relapsed/refractory (RR) B-NHL patients. In the presence of T-cells from a healthy donor, epcoritamab demonstrated potent activity against primary tumor cells, irrespective of prior treatments, including CD20 mAbs. Median lysis of 65, 74, and 84% were achieved in diffuse large B-cell lymphoma (n = 16), follicular lymphoma (n = 15), and mantle cell lymphoma (n = 8), respectively. Furthermore, in this allogeneic setting, we discovered that the capacity of B-cell tumors to activate T-cells was heterogeneous and showed an inverse association with their surface expression levels of the immune checkpoint molecule Herpesvirus Entry Mediator (HVEM). In the autologous setting, when lymph node (LN)-residing T-cells were the only source of effector cells, the epcoritamab-dependent cytotoxicity strongly correlated with local effector cell-to-target cell ratios. Further analyses revealed that LN-residing-derived or peripheral blood-derived T-cells of B-NHL patients, as well as heathy donor T-cells equally mediated epcoritamab-dependent cytotoxicity. These results show the promise of epcoritamab for treatment of newly-diagnosed or relapsed/refractory B-NHL patients, including those who became refractory to previous CD20-directed therapies.

Fig. 1. Epcoritamab mediates comparable levels of cytotoxicity in various B-NHL subtypes and in ND and RR samples, including samples from patients who received prior CD20-antibody containing treatment.

A Percentages cytotoxicity mediated by epcoritamab and CD3xCtrl (30 ng/mL) and B EC₅₀ values (ng/mL), shown for samples with clear dose-response curves, for cytotoxicity in the presence of allogeneic PBMCs (E:T ratio 10:1) in ND and RR DLBCL (n = 16), FL (n = 15), and MCL (n = 8) samples (median ± interquartile range), including patients who relapsed from or were refractory to CD20 therapy ( and , respectively). Statistical analysis was performed with Kruskal–Wallis and Dunn’s multiple comparisons test to compare epcoritamab with CD3xCtrl for each indication (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001) and epcoritamab-dependent cytotoxicity between subtypes (ns (C) CD20 expression (antibody molecules per cell) on tumor cells in samples from B-NHL patients unexposed to prior CD20 therapy correlated to epcoritamab-dependent cytotoxicity (30 ng/mL) (Spearman’s; ns). D Time between last CD20-therapy and moment of lymph node biopsy (months) of patients previously treated with CD20-antibody containing treatment (n = 8) correlated to cytotoxicity mediated by epcoritamab (30 ng/mL) (Spearman’s; r = 0.78, *p = 0.03).

Fig. 2. T-cell activation by epcoritamab does not correlate strongly with cytotoxic activity.

A Gating of CD4⁺ and CD8⁺ cells and example of a flow histogram for CD69 positivity and B percentages of CD69-positive CD4⁺ and CD8⁺ allogeneic T-cells after treatment of B-NHL samples with epcoritamab and CD3xCtrl (30 ng/mL) (median ± interquartile range). C Granzyme B levels (pg/mL) in supernatants of coculture of B-NHL sample and allogeneic T-cells stimulated with epcoritamab and CD3xCtrl (30 ng/mL) (median ± interquartile range). D, E Correlation between percentage CD69-positive (CD4⁺ and CD8⁺) allogeneic T-cells (D) or granzyme B release (pg/mL) (E), and cytotoxicity induced by epcoritamab (30 ng/mL) (Spearman’s; ns and r = 0.35; *p = 0.04, respectively). All samples from different B-NHL subtypes and from ND and RR are plotted together.

Fig. 3. Lower T-cell activation by epcoritamab in presence of tumor cells with high expression of the immune regulatory molecule HVEM.

A CD69-positive CD4⁺ and CD8⁺ allogeneic T-cells after treatment of B-NHL samples with epcoritamab (30 ng/mL) and CD3xCtrl (30 ng/mL) in samples with low or high tumor HVEM expression levels (Mann–Whitney U-test; ***p < 0.001, **p < 0.01). B Epcoritamab-dependent cytotoxicity (30 ng/mL) in B-NHL samples with low and high tumor cell HVEM expression (Mann–Whitney U-test; ns). HVEM expression on healthy donor B-cells was used as cut-off value to stratify between low and high expression. Data are shown as median ±interquartile range. All samples from different B-NHL subtypes and from ND and RR are plotted together.

Fig. 4. The intrinsic capacity of T-cells from B-NHL LN samples to induce cytotoxicity is comparable to that of healthy donor T-cells.

A Percentages cytotoxicity mediated by epcoritamab and CD3xCtrl (30 ng/mL) without addition of PBMCs in ND and RR DLBCL (n = 15), FL (n = 15), and MCL (n = 8) samples (median ± interquartile range), including patients who relapsed from or were refractory to CD20 therapy ( and , respectively). Statistical analysis was performed with Kruskal–Wallis and Dunn’s multiple comparisons test to compare epcoritamab with CD3xCtrl (*p < 0.05) and to compare epcoritamab-dependent cytotoxicity (subtracted with CD3xCTRL values) between subtypes (ns). B Ratio of CD4⁺ T helper cells (CD4⁺CD25⁻) to CD8⁺ T-cells in DLBCL, FL, and MCL samples (Kruskal–Wallis with Dunn’s multiple comparisons test; *p < 0.05, **p ≤ 0.01) (median ± interquartile range). C PD-1 expression (MFI values normalized to expression on T-cells of internal healthy donor control) on CD4⁺ T-cells of DLBCL, FL, including PD-1 bright and dim populations, and MCL (Kruskal–Wallis with Dunn’s multiple comparisons test; *p < 0.05, **p < 0.01, ***p < 0.001) (median ± interquartile range). D Spearman’s correlation between CD3:Target ratio and cytotoxicity mediated by epcoritamab (30 ng/mL) (r = 0.63; ****p < 0.0001). E Lymph node T-cells of B-NHL patients isolated by magnetic-activated cell-sorting (MACS) showed similar percentages B-cell lymphoma cell line (WSU-DLCL2 and Daudi, left to right) cytotoxicity as T-cells isolated from healthy donor PB (mean ± SEM) (Mann–Whitney U-test; ns).

Fig. 5. The intrinsic capacity of T-cells from peripheral blood of B-NHL patients to induce cytotoxicity is comparable to that of healthy donor T-cells.

A Dose-response of and B maximal cytotoxicity induced by epcoritamab in B-NHL cells in the presence of allogeneic healthy donor PBMCs or autologous B-NHL patient PBMCs in a 10:1 E:T ratio (n = 5; Wilcoxon matched-pairs signed rank test; ns).