Antibody-Dependent Cellular Phagocytosis by Macrophages is a Novel Mechanism of Action of Elotuzumab

Abstract

Elotuzumab, a recently approved antibody for the treatment of multiple myeloma, has been shown to stimulate Fcγ receptor (FcγR)-mediated antibody-dependent cellular cytotoxicity by natural killer (NK) cells toward myeloma cells. The modulatory effects of elotuzumab on other effector cells in the tumor microenvironment, however, has not been fully explored. Antibody-dependent cellular phagocytosis (ADCP) is a mechanism by which macrophages contribute to antitumor potency of monoclonal antibodies. Herein, we studied the NK cell independent effect of elotuzumab on tumor-associated macrophages using a xenograft tumor model deficient in NK and adaptive immune cells. We demonstrate significant antitumor efficacy of single-agent elotuzumab in immunocompromised xenograft models of multiple myeloma, which is in part mediated by Fc-FcγR interaction of elotuzumab with macrophages. Elotuzumab is shown in this study to induce phenotypic activation of macrophages in vivo and mediates ADCP of myeloma cells though a FcγR-dependent manner in vitro Together, these findings propose a novel immune-mediated mechanism by which elotuzumab exerts anti-myeloma activity and helps to provide rationale for combination therapies that can enhance macrophage activity. Mol Cancer Ther; 17(7); 1454-63. ©2018 AACR.

Figure 1. Elotuzumab inhibits tumor progression and promotes mouse survival in early and late myeloma xenograft models

(a) Weekly ventral BLI acquired over the course of 6 weeks of SCID-beige mice (n=8 per group) injected i.v. with 5 million MM1S GFP⁺ Luc⁺ cells. Treatment in the early xenograft model was initiated 48 hours later and comprised the administration of 10mg/kg of hIgG1 or elotuzumab i.p. twice weekly. (b) Early xenograft model Kaplan-Meier survival analysis showing significantly prolonged survival among elotuzumab treated mice (mean survival= 87 days) compared to hIgG1 treated group (mean survival=38 days) P<0.001. (c) Ventral BLI imaging at week 5 of late xenograft model where mice (n=6-7 per group) were randomized into the two treatment groups upon showing first positive BLI signal at week 2. A total of 5 doses of 10mg/kg of elotuzumab or hIgG1 were administered resulting in significant decrease in BLI intensity. (d) Late xenograft model Kaplan-Meier survival analysis showing significantly prolonged survival among elotuzumab treated mice (mean survival= 58.5 days) compared to hIgG1 treated group (mean survival=38 days) P= 0.027.

Figure 2. Elotuzumab induces FcγR mediated immunostimulation of macrophages

(a) Weekly ventral BLI acquired over the course of 6 weeks of SCID-beige mice (n=5 per group) injected i.v. with 5 million MM1S GFP⁺ Luc⁺ cells. Treatment in the early xenograft model was initiated 48 hours later and comprised the administration of 10mg/kg of hIgG1, Fc-inert variant of elotuzumab, or elotuzumab i.p. twice weekly. (b) Early xenograft model Kaplan-Meier survival analysis showing significantly prolonged survival among elotuzumab treated mice (mean survival= 50 days) compared to hIgG1 and Fc-inert variant of elotuzumab treated groups (mean survival=38 and 36 days respectively) P<0.001. (c) Dorsal BLI imaging of early tumor xenograft model among NSG mice (n=3 per group) treated with elotuzumab or its Fc-inert variant at dose of 10mg/kg administered i.p. twice per week.

Figure 3. Elotuzumab promotes recruitment of macrophages capable of mediating a potent anti-tumor effect comparable to that of NK cells

(a) Tumor growth curves of EG7-hSLAMF7 injected subcutaneously to C57Bl/6j mice (n= 8 per group). Where indicated, mice were either depleted of macrophages (anti-CSF1R antibody) or NK cells (asialo GM1 antibody) and concurrently received elotuzumab-g2a or mIgG2a isotype control treatment. Analysis of the median tumor volume among the different treatment groups was conducted on day 18 post treatment. Mice receiving elotuzumab treatment (group: elotuzumab-g2a) had the significant reduction in median tumor volume (111.25 mm³) compared to that of isotype control treated group (mIgG2a) (667.17 mm³, P=0.028). NK cell depleted mice that concurrently received elotuzumab (group: elotuzumab-g2a + asialo GM1) demonstrated a significant reduction in tumor volume (686.56 mm³) compared to mice that were only depleted of NK cells (group: asialo GM1, median tumor volume= 1,264.07 mm³, P=0.005) signifying the involvement of other effector cells in mediating anti-tumor effects of elotuzumab. A similar reduction of tumor volume among macrophage depleted mice treated with elotuzumab (group: elotuzumab + CSF-1R, 669.51 mm³) compared to macrophage depleted mice (group: CSF1R, 954.56 mm³) that was not statistically significant. The depletion of macrophages in elotuzumab treated mice (group: elotuzumab-g2a + CSF1R, median tumor volume = 669.51 mm³) abrogated the elotuzumab induced reduction in median tumor volume in immunocompetent mice (group: elotuzumab-g2a, 111.25 mm³, P=0.037). NK cell depletion also resulted in a comparable loss of elotuzumab efficacy (686.56 mm³, P=0.037). (b) Among tumor infiltrating leukocytes (TIL), the percentage of F4/80⁺ macrophages was analyzed 11 days after treatment initiation among mice treated with elotuzumab or isotype control (n= 7-8 mice per group) with and without antibody based depletion of CSF1R bearing macrophages (n= 3 mice per group). There is a 1.59-fold increase in the percentage of macrophages accumulating at the tumor site among mice treated with elotuzumab compared to isotype control (P=0.042). Efficacious depletion of macrophages using CSF1R depleting antibody among mice that received treatment with elotuzumab or isotype control is demonstrated with the significant reduction in F4/80⁺ populations (P=0.026 and P=0.009 respectively). Statistical significance was determined by *P<0.05, **P<0.01 by two-tailed nonparametric Mann Whitney U test (A) and one-tailed unpaired t-test (B). Bars on day 18 represent median ± deviation

Figure 4. Dose dependent accumulation and phenotypic activation of TAMs by elotuzumab

(a) Individual tumor growth curves of subcutaneously injected OPM2 cells in SCID mice (n= 10 per group) randomized for hIgG1 or elotuzumab (0.5mg/kg or 1mg/kg) daily treatment for a total of 5 doses. (b, c) Flow cytometry analysis of TAMs and spleen macrophages at days 1 and 8 post treatment (n=3-5 per group per timepoint) for the expression of F4/80⁺ macrophages and CD86 and PD-L1 surface markers. (b) TAMs from elotuzumab (1mg/kg) treated mice showed statistically significant increase in the percentage of F4/80+ macrophages (P=0.007) expressing CD86 (P=0.003) and PD-L1 (P=0.018) 1 day after treatment compared to the hIgG1 control. A dose dependent response was also noted with the 1mg/kg elotuzumab dosed group showing a statistically significant increase in the percentage of F4/80+TAMs expressing CD86 (P=0.009) and PD-L1. While the increase in F4/80+ TAM percentage was not sustained till day 8 after treatment, phenotypic changes were preserved. Also, the percentage of CD86 and PD-L1 expressing macrophages became comparable among groups receiving elotuzumab at both dosages. (c) Among macrophages isolated from the spleen, elotuzumab treated mice did not demonstrate an increase in the percentage of F4/80⁺ macrophages or the percentage of CD86⁺ or PD-L1⁺ macrophages compared to the hIgG1 control. The analysis of marker expression on days 1 and day 8 was separately normalized to the hIgG1 control of the same day experiment. The data is representative of two independent experiments and statistical significance was determined by *P<0.05, **P<0.001 by two tailed unpaired t-test.

Figure 5. Flow cytometry and confocal microscopy reveal FcγR mediated ADCP of myeloma cells

(a) ADCP flow cytometry analysis of undifferentiated TAMs and M1 polarized TAMs that were co-cultured with MM1S GFP⁺ Luc⁺ opsonized with hIgG1, Fc-inert variant of elotuzumab, or elotuzumab. The double positive quadrant denotes the percentage of phagocytosis after 18-24 hrs of co-culture. (b) Bar graph representation of ADCP from 3 independent experiments. M1 polarized macrophages with LPS and IFNγ show significant increase in FcγR dependent phagocytosis of myeloma cells opsonized with elotuzumab as compared to hIgG1 and Fc-inert variant of elotuzumab. Despite treatment with elotuzumab, ADCP was significantly reduced among unpolarized TAMs. (c) Confocal microscopy of M1 polarized TAMs stained for F4/80 (Alexa Fluor 594) co-cultured with MM1S GFP⁺ Luc⁺ opsonized with hIgG1, Fc-inert variant of elotuzumab, or elotuzumab. Robust internalization of multiple elotuzumab opsonized myeloma cells by M1 polarized macrophages is shown compared to hIgG1 and Fc-inert elotuzumab treated cells. Indicated calibration bars corresponding to 30 μm. Data represents three independent experiments and the bars represent the mean ± S.E.M. *P<0.05, **P<0.01 by two-tailed unpaired t-test.