Membrane-Proximal Epitope Facilitates Efficient T Cell Synapse Formation by Anti-FcRH5/CD3 and Is a Requirement for Myeloma Cell Killing

Abstract

The anti-FcRH5/CD3 T cell-dependent bispecific antibody (TDB) targets the B cell lineage marker FcRH5 expressed in multiple myeloma (MM) tumor cells. We demonstrate that TDBs trigger T cell receptor activation by inducing target clustering and exclusion of CD45 phosphatase from the synapse. The dimensions of the target molecule play a key role in the efficiency of the synapse formation. The anti-FcRH5/CD3 TDB kills human plasma cells and patient-derived myeloma cells at picomolar concentrations and results in complete depletion of B cells and bone marrow plasma cells in cynomolgus monkeys. These data demonstrate the potential for the anti-FcRH5/CD3 TDB, alone or in combination with inhibition of PD-1/PD-L1 signaling, in the treatment of MM and other B cell malignancies.

Keywords: CD3; FCRL5; FcRH5; T cell; bispecific antibody; multiple myeloma.

Figure 1

Anti-FcRH5/CD3 TDB Induces Target Clustering and Exclusion of CD45 from the Synapse Resulting in TCR Triggering (A) Schematic of the interface between the FcRH5-expressing target cell and the reconstituted HEK-T cell, with protein dimensions shown approximately to scale. The remaining components of the HEK-T cells, including the normally cytoplasmic kinase, ZAP70, have been omitted for clarity. (B) Representative images of TDB-mediated conjugates (1G7-TDB) between an FcRH5-expressing HEK cell and a reconstituted HEK-T cell, showing CD45 segregation, FcRH5 clustering, and ZAP70 recruitment at the cell interface. (C) Schematic of the interface between the pMHC-expressing target cell and the reconstituted HEK-T cell, with protein dimensions shown approximately to scale. The remaining components of the cells, including the normally cytoplasmic kinase, ZAP70, have been omitted for clarity. (D) Representative images of conjugates between a pMHC-expressing Raji B cell and a reconstituted HEK-T cell, showing CD45 segregation, pMHC clustering, and ZAP70 recruitment at the cell interface. (E) A reconstructed en face view of a conjugate interface equivalent to that shown in (B). (F) Line profiles of CD45 (green), FcRH5 (blue), and TDB (red) intensities across a conjugate interface equivalent to that shown in (B). In all images, box color denotes that used in the overlay image. Scale bars, 5 μm (B) and (D) or 2 μm (E) in length.

Figure 2

Membrane-Proximal Epitope Is Required for Efficient Synapse Formation and Killing Activity of Anti-FcRH5/CD3 (A) Schematic of the epitope locations of the three TDBs used in the figure, including the gD epitope at the N terminus. (B) Representative confocal images of the interface between a HEK-T cell and an FcRH5-expressing HEK cell, where conjugate formation has been induced using the indicated fluorescently labeled TDB. The overlay image identifies the cell-cell interface over which the intensities of CD45, FcRH5, and TDB were measured, relative to the rest of the plasma membrane. Box color denotes that used in the overlay image. Scale bars, 5 μm. (C) Quantification of CD45 intensity in the conjugate interface relative to CD45 intensity in the rest of the plasma membrane. Data shown are the mean from 12 to 17 conjugates for each TDB is shown ±SEM. (D) Equivalent quantification of FcRH5 intensity in the same cells used in (C), presented as fold-increase of intensity compared with rest of target plasma membrane ± SEM. (E) Phospho-SLP76 western blot of peripheral CD8+ T cells from a healthy donor stimulated with 1 μg/mL TDBs and cells expressing human FcRH5 with an N-terminal gD expression tag. Total SLP76 is shown to confirm equal sample loading. (F) FcRH5 target cell killing using either 1G7 (blue) or gD (red) TDBs using CD8+ T cells. (G) Schematic of the truncated FcRH5 construct with the gD tag now proximal to the membrane. (H) Target cell killing using the truncated FcRH5 construct using 1G7 (blue) or gD (red) TDBs. The truncated construct was expressed in HEK293 cells. (I) Target cell killing using 1G7 TDB (blue), alternate TDBs that recognize the membrane-proximal epitope (dashed lines), or 10A8 TDB (green). Data in (F), (H), and (I) are represented as the mean ± SD. See also Figure S1.

Figure 3

Anti-FcRH5/CD3 TDB Induces T Cell Activation, Target-Dependent Cell Killing, and T Cell Proliferation (A) Dose-dependent activation of CD4⁺ and CD8⁺ T cells upon stimulation with target cells (MOLP-2) and anti-FcRH5/CD3 TDB detected by flow cytometry analysis. (B) Target cell (MOLP-2) killing by anti-FcRH5/CD3 TDB and effector cells (CD4⁺ or CD8⁺). (C) Target-dependent killing by anti-FcRH5/CD3 TDB and flow cytometry (insert) of a parental Fox-NY cell line (green) and clones transfected to express low (blue) or high level (red) of FcRH5. (A), (B), and (C) Data are represented as the mean ± SD. (D) Anti-FcRH5/CD3 TDB-induced CD8⁺ T cell proliferation response (5 days) was detected by measuring carboxyfluorescein succinimidyl ester (CFSE) fluorescence intensity dilution. Only CFSE-labeled human CD8⁺ T cells and those co-cultured with MOLP-2 ± anti-FcRH5/CD3 TDB (1 μg/mL) are shown. See also Figure S2.

Figure 4

Expression of FcRH5 in Normal Tissues and Multiple Myeloma (A) FcRH5 RNA-seq analysis of the Genotype-Tissue Expression (GTEx) sample set. Boxes represent the interquartile range (IQR, 25th to 75th percentile), the horizontal line is the median. Whiskers extend to the most extreme data point that is within 1.5^∗IQR of the 25th and 75th percentiles. Data points beyond the whiskers represent outliers. (B) Correlation of FcRH5 expression with CD19, CD20, and BCMA expression in the GTEx samples. (C) FcRH5 protein expression in primary multiple myeloma tumor cells and healthy donor peripheral B cells and bone marrow plasma cells. FcRH5 expression was analyzed by flow cytometry normalized to expression in MOLP-2 internal and assay control. Data are represented as the mean ± SEM. (D) FcRH5 mRNA expression in anti-CD138 purified plasma cells from newly diagnosed non-treated multiple myeloma patients measured using Affymetrix GeneChip Human Exon 1.0 ST. (E) qRT-PCR analysis of FcRH5 mRNA level in bone marrow biopsies from patient myeloma samples with or without 1q21 gain. mRNA expression level was calculated by the delta Ct (dCt) method. Statistical analysis was performed using a Mann-Whitney U test. The box represents the 25th to 75th percentiles and median. Whiskers represent the range of minimum and maximum values. (F) Representative images of FISH analysis on primary multiple myeloma biopsies showing normal diploid of 1q21 (red; top panel) and a mixture of three to six copies of 1q21 (bottom). A tumor sample was identified as 1q21 gain when >20% of the tumor cells scored had three or more copies of the 1q21.3 locus. Scale bars, 10 μm in length.

Figure 5

Anti-FcRH5/CD3 TDB Mediates Potent Killing of Normal Plasma Cells and Patient-Derived Primary Myeloma Cells (A) Cytotoxic activity on human plasma cells was evaluated by culturing human bone marrow mononuclear cells (BMMC) for 72 hr with anti-FcRH5/CD3 TDBs and analyzing the number of live CD38⁺CD138⁺ cells by flow cytometry. (B) Cytotoxic activity of anti-FcRH5/CD3 TDB on patient-derived primary myeloma cell was evaluated by co-culturing human myeloma BMMCs with CD8⁺ T cells isolated from healthy donor and anti-FcRH5/CD3 TDB. The killing activity was analyzed as in (A). (C) The number of anti-FcRH5/CD3 TDB binding sites per cell was determined using Scatchard analysis and is indicated in the panel. Cytotoxic activity of anti-FcRH5/CD3 TDB against target cells expressing low levels of FcRH5 was assessed as in (B). Data are represented as the mean ± SD.

Figure 6

Anti-FcRH5/CD3 TDB Suppresses Growth of Established MOLP-2 Tumors in Mice Reconstituted with Human Immune Cells (A) MOLP-2 killing activity of peripheral T cells from healthy human donors and of human CD8⁺ T cells isolated from spleens of humanized NOD/SCID gamma (huNSG) mice. Data are represented as the mean ± SD. (B) The mice were treated with a weekly intravenous dose of vehicle or anti-FcRH5/CD3 TDB at 0.5 mg/kg. The individual tumor volumes (gray), mean tumor volume (blue bold line in the vehicle-treated group; black bold line in the anti-FcRH5/CD3 TDB-treated group), and mean of control group (blue dashed line) are shown.

Figure 7

Anti-FcRH5/CD3 TDB Depletes B Cells and Bone Marrow Plasma Cells in Cynomolgus Monkey A single intravenous dose of vehicle (blue) or anti-FcRH5/CD3 TDB (red, 1 mg/kg; brown, 2 mg/kg; green, 4 mg/kg) was administered to three cynomolgus monkeys/group. (A) Blood samples were collected at indicated time points and human IgG was detected by ELISA. (B and C) The effect of anti-FcRH5/CD3 TDB on CD4⁺ (B) and CD8⁺ (C) T cell activation in peripheral blood. (D–G) The absolute count of CD20⁺ B cells in peripheral blood (D), spleen (E), bone marrow (F), and mandibular lymph node (G). (H and I) The effect of anti-FcRH5/CD3 TDB on bone marrow plasma cells (H) and serum IgG (I) levels. The difference between before and after treatment was analyzed by unpaired t test. Data in E–I are represented as the mean ± SEM. See also Figures S3–S5.

Figure 8

PD-L1 Blockade Enhances Activity of Anti-FcRH5/CD3 TDB (A) CD8⁺ T cells were stimulated for 48 hr with 1 μg/mL anti-FcRH5/CD3 TDB and MOLP-2 target cells, and analyzed by flow cytometry for the PD-1 expression. (B) The percentage of PD-1⁺ cells in CD8⁺ T cells in cynomolgus monkeys after single-dose intravenous administration of vehicle (blue), 1 mg/kg (red), 2 mg/kg (brown), and 4 mg/kg (green) of anti-FcRH5/CD3 TDB. Individual animal data are shown. (C) The ability of anti-FcRH5/CD3 TDB to redirect activity of pre-stimulated CD8⁺ T cells (A) to kill 293-FcRH5-PD-L1 cells was tested in the presence (blue) or absence (red) of 10 μg/mL anti-PD-L1. Data are represented as the mean ± SEM. See also Figure S6.