A Defucosylated Anti-EpCAM Monoclonal Antibody (EpMab-37-mG2a-f) Exerts Antitumor Activity in Xenograft Model

The epithelial cell adhesion molecule (EpCAM) is a stem cell and carcinoma antigen, which mediates cellular adhesion and proliferative signaling by the proteolytic cleavage. In contrast to low expression in normal epithelium, EpCAM is frequently overexpressed in various carcinomas, which correlates with poor prognosis. Therefore, EpCAM has been considered as a promising target for tumor diagnosis and therapy. Using the Cell-Based Immunization and Screening (CBIS) method, we previously established an anti-EpCAM monoclonal antibody (EpMab-37; mouse IgG1, kappa). In this study, we investigated the antibody-dependent cellular cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC), and an antitumor activity by a defucosylated mouse IgG2a-type of EpMab-37 (EpMab-37-mG2a-f) against a breast cancer cell line (BT-474) and a pancreatic cancer cell line (Capan-2), both of which express EpCAM. EpMab-37-mG2a-f recognized BT-474 and Capan-2 cells with a moderate binding-affinity [apparent dissociation constant (KD): 2.9 × 10−8 M and 1.8 × 10−8 M, respectively] by flow cytometry. EpMab-37-mG2a-f exhibited ADCC and CDC for both cells by murine splenocytes and complements, respectively. Furthermore, administration of EpMab-37-mG2a-f significantly suppressed the xenograft tumor development compared with the control mouse IgG. These results indicated that EpMab-37-mG2a-f exerts antitumor activities and could provide valuable therapeutic regimen for breast and pancreatic cancers.


Introduction
EpCAM is a unique type I transmembrane glycoprotein which is expressed on the basolateral membrane of epithelial cells [1]. EpCAM mediates homophilic and intercellular adhesion through the extracellular domain, which is essential for the epithelial integrity [2]. EpCAM was the first identified human tumor antigen [3], and the expression is correlated with poor prognosis in various tumors [4][5][6][7]. EpCAM also functions as a signaling molecule. The formation in intercellular EpCAM oligomers triggers the transmembrane proteolytic cleavage by a membrane protease complex. The EpCAM intracellular domain serves as a

Animals
All animal experiments were performed following regulations and guidelines to minimize animal distress and suffering in the laboratory by the Institutional Committee for Experiments of the Institute of Microbial Chemistry (Numazu, Japan) (approval no. 2022-024). Mice were maintained on an 11 h light/13 h dark cycle in a specific pathogen-free environment across the experimental period. Food and water were supplied ad libitum. Mice weight was monitored twice per week and health was monitored three times per week. The loss of original body weight was determined to a point >25% and/or a maximum tumor size >2000 mm 3 and/or significant changes in the appearance of tumors as humane endpoints for euthanasia. Cervical dislocation was used for euthanasia. Mice death was confirmed by respiratory arrest and rigor mortis.

ADCC
ADCC induction by EpMab-37-mG 2a -f was assayed as follows. Six female BALB/c nude mice (five-week-old) were purchased from Charles River Laboratories, Inc. Spleens were aseptically removed, and single-cell suspensions were obtained through a sterile cell strainer (352360, BD Falcon). Erythrocytes were removed with treatment of ice-cold distilled water. The splenocytes were resuspended in the medium; this preparation was designated as effector cells. Target cells (BT-474 and Capan-2) were treated with Calcein AM (10 µg/mL, Thermo Fisher Scientific, Inc.). The target cells (2 × 10 4 cells) were mixed with effector cells (effector-to-target ratio, 100:1), 100 µg/mL of EpMab-37-mG 2a -f or control mouse IgG 2a in 96-well plates. After incubation for 4.5 h at 37 • C, the Calcein release into the medium was measured with an excitation wavelength (485 nm) and an emission wavelength (538 nm) using a microplate reader (Power Scan HT; BioTek Instruments, Inc., Winooski, VT, USA).
Cytolyticity (% lysis) was determined as follows: % lysis = (E − S)/(M − S) × 100. "E" is the fluorescence in the presence of both effector and target cells. "S" is the spontaneous fluorescence in the presence of only target cells. "M" is the maximum fluorescence by the treatment with a lysis buffer (10 mM Tris-HCl (pH 7.4), 10 mM of EDTA, and 0.5% Triton X-100).

ADCC Reporter Bioassay
The ADCC reporter bioassay was performed using an ADCC Reporter Bioassay kit from Promega (Madison, WI, USA), following the manufacturer's instructions [51]. Target cells (12,500 cells per well) were inoculated into a 96 well white solid plate. EpMab-37-mG 2af, EpMab-37, and 281-mG 2a -f were serially diluted and added to the target cells. Jurkat cells stably expressing the human FcγRIIIa receptor, and a Nuclear Factor of Activated T-Cells (NFAT) response element driving firefly luciferase, were used as effector cells. The engineered Jurkat cells (75,000 cells in 25 µL) were then added and co-cultured with antibody-treated target cells at 37 • C for 6 h. Luminescence using the Bio-Glo Luciferase Assay System (Promega) was measured with a GloMax luminometer (Promega).

Statistical Analysis
All data are represented as mean ± standard error of the mean (SEM). Welch's t test was conducted for ADCC activity, CDC activity, and tumor weight. ANOVA with Sidak's post hoc test was conducted for tumor volume and mouse weight. GraphPad Prism 8 (GraphPad Software, Inc.) was used for all calculations. p < 0.05 was considered to indicate a statistically significant difference.

Flow Cytometric Analysis
In our previous study, a sensitive and specific anti-EpCAM mAb, EpMab-37 (mouse IgG 1 , kappa), was established using the CBIS method [33]. We first performed flow cytometric analysis using EpMab-37-mG 2a -f against BT-474 and Capan-2 cells and found that EpMab-37-mG 2a -f recognized both cells ( Figure 1A). All data are represented as mean ± standard error of the mean (SEM). Welch's t test was conducted for ADCC activity, CDC activity, and tumor weight. ANOVA with Sidak's post hoc test was conducted for tumor volume and mouse weight. GraphPad Prism 8 (GraphPad Software, Inc.) was used for all calculations. p < 0.05 was considered to indicate a statistically significant difference.

Antitumor Effects of EpMab-37-mG2a-f in Mouse Xenograft Models of BT-474 and Capan-2
BT-474 cells were inoculated into the left flank of mice, followed by the intraperitoneally injection of EpMab-37-mG2a-f or control mouse IgG on days 6, 13, and 18. The tumor volume was measured after the inoculation. EpMab-37-mG2a-f-treated mice exhibited significantly less tumor volume on day 11 (p < 0.01), day 13 (p < 0.01), day 18 (p < 0.01), day 22 (p < 0.01), and day 24 (p < 0.01), compared with control mouse IgG-treated control mice ( Figure 4A). The EpMab-37-mG2a-f treatment resulted in a 51.3% reduction of the tumor volume compared with that of the control mouse IgG on day 24 ( Figure 4C). Tumors from EpMab-37-mG2a-f-treated mice weighed significantly less than tumors from control IgGtreated control mice (52.4% reduction, p < 0.01; Figure 4E). Resected tumors on day 24 are presented in Figure 4E. The total body weights did not significantly differ between the EpMab-37-mG2a-f-treatment and the control groups ( Figure 5A). The body appearance of mice on day 24 post inoculation is shown in Figure 5C, and the body weights' loss and skin disorder were not observed.

Antitumor Effects of EpMab-37-mG 2a -f in Mouse Xenograft Models of BT-474 and Capan-2
BT-474 cells were inoculated into the left flank of mice, followed by the intraperitoneally injection of EpMab-37-mG 2a -f or control mouse IgG on days 6, 13, and 18. The tumor volume was measured after the inoculation. EpMab-37-mG 2a -f-treated mice exhibited significantly less tumor volume on day 11 (p < 0.01), day 13 (p < 0.01), day 18 (p < 0.01), day 22 (p < 0.01), and day 24 (p < 0.01), compared with control mouse IgG-treated control mice ( Figure 4A). The EpMab-37-mG 2a -f treatment resulted in a 51.3% reduction of the tumor volume compared with that of the control mouse IgG on day 24 ( Figure 4C). Tumors from EpMab-37-mG 2a -f-treated mice weighed significantly less than tumors from control IgG-treated control mice (52.4% reduction, p < 0.01; Figure 4E). Resected tumors on day 24 are presented in Figure 4E. The total body weights did not significantly differ between the EpMab-37-mG 2a -f-treatment and the control groups ( Figure 5A). The body appearance of mice on day 24 post inoculation is shown in Figure 5C, and the body weights' loss and skin disorder were not observed. decreased than that of control IgG-treated mice (49.1% reduction; p < 0.01, Figure 4D). The excised tumors of control and EpMab-37-mG2a-f -treated groups on day 27 are shown in Figure 4F. Total body weights were almost similar in both groups ( Figure 5B). Appearance of mice on day 27 after inoculation of cells are shown in Figure 5D.
These results indicate that administration of EpMab-37-mG2a-f effectively suppresses the tumor growth of BT-474 and Capan-2 xenografts.

Discussion
The impact of EpCAM expression on breast cancer prognosis is dependent on intrinsic subtype. In the luminal B HER2-positive and triple negative subtypes, EpCAM expression is associated with an unfavorable prognosis. In contrast, EpCAM expression is associated with a favorable prognosis in the HER2-positive non-luminal subtype [15]. Therefore, the luminal B HER2-positive and triple negative subtypes are potential groups for treatment with EpCAM-targeting therapy. In this study, we investigated the antitumor effect of a defucosylated anti-EpCAM mAb (EpMab-37-mG2a-f) against a breast cancer cell line, BT-474 derived from luminal B HER2-positive subtype [34]. EpMab-37-mG2a-f exhibited superior ADCC and CDC activities in vitro (Figures 2 and 3), and antitumor activity against BT-474 xenograft in nude mice (Figure 4). We previously developed an anti-HER2 mAb (H2Mab-19) and examined ADCC, CDC, and antitumor activities against BT-474 cells [45]. Although the binding affinity of H2Mab-19 and EpMab-37-mG2a-f to BT-474 cells were comparable, EpMab-37-mG2a-f exerted more potent ADCC activity and antitumor effect in vivo. These results are probably due to the defucosylation in EpMab-37-mG2a-f, but not in H2Mab-19. Moreover, EpCAM forms a cis-dimer which further makes In the Capan-2 xenograft, EpMab-37-mG 2a -f and control mouse IgG were injected intraperitoneally into mice on days 6, 14, and 18 after the inoculation of Capan-2 cells. The tumor volume was measured on days 6, 11, 14, 18, 20, 25, and 27. The treatment of EpMab-37-mG 2a -f resulted in a significant inhibition in tumor development on days 11 (p < 0.05), 14 (p < 0.01), 18 (p < 0.01), 20 (p < 0.01), 25 (p < 0.01), and 27 (p < 0.01) compared with that of the control mouse IgG ( Figure 4B). The administration of EpMab-37-mG 2a -f resulted in a 51.3% reduction of tumor volume compared with that of the control mouse IgG on day 27. Furthermore, the tumor weight of the EpMab-37-mG 2a -f -treated mice was significantly decreased than that of control IgG-treated mice (49.1% reduction; p < 0.01, Figure 4D). The excised tumors of control and EpMab-37-mG 2a -f -treated groups on day 27 are shown in Figure 4F. Total body weights were almost similar in both groups ( Figure 5B). Appearance of mice on day 27 after inoculation of cells are shown in Figure 5D.
These results indicate that administration of EpMab-37-mG 2a -f effectively suppresses the tumor growth of BT-474 and Capan-2 xenografts.

Discussion
The impact of EpCAM expression on breast cancer prognosis is dependent on intrinsic subtype. In the luminal B HER2-positive and triple negative subtypes, EpCAM expression is associated with an unfavorable prognosis. In contrast, EpCAM expression is associated with a favorable prognosis in the HER2-positive non-luminal subtype [15]. Therefore, the luminal B HER2-positive and triple negative subtypes are potential groups for treatment with EpCAM-targeting therapy. In this study, we investigated the antitumor effect of a defucosylated anti-EpCAM mAb (EpMab-37-mG 2a -f) against a breast cancer cell line, BT-474 derived from luminal B HER2-positive subtype [34]. EpMab-37-mG 2a -f exhibited superior ADCC and CDC activities in vitro (Figures 2 and 3), and antitumor activity against BT-474 xenograft in nude mice (Figure 4). We previously developed an anti-HER2 mAb (H 2 Mab-19) and examined ADCC, CDC, and antitumor activities against BT-474 cells [45]. Although the binding affinity of H 2 Mab-19 and EpMab-37-mG 2a -f to BT-474 cells were comparable, EpMab-37-mG 2a -f exerted more potent ADCC activity and antitumor effect in vivo. These results are probably due to the defucosylation in EpMab-37-mG 2a -f, but not in H 2 Mab-19. Moreover, EpCAM forms a cis-dimer which further makes a biologically relevant oligomeric state (e.g., cis-tetramer, trans-tetramer, and trans-octamer) according to several experimental observations [56]. These oligomeric structures of EpCAM could promote the clustering of anti-EpCAM mAbs, which might help the FcγRIIIa engagement on effector cells, and potentiate the ADCC activity.
EpCAM is an important cell surface molecule to collect CTCs [10]. In a prospective study of pancreatic cancer patients, the CTC in the peripheral blood affect the outcome of patients independent from other risk factors, including adjuvant chemotherapy [12]. Furthermore, EpCAM-sorted pancreatic adenocarcinoma cells from surgically resected tumors could be applied to the analysis of tumor-cell-intrinsic chromatin accessibility patterns. A chromatin accessibility signature and associated transcriptional factors (ZKSCAN1 and HNF1β) are significantly correlated with pancreatic cancer prognosis [57]. This information could contribute to the selection of patients to be applied to anti-EpCAM mAb therapy. Recently, CTC expansion techniques have been developed to evaluate the characteristics of CTCs. The techniques include the two-dimensional (2D) long-term expansion, 3D organoids/spheroids culture, and in vivo xenografts/metastasis formation in immunodeficient mice [58]. It would be worthwhile to investigate the effect of EpMab-37-mG 2a -f on the 2D and 3D CTC expansion in vitro and anti-metastatic activity in vivo.
We have been investigating the critical epitope of EpMab-37 [33], and recently identified that Arg163 of EpCAM is the most important residue of the EpMab-37 epitope (submitted). Among clinically tested mAbs, no mAb recognized above region, suggesting that EpMab-37 possesses a unique epitope and a different mode of actions. In this study, we did not examine the EpCAM-internalizing activity by EpMab-37-mG 2a -f. Furthermore, the relationship between the internalizing activity and the epitope has not been investigated. In future study, we would like to evaluate it for the development of antibody-drug conjugates.
Anti-EpCAM mAb can be used for a bispecific Ab with anti-MET mAb [59]. MM-131 is a bispecific Ab that is monovalent against MET, but it exhibits high avidity to EpCAM through binding to a single chain Fv of an anti-EpCAM mAb, MOC31 [60]. MM-131 exhibits antagonistic activity that interferes both ligand-dependent and ligand-independent MET signaling and induces the receptor down-regulation [59]. MCLA-128 is a bispecific Ab for HER2 and HER3. MCLA-128 can inhibit heregulin (a HER3 ligand)-mediated signaling of HER2/HER3 heterodimer and suppress tumor cell growth via the suppression of PI3K/Akt signaling [61]. Clinical studies on MCLA-128 are ongoing in patients with breast cancer, pancreatic cancer, and non-small cell lung cancers [62]. In the future study, we would like to apply EpMab-37 for the combination therapy with anti-HER2 mAbs or develop a bispecific mAb targeting EpCAM and HER2.