Potent inhibition of local and disseminated tumor growth in immunocompetent mouse models by a bispecific antibody construct specific for Murine CD3

Abstract

Bispecific single-chain antibody constructs specific for human CD3 have been extensively studied for antitumor activity in human xenograft models using severe combined immunodeficient mice supplemented with human T cells. High efficacy at low effector-to-target ratios, independence of T cell costimuli and a potent activation of previously unstimulated polyclonal T cells were identified as hallmarks of this class of bispecific antibodies. Here we studied a bispecific single-chain antibody construct (referred to as 'bispecific T cell engager', BiTE) in an immunocompetent mouse model. This was possible by the use of a murine CD3-specific BiTE, and a syngeneic melanoma cell line (B16F10) expressing the human Ep-CAM target. The murine CD3-specific BiTE, called 2C11x4-7 prevented in a dose-dependent fashion the outgrowth of subcutaneously growing B16/Ep-CAM tumors with daily i.v. injections of 5 or 50 microg BiTE which was most effective. Treatment with 2C11x4-7 was effective even when it was started 10 days after tumor cell inoculation but delayed treatments showed a reduction in the number of cured animals. 2C11x4-7 was also highly active in a lung tumor colony model. When treatment was started on the day of intravenous tumor cell injection, seven out of eight animals stayed free of lung tumors, and three out of eight animals when treatment was started on day 5. Our study shows that BiTEs also have a high antitumor activity in immunocompetent mice and that there is no obvious need for costimulation of T cells by secondary agents.

Fig. 1

Structure and in vitro efficacy of BiTEs diL2Kx4-7 and 2C11x4-7 a Molecular design of human CD3-specific BiTE diL2Kx4-7 and murine CD3-specific BiTE 2C11x4-7. Rectangles depict variable immunoglobulin domains, and connecting lines are linker sequences. N, N terminus; C, C terminus. b In vitro efficacy of BiTE diL2Kx4-7. Redirected lysis of Kato III tumor cells was tested with human PBMC in the presence of increasing BiTE concentrations for a 20-h assay period. c In vitro efficacy of BiTE 2C11x4-7. Redirected lysis of human Ep-CAM cDNA transfected NALM-6 B lymphoma cells was tested with murine CD3 cell-enriched splenocytes in the presence of increasing BiTE concentrations for a 45-h assay period. Effector and target cells were mixed at an E:T ratio of 10:1 Error bars indicate SEM of triplicate measurements. Cell lysis was assessed via a FACS based cytotoxicity assay described in Materials and methods

Fig. 2

Pharmacokinetic parameters of 2C11x4-7 and diL2Kx4-7 C57BL/6 mice were intravenously injected with 150 μg 2C11x4-7 or diL2Kx4-7 and groups of five mice each bled at different time points. Plasma concentrations of BiTEs were quantified by specific ELISAs. a Plasma concentration versus time profiles of 2C11x4-7 and diL2Kx4-7 in C57BL/6 mice. b Mean pharmacokinetic parameters of 2C11x4-7 and diL2Kx4-7. The dashed line indicates the lower limit of quantification of 1 μg/ml

Fig. 3

Dose-dependent inhibition of subcutaneous SW480 tumor growth in NOD/SCID mice by diL2Kx4-7. Five ×10⁶ SW480 cells were mixed with 5×10⁶ human PBMCs (PBMC:T ratio of 1:1) and subcutaneously injected into the right flanks of six female NOD/SCID mice per group. Treatment with 0.1 μg/mouse a and 1 μg/mouse b of diL2Kx4-7 (open triangles) and PBS control buffer (black diamonds) was started 1 h after the injection of tumor cells and treatment was repeated for four consecutive days. Tumor size was measured three times a week with calipers. Arrows indicate time points of diL2Kx4-7 administration. Asterisks indicate statistical differences between the treatment and the control group (*P≤0.05; **P≤0.01 as determined with the Student’s t test)

Fig. 4

Effect of the 2C11x4-7 dose on subcutaneous B16F10/Ep-CAM tumor growth in immunocompetent C57BL/6 mice. B16F10/Ep-CAM or B16F10 wildtype (B16F10WT) tumor cells (7.5×10⁴) were subcutaneously injected into the right flanks of 5 C57BL/6 mice per group and tumor size measured three times a week. Treatment with PBS control buffer and different doses of 2C11x4-7 (a 0.05), (b 0.5), (c 5) and (d 50 μg/injection) was started 1 h after the injection of tumor cells and treatment was repeated for four consecutive days. One group of control animals (black diamonds) was implanted with B16F10/Ep-CAM and treated with PBS control buffer, whereas another group of control animals (black squares; c) was implanted with B16F10WT target-negative tumor cells and treated with 5 μg 2C11x4-7/injection. For the control groups, mean values of tumor volume are shown, whereas for the 2C11x4-7 treatment groups, individual tumor growth curves are depicted (open triangles). Error bars indicate standard deviation calculated for the mean value of tumor growth curves. Arrows indicate time points of 2C11x4-7 administration

Fig. 5

Effect of delayed 2C11x4-7 treatments on the growth of established subcutaneous B16F10/Ep-CAM tumors in immunocompetent mice. B16F10/Ep-CAM cells (7.5×10⁴) were subcutaneously injected into the right flanks of eight immunocompetent C57BL/6 mice per group and tumor size measured three times a week. Treatment with 5 μg 2C11x4-7/mouse (open triangles) was started at different time points (day 0, 5, 7 and 10) after the injection of tumor cells and treatment was repeated for nine consecutive days. One group of control animals (black diamonds) was treated with PBS control buffer whereas another group of control animals (black squares) was treated with 5 μg/injection of an irrelevant BiTE molecule from days 5–14. Small double asterisks indicate highly significant differences (P<0.0005) compared to the PBS control group

Fig. 6

Effects of early and late treatment with BiTE 2C11x4-7 on the development of B16F10/Ep-CAM lung tumor colonies in C57BL/6 mice. B16F10/Ep-CAM cells (1×10⁵) were intravenously injected into the tail vein of eight immunocompetent C57BL/6 mice per group. Animals were treated with 5 μg/mouse 2C11x4-7 on day 0 (early treatment) or day 5 (late treatment), or PBS control buffer 1 h after intravenous B16F10/Ep-CAM inoculation, and treatment was repeated for nine consecutive days. 2C11x4-7 treatment groups were split into an early (day 0–9) and late treatment group (day 5–14). Twenty-one days after tumor cell inoculation animals were sacrificed, lungs removed, and the number of lung tumor colonies quantified as a measure of efficacy. a Photographs of lungs of animals from day 21. b Number of lung tumor colonies as individual readings (symbols) and the mean numbers (short lines) are given for each treatment group (P=7×10⁻¹² and 2×10⁻¹¹ for early and late treatment, respectively, as determined by the Student’s t test).