Abstract B122: Humanized SCID mouse as tool for evaluation of B-cell-targeting therapies.

Abstract

Abstract Background: Antibody strategies are of huge importance in targeted therapies drug development. However, most of the time, antigen is specific of human species with bad cross-reactivity with mouse antigen, leading to difficulties to apply regular mouse models. The “gold” model for the preclinical evaluation of antibody strategy should have cells expressing human antigen, to analyze Fab binding activity but should also contain human immune effector cells in order to be able to test the antibody Fc mediated activity (Antibody Dependent Cell Cytotoxicity, ADCC). We here propose the use of CB17-SCID mouse strain having the ability to exhibit complement activation, useful to test the CDC (Complement Dependent Cytotoxicity) activity of antibody. This strain of mice supports the engraftment of human peripheral blood mononuclear cells (hPBMCs), containing human effector cells as NK cells and moncoytes, responsible of ADCC activity. In our case, the target of interest is expressed on human B lymphocytes, whatever the tumoral or healthy origin. The engrafted hPBMCs contained both the NKs as effector cells but also the human B cells as target cells. Methods: Whole body irradiated (D-3) female CB17-SCID mice were treated twice (Q7D×2) (starting at D-2) with subcutaneous injections of mouse NK cell-depleting antibody TM-Beta 1 and received an intraperitoneal injection of freshly prepared hPBMCs at D0. At D11 mice received a single intravenous injection of trastuzumab used as negative control, of new B cell targeting antibody or rituximab used as positive control (10 mice per group). At D15, mice were sacrificed to collect spleen. Single cell suspensions were prepared from each collected spleen, labeled with mCD45, hCD45, hCD19 and hCD20 antibodies before FACS analysis. Absolute cell number as well as percentage of human B cells were calculated. Statistical comparisons were performed using Mann-Whitney U test after having discarded outlier values by Dixon Q test. Results: All mice injected with hPBMCs were found humanized at the time of sacrifice. Percentage of human leukocytes in mouse spleen ranged from 0.6 to 59.4% with a mean value of 12.9 ± 12.5 % and a median value of 8.3 %. Considering the human B-cell percentage, only 3 out of 80 values were considered as outliers. In the trastuzumab treated group (10 mg/kg), the spleen contained about 6 % of human B cells (within the human leukocytes, i.e. hCD45+). When the mice were treated with rituximab at 2 mg/kg, a significant decrease in the percentage of human B cells was evidenced in the mouse spleen, with about 1.5% of human B cells. A dose-dependent decrease in human B cells was also observed in mouse spleen when mice were treated with the new B-cell targeting antibody (1.4 ± 1.1 % and 3.8 ± 2.2 % of human B cells for 10 mg/kg and 0.02 mg/kg respectively). The decrease starts to be significant from 0.1 mg/kg of the new B-cell targeting antibody. Conclusions: These results show that this humanized mouse model is suitable to evaluate B-cell targeting therapies. Humanization level as well as sensitivity to treatment are compatible with dose-response analysis and then offer the possibility to use this model for benchmarking of human B-cell targeting therapy tested on human target in the presence human effectors. Moreover, this model could be applied to other targets expressed on human immune cells. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr B122.