Abstract

CD19 is a B cell lineage-specific transmembrane signaling protein that controls differentiation and proliferation. CD19 is an attractive therapeutic target due to its high level of expression in numerous B cell malignancies, as well as its lack of expression on non-B cells. Here we report the in vitro anti-tumor activity of a novel humanized monoclonal anti-CD19 Ab (CD19-IgG1, aka XENP5603) and its Fc engineered counterpart (XmAb™CD19, aka XENP5574). XENP5603 induced direct apoptosis in normal CD19+ B cells, but not NK cells, T cells, or monocytes, as determined by flow cytometric staining with annexin V and propidium iodide. XENP5603 also induced significant levels of apoptosis in a number of lymphoblastoid cell lines, including Ramos, Raji, 697, NALM6, and RS4;11 cells. Treatment of primary chronic lymphocytic leukemia (CLL) cells with XENP5603 induced significant cell death in all patients tested (mean, 36% apoptotic cells at 24 hours; range, 13–66%, p < 0.001). Similar apoptosis was noted in cells from a subset of patients (4 of 9) with CD19+ primary acute lymphoblastic leukemia (ALL). Apoptosis of CLL cells treated with XENP5603 was not associated with cleavage of caspase-3, caspase-8, caspase-9, or PARP, but was associated with upregulation of Bim, suggesting a caspase-independent mechanism of cell death. NK cells from normal donors exhibited high levels of ADCC in response to B cell lines coated with XENP5603. Furthermore, NK cells from CLL patients mediated significant ADCC against autologous CLL cells in the presence of XENP5603 (mean, 15% specific lysis at an E:T ratio of 25:1; range, 8–24%; p = 0.04 vs. the negative control Ab). ADCC activity was further increased in the presence of XENP5574, which has the same antigen-recognition sequences as XENP5603 but which contains two mutations in the Fc region that increase FcγRIIIa affinity (mean, 39% specific lysis at an E:T ratio of 25:1; range, 29–51%; p = 0.02 vs. the negative control Ab). ADCC mediated by either CD19 Ab was also significantly higher than that mediated by an equivalent concentration of rituximab (mean, 39% specific lysis with XENP5574 vs. 12% with rituximab; p < 0.001). ADCC in the presence of either Ab was further increased in the presence of the NK cell-activating cytokine IL-2, suggesting that these antibodies might be effectively combined with immune stimulatory adjuvants. Furthermore, NK cell ADCC against CLL cells in the presence of CD19 Abs was found to be dependent on perforin/granzyme release, as treatment with 3,4-dichloroisocoumarin (which inhibits granzyme enzymatic activity) or EGTA (which prevents release of cytotoxic vesicles) potently inhibited ADCC activity. Collectively, these studies provide evidence of the autologous innate immune-mediated cytotoxicity and direct apoptotic activity of XENP5603 and XENP5574. In addition, engineering to enhance FcγRIIIa binding enhances autologous ADCC, providing support for further pre-clinical development of XENP5574 in CD19+ malignancies, including CLL and ALL.

Author notes

Disclosure: No relevant conflicts of interest to declare.