Comparative Assessment of Different Clinically Utilized CD20 Directed Antibodies in Chronic Lymphocytic Leukemia (CLL) Cells Reveals Divergent NK-Cell, Monocyte and Macrophage Properties,

Abstract 3717

The anti-CD20 antibody rituximab represents the first therapy to contribute to prolongation of survival in CLL yet controversy exists as to how it promotes tumor elimination. Pre-clinical studies in CLL have suggested that innate immune cells, complement, and direct antibody killing might contribute to anti-CD20 antibody efficacy. Murine models of B-cell depletion by CD20 antibodies have shown monocytes to be the most important effector cell, although significant controversy around this point exists. Given the success of rituximab, second generation anti-CD20 antibodies (Ofatumumab and GA101 are now in phase III testing in CLL with reported improved direct (GA101), complement (Ofatumumab), or NK cell effector cell killing (GA101). To date, direct comparison of these three antibodies for direct, complement, and effector cell engagement has not been performed. Additionally, the impact of afucosylation engineering of GA101 on monocyte and macrophage function has not been reported. As a Type II anti-CD20 antibody, GA101 mediated significantly increased cell death (∼25%) without Fc crosslinking when compared to Ofatumumab or rituximab (5–8%). Direct cytotoxicity assessment of 19 pts with cross-linking demonstrated GA101 (25%) to mediate significantly greater (p=0.0003) killing than rituximab (15%) but not Ofatumumab (20%). Complement mediated killing was significantly increased over media with Ofatumumab (∼30%), whereas an average of only 5–10% killing was observed with rituximab or GA101. Immobilized GA101 significantly increased NK cell activation as detected by IFNg production and CD107a induction (p=0.005) more than rituximab or Ofatumumab. In addition, GA101 mediates 2 to 3 fold greater NK cell-mediated Antibody Dependent Cellular Cytotoxicity (ADCC) compared to Ofatumumab or rituximab at higher antibody concentrations, but not at concentrations less than 0.05 mg/ml. This enhanced ADCC with GA101 is seen with normal NK cells as well as NK cells from CLL patients. Ofatumumab mediated significantly greater (p=0.0001) NK cell ADCC than rituximab. Given the importance of macrophages in depletion of anti-CD20 tumors in mouse models of lymphoma, we next focused on this. The three anti-CD20 antibodies show Antibody Dependent Cellular Phagocytosis (ADCP) capability with Monocyte Derived Macrophages (MDM) against CLL B cells, with Ofatumumab exhibiting the greatest ADCP compared to GA101 (60% vs 40%; p=0.0036). In addition, primary monocytes stimulated with immobilized GA101 show less TNFa release, when compared to rituximab or Ofatumumab. To investigate this mechanistically, monocytes were stimulated with plate bound rituximab, Ofatumumab or GA101. Compared to the non Fc engineered antibodies (rituximab and Ofatumumab) Fc engineered GA101 induces reduced pan tyrosine phosphorylation and phosphorylation of ERK. No differences in phosphorylation of FcgRIIa or FcgRIIb was observed between the engineered and non-engineered antibodies. Further mechanistic studies to elucidate these differences in monocytes based upon antibody afucosylation changes, or differential phosphorylation of FcgR are ongoing. Collectively, our data indicate that GA101 and Ofatumumab are both superior to rituximab against CLL cells and have differential properties with respect to apoptosis, CDC, and effector cell-mediated killing. GA-101 mediates the most potent NK cell mediated killing at high concentrations whereas Ofatumumab has the greatest monocyte activation and phagocytosis despite absence of any engineering. These findings have relevance in the choice of the optimal CD20 antibody for treatment of CLL and combination strategies used.