Ofatumumab, a Human Mab Targeting a Membrane-Proximal Small-Loop Epitope On CD20, Induces Potent NK Cell-Mediated ADCC.

CD20 represents a well established target for immunotherapy of B-cell malignancies such as follicular lymphoma (FL). The standard therapy for FL is the monoclonal antibody (mAb) rituximab given as a single agent or combined with chemotherapy. Antibody-dependent cytotoxicity (ADCC) has been suggested to be an important in vivo mechanism of action of CD20 mAb as differences in response rates to rituximab exist between individuals expressing different allotypes of the polymorphic Fc receptor FcgRIIIa. Individuals homozygous for the FcgRIIIa 158V allotype respond significantly better than individuals expressing the FcgRIIIa 158F allotype. Ofatumumab, a unique human mAb targeting a novel membrane-proximal small-loop epitope on CD20, has been shown to be associated with highly efficient killing of primary tumor cells through complement-dependent cytotoxicity. Here we studied the ability of ofatumumab to induce ADCC by NK cells from FcgRIIIa 158V and 158F homozygous healthy donors and compared it to that of rituximab.

Blood was drawn (with informed consent) from ten healthy volunteers homozygous for FcgRIIIa 158V, and ten homozygous for FcgRIIIa 158F, selected from a panel of 479 donors and matched for age, sex and race. NK cells were purified, after which we assessed Fc-mediated antibody binding to NK cells, and ADCC, in a blinded study. Glycan profiling was performed for ofatumumab and rituximab by HPAEC-PAD analysis. MAb binding was measured by flow cytometry in a competition binding assay in which binding of FITC-labelled CD16 mAb 3G8 was blocked by the binding of ofatumumab or rituximab. ADCC was assessed by measuring Europium release from the B cell line ARH77 target cells upon incubation with a concentration curve of ofatumumab or rituximab in the presence of NK effector cells (effector:target ratio 5:1).

Statistical analysis was performed using a non-linear mixed effect model fitting sigmoidal concentration-responses curves to data from each donor for each antibody. The model provided estimates of mean EC50 for each antibody and donor group, and allowed differences between these means to be tested for statistical significance.

Biochemical analysis of clinical grade ofatumumab and rituximab batches indicated their quality to be comparable with similar amounts of aggregates present. The level of core-fucosylation was determined to be similar and ranged from 4 – 6 % for both mAbs. Monomeric ofatumumab bound more strongly to NK cells expressing FcgRIIIa 158 V/V (EC50: 900 mg/ml [95% CI 680-1200 mg/ml]) than FcgRIIIa 158 F/F (EC50: 3970 mg/ml [95% CI 2940-5370 mg/ml]). The approximately 4.4 fold difference in affinity was statistically significant (p<0.0001). A similar 4.2 fold difference (p<0.0001) was found for rituximab (FcgRIIIa 158 V/V EC50: 1370 mg/ml [95% CI 1020-1850 mg/ml] and FcgRIIIa 158 F/F EC50: 5720 mg/ml [95% CI 4170-7850 mg/ml]). Rituximab bound 1.4 fold less tightly to both FcgRIIIa allotypes than ofatumumab (p<0.0001).

Ofatumumab induced a potent NK-mediated ADCC with both FcgRIIIa 158V/V expressing (EC50: 6.4 ng/ml [95% CI 4.8-8.5 ng/ml]) as well as FcgRIIIa 158 F/F expressing NK cells (EC50: 17.6 ng/ml [95% CI 13.4-23.1 ng/ml]). The observed ∼2.7 fold difference in potency was significant (p< 0.0001). Notably, higher concentrations of rituximab compared to ofatumumab were required to induce ADCC by NK cells obtained from both types of donors. Thus, rituximab induced ADCC by FcgRIIIa 158V/V and 158F/F expressing NK cells at EC50s of 12 ng/ml [95% CI 11-13 ng/ml] and 31 ng/ml [95% CI 28-34 ng/ml], respectively. This 1.8 fold difference in ADCC potency between ofatumumab and rituximab was statistically significant (p<0.0001).

We assessed FcgRIIIa affinity and potency to induce ADCC by purified NK cells for ofatumumab and rituximab in a blinded study. Expected differences in affinity for the 158V and 158F allotypes of FcgRIIIa were observed for both ofatumumab and rituximab. These differences correlated with a stronger ADCC by FcgRIIIa 158V/V compared to 158F/F expressing NK cells. Significantly, ofatumumab was able to induce ADCC more potently than rituximab for both Fc receptor allotypes. Ofatumumab binds CD20 stably and at a distinct membrane-proximal epitope compared to rituximab. Our data suggest that these binding characteristics may positively impact ofatumumab's ability to direct killing of tumor cells via ADCC.

Craigen:GSK: Employment, Equity Ownership. Mackus:Genmab: Employment, Equity Ownership. Engleberts:Genmab: Employment, Equity Ownership. Miller:GSK: Employment, Equity Ownership. Speller:GSK: Employment, Equity Ownership. Chamberlain:GSK: Employment, Equity Ownership. Davis:GSK: Employment, Equity Ownership. McHugh:GSK: Employment, Equity Ownership. Bullmore:GSK: Employment, Equity Ownership. Cox:GSK: Employment, Equity Ownership. Wetten:GSK: Employment, Equity Ownership. Perdock:Genmab: Employment, Equity Ownership. Bakker:Genmab: Employment. van de Winkel:Genmab: Employment, Equity Ownership. Parren:Genmab: Employment.