Abstract
Abstract 2404
The use of the anti-CD20 monoclonal antibody (mAb) rituximab (RTX) in the immunotherapy of cancer has led to substantial advances in the treatment of B cell malignancies. However, many patients are refractory to RTX therapy and a substantial fraction of responding patients suffer relapses. In some patients the loss or lack of efficacy can be explained because levels of CD20 are substantially reduced on target B cells as a consequence of RTX therapy. This phenomenon may be due to growth and proliferation, over a period of at least several months, of a CD20-negative B cell clone. In contrast, we have reported that levels of CD20 expressed on circulating CLL cells can be reduced > 90% in less than one hour following RTX infusion. These clinical observations, as well as our results seen in vitro and in mouse models, suggest that after the body's effector mechanisms that promote mAb-mediated cell killing are exhausted or saturated, an alternative reaction predominates. RTX-CD20 complexes on targeted B cells are rapidly removed from the B cells by monocytes and macrophages through an Fc receptor-mediated endocytic process known as trogocytosis, or the shaving reaction. Beers et al. recently reported (Blood, 2010) that human CD20 transgenic mouse B cells opsonized with Rit m2a, a mouse IgG2a mAb with RTX activity, internalized both bound mAb and up to half of cell-surface CD20 after incubation for 2 hrs at 37C. They also reported that on average 40% of cell-surface CD20 was internalized when CLL cells opsonized with RTX were similarly treated. It would thus appear that as a consequence of RTX therapy, both trogocytosis by effector cells and direct internalization by B cells can promote loss of CD20 from B cells targeted by RTX. Both of these reactions will reduce the efficacy of RTX therapy, and a key question must focus on the relative rates of trogocytosis and internalization in promoting CD20 loss in vivo.
To address these questions we examined human CD20 transgenic mouse 38C13 B cells (a kind gift from Dr. Josee Golay) for RTX-mediated CD20 internalization versus shaving. To test for internalization of cell-bound RTX, CD20+ 38C13 cells were reacted with Al488- or biotinylated RTX for 10 min to 2 hours at 37C, then washed and secondarily probed with the following reagents: 1, Al647 mAb HB43, specific for the Fc region of human IgG; 2, Al488-streptavidin to detect surface-bound biotinylated RTX; 3, Anti-Al488 antibodies that induce quenching of surface-bound Al488 RTX; or 4, The cells were subjected to an acid wash, re-equilibrated at neutral pH, and re-tested for Al488 RTX binding. Our flow cytometry results reveal rapid and stable binding of Al488-labeled or biotinylated RTX to the CD20+ 38C13 cells. After two hours at 37C, the amount of cell-bound RTX that had been internalized, defined by the 4 different assays, was much less than that reported by Beers, and ranged between 5 and 25%, suggesting that the degree of internalization may be model-dependent. Moreover, reaction of acceptor THP-1 cells (a monocytic line) with the RTX-opsonized cells leads to rapid shaving and removal of ~ 90% of bound RTX and CD20 from the 38C13 cells. After just 45 min at 37C in the presence of adhered THP-1 monocytes, there is a 10-fold reduction in fluorescence intensity signal (molecules of equivalent soluble fluorochrome) on the 38C13 cells attributable to either bound Al488 RTX or due to secondary development with Al647 anti-human IgG mAb HB43. Alternatively, Al488 RTX-opsonized CD20+ 38C13 cells were reacted in solution with THP-1 monocytes at varying THP-1/38C13 cell ratios. Within 45 min at 37C at a 3/1 ratio more than half of cell bound Al488 RTX and CD20 were removed from the 38C13 cells, and large amounts of Al488 RTX were found on the acceptor THP-1 cells. Our results indicate that the shaving reaction is considerably more rapid and indeed leads to the removal of much more bound anti-CD20 mAb RTX (and CD20) than the internalization reaction, suggesting that the shaving reaction is most likely primarily responsible for the rapid down-modulation of CD20 that is observed when CLL patients are treated with RTX.
No relevant conflicts of interest to declare.
This icon denotes an abstract that is clinically relevant.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal