Commentary on Abstracts #633 and #2523
Commentary on Abstracts #633 and #2523
How the monoclonal antibodies work in lymphoid malignancies remains unclear. Several studies presented at the 2001 American Society of Hematology (ASH) meeting provided insight into their mechanisms of action, and also into how cells become resistant to their effects.
Rituximab (Rituxan) and other monoclonal antibodies are presumed to act through antibody-dependent cellular cytotoxicity (ADCC) with activation of natural killer cells and macrophages, complement-dependent cytotoxicity (CDC), and induction of apoptosis. However, the relative roles of ADCC and complement activation are controversial.
Although virtually all patients with follicular non-Hodgkin’s lymphoma (NHL) are CD20-positive, only half of patients respond to treatment with rituximab (McLaughlin et al: J Clin Oncol 16:2825-2833, 1998). The mechanisms involved in cellular resistance to rituximab are also unknown, but may involve differences in pharmacokinetics, abnormalities of effector cells, poor tumor penetration, or other factors. Golay et al (Blood 95:3900-3908, 2000) studied four follicular NHL and one Burkitt cell lines, three fresh follicular NHL samples, and normal B cells in vitro to determine how these cells become resistant to the effects of rituximab. Although apoptosis in response to the antibody was not detected, all of the cells were affected by ADCC. Human complement-mediated lysis was highly variable among the cell lines. Blocking CD55 and CD59 with specific antibodies significantly increased CDC in follicular NHL cells. They suggested that the levels of these factors might be useful markers of potential response to the antibody.
Treon et al (J Immunother 24:263-271, 2001) also found that expression of the complement regulator CD59 was associated with resistance to rituximab-mediated lysis of multiple myeloma and NHL cell lines. However, more recent studies have failed to detect a correlation between CD46, CD55, and CD59 on follicular lymphoma cells, or complement-dependent lysis and clinical outcome following rituximab therapy (Weng and Levy: Blood 98:1352-1357, 2001). Additional information was presented at the ASH meetings by Golay et al (abstract #633), who presented an evaluation of cells from patients with CLL, prolymphocytic leukemia, and MCL for their expression of CD20 and found a correlation with lysis in vitro, regardless of diagnosis. Levels of CD55 and CD59 did not correlate with responsiveness to complement lysis, but blocking those antigens resulted in greater cell kill. Fludarabine (Fludara) may up-regulate CD55 and CD59, thus enhancing the activity of the antibody (Golay et al: Blood 96:339a[abstract 1463], 2000). Whether these in vitro observations are clinically relevant remains to be determined.
It is possible that in some cases, the antibody fails to bind to the tumor. However, it is more likely that binding occurs but neither activates the necessary signaling pathways nor recruits the necessary effector cells. Natural killer (NK) cells and macrophages that express the FcgRIIIa-positive receptor are believed to play an important role in the efficacy of rituximab. This receptor is encoded by the FCGR3A gene, which can have either phenylalanine (F) or valine (V) at the 158 position. When valine is present, the affinity for IgG1 is higher and ADCC is increased.
Cartron and coworkers (abstract #2523) studied the pharmacogenomics in 49 patients treated with rituximab and found that 20% were homozygous for FCGR3A-158V and 35% were homozygous for FCGR3A-158F, whereas 45% were heterozygous for FCGR3A-158F. Higher response was associated with FCGR3A-158V. This assay may lead to a better understanding of the effects of rituximab on the lymphoma cell, and has potential to predict which patients are more suited for rituximab therapy.