Complement activation impacts B-cell depletion by both type I and type II CD20 monoclonal antibodies

To the editor:

With great interest, we have read the article by Beers et al¹  documenting a potent B-cell depleting ability for type II (or tositumumab-like) CD20 antibodies. The authors conclude that complement-dependent cytotoxicity (CDC) is of little importance for B-cell depletion induced not only by type II, but also by type I (or rituximab-like) CD20 antibodies. Unfortunately, the experimental design chosen by Beers et al does not allow such a conclusion, and the data in our view appear to suggest the reverse.

First, Beers et al show that tositumomab also induces CDC, albeit at a lower level than rituximab. Similarly, we also routinely observe significant complement-mediated lysis of lymphoma cells by tositumomab (Figure 1). Second, to specifically address the role of complement, the authors generated a mouse IgG2a version of rituximab (Rit-m2a) and introduced a lysine-into-alanine mutation at position 322 (K322A) to abrogate C1q binding and CDC. A study by Idusogie et al,²  however, previously demonstrated this mutation to be insufficient to remove CDC activity, and K322A-mutated rituximab retained more than 60% of its CDC capacity at near-physiologic complement levels. Indeed, we have confirmed that for other CD20 antibodies, such as HuMab 7D8,³  significant CDC activity remains after mutating K322 (Figure 1). Hence, on basis of their data, Beers et al cannot exclude complement activation as an in vivo mechanism of action for B-cell depletion by either tositumomab or rituximab.

Notably, the authors point out that it is not understood why the genetic background of the hCD20-transgenic mice used, strongly influenced CD20-induced B-cell depletion, with stronger depletion in BALB/c than in C57BL/6 mice. In this context, it is important to note that serum complement activity differs significantly between mouse strains and varies with gender and age.^4-6  In our own experience, BALB/c mice (males in particular) have relatively high levels of complement activity compared with C57BL/6 mice. In our view, this may explain the observed differences and identify complement activation as a significantly contributing factor to the in vivo mechanism of action of CD20 antibodies.

The contribution of complement activation to B-cell depletion by CD20 antibodies is strongly supported by several studies. Kennedy et al have shown that CD20⁺ cells are depleted concomitantly with the consumption of complement in patients during treatment with rituximab.⁷  Golay et al demonstrated that therapeutic activity against established B-cell tumors by rituximab required complement.⁸  Furthermore, Gong et al⁹  showed depletion of resident B-cells, such as those in the splenic marginal zone, to be critically dependent on complement. Notably, as above, the magnitude of complement-mediated B-cell depletion differed between mice of distinct genetic backgrounds. Circulatory B-cells (such as peripheral blood, lymph node, and splenic follicular B-cells), in contrast, can also be effectively cleared by Fc-receptor–mediated mechanisms in the absence of complement.^9,10  The observations by Beers et al, which document depletion of circulating (CFSE-labeled) B-cells in mice deficient for C1q serves to confirm this notion.

We conclude that the data presented by Beers et al do not support conclusions that minimize a role for complement in B-cell depletion by CD20 antibodies. In contrast, the increased B-cell depletion in complement-sufficient BALB/c mice in fact supports a critical contribution. Therapeutic CD20 antibodies may engage multiple mechanisms to deplete B-cells, of which Fc-mediated mechanisms seem sufficient for clearance of circulatory cells, whereas depletion of resident normal and tumor B-cells appears to be critically dependent on complement.