Do Data from BLyD Support the Trojan Horse Concept

The use of BLyD has, therefore, provided us with a powerful new tool with which to treat a number of autoimmune conditions and, perhaps more importantly for the future, to probe underlying pathogenic mechanisms. It was initially thought by some that BLyD would not work in such autoantibody-associated conditions, since rituximab-based ablation of circulating B cells in lymphoma for a period of many months was not associated with major falls in immunoglobulin levels (McLaughlin, 2001; Grillo-Lopez et al., 2002). Clinical responses seen in a number of conditions showed this prediction to be wrong.

4.1. Autoantibody Levels Fall Selectively Compared with

Antimicrobial and Total Immunoglobulin Levels

Early experience showed that significant, apparently selective falls in autoantibody levels could occur (Edwards and Cambridge, 2001; Specks et al., 2001). This is confirmed by immunodynamic studies in RA and SLE (Leandro et al., 2002; Cambridge et al., 2003; Edwards et al., 2004). It appears that B cell depletion may be more useful than predicted because of a differential sensitivity of autoantibody-secreting cells. Correlations between decline in C-reactive protein and RhF levels are almost linear in many RA patients. In SLE, the drop in anti-DNA antibodies mirrors renal disease most closely (Leandro et al., 2002a). In other conditions antibodies are more difficult to quantify. In contrast, levels of antimicrobial antibodies, to pneumococcal capsular polysaccharide (PCP) and tetanus toxoid, in most patients with RA and lupus did not decrease significantly, even after several cycles of treatment (Cambridge et al., 2003).

Since circulating antibody comes chiefly from plasma cells rather than B cells, it seems likely that differential attrition of autoantibody levels, compared with antimicrobial responses, following BLyD reflects relatively short half-lives for at least some autoantibody-secreting plasma cells. Plasma cells do not themselves express significant levels of CD20 and would be unlikely to be ablated by the protocols used, although cyclophosphamide may be modestly plasmacytolytic.

There is also a suggestion that marginal zone B cells in the spleen, which secrete antibodies to T cell-independent, often carbohydrate, antigens, may be relatively resistant to rituximab. This could explain preservation of levels of antibodies to PCP seen in RA and lupus (Cambridge et al., 2003; G. Cambridge, M. J. Leandro, J. C. W. Edwards, manuscript in preparation). In fact there is sometimes a transient rise in such levels after B cell depletion, suggesting that marginal zone-derived plasma cells may take over space made available by plasma cells derived from rituximab-sensitive B cells (G. Cambridge, unpublished data).

4.2. Total Immunoglobulin Levels May Fall after Repeat Cycles

An observation of particular interest is that after three courses of BLyD based on rituximab serum IgM in several RA patients has fallen to undetectable levels (M. J. Leandro, manuscript in preparation). During the first cycle of depletion, falls of up to 40% are seen. It seems that certain subpopulations of B cells, and their daughter plasma cells, may not be regenerated and that repeated cycles may deplete any remaining relatively long-lived IgM-secreting plasma cell populations. IgM is thought to provide a frontline defense against infection through "natural antibody," capable of recognizing, with low affinity, microbes to which the host does not have an adaptive immune response, and facilitating scavenging and presentation of microbial antigen by macrophages and dendritic cells to T cells. Thus IgM is important for T cell-mediated immunity, quite apart from its role in humoral immunity. A defect in both systems following B cell depletion is of potential concern. Nevertheless, in the experience at UCL, and also apparently elsewhere, there is little or no indication that B cell depletion makes patients susceptible to the classical opportunistic infections such as pneumocys-tis carinii, tuberculosis, commensal fungi, or viruses. To date, none of the patients with absent IgM levels in our cohort have suffered significant infective episodes.

Serum IgG levels tend to fall by a mean of 20-30%, but remain within the normal range, even following repeated cycles of rituximab (Cambridge et al., 2003). Interestingly, following rituximab patients with both RA and SLE may normalize their IgG levels from either very high or abnormally low levels. This can be dramatic and encourages the hope that some form of restoration of immune regulation or removal of inhibitory or cytotoxic antibodies is occurring.

The broad conclusion, therefore, is that the simplest explanation for the benefit of BLyD is the reduction in levels of pathogenic autoantibodies. Our recent immunodynamic monitoring of patients with autoimmune disease undergoing BLyD has, however, produced a number of findings that suggest that the relationship between B cells and disease is even more complex (Cambridge et al., 2003).

4.3. Clinical Response Follows Serological Response, Not B Cell Numbers

If the therapeutic action of BLyD was through the removal of B cells responsible for presenting autoantigens to T cells, benefit would be expected only for as long as B cells were depleted. B lymphocytes are often sparse in synovium and even when present form a minority of antigen-presenting cells, but this would not preclude the possibility that they were functioning in extrasynovial sites. Observations following BLyD make this explanation unlikely. B lymphocyte numbers fall within days of rituximab therapy but clinical improvement occurs over a period of several months (Cambridge et al., 2003). If B cells are functioning as the key providers of antigen presentation to T cells capable of inducing cytokine production from synovial or other macrophage populations, one would expect a rapid and dramatic response to BLyD, as observed with TNFa antagonists. This does not occur.

4.4. The Kinetics of Relapse Follow Autoantibody Rises Rather than B Cell Return

Although evidence to date suggests that B cell depletion as currently performed is unlikely to produce long-term remission in RA it is not uncommon for patients to remain in remission for 1-2 years following the return of circulating B cells (Cambridge et al., 2003). In almost exactly half of cases relapse occurs within a few weeks of B cell return, but in the other half there is a delay of several months (see Figure 21.3). B cell return is therefore necessary but not sufficient for relapse. However, return of circulating autoantibodies does appear to be both necessary and sufficient in that a return to pretreatment levels is consistently associated with relapse. This supports the view that autoantibodies are directly involved both in the propagation of the autoimmune response and in clinical disease expression, as in the Trojan horse model (see Figure 21.1).

The long delay between B cell return and clinical relapse indicates the presence of a rate-limiting step (or steps) in the returning pathological immune response. The often-protracted period of increasing autoantibody production, which is also known to precede the initial onset of clinical RA in many cases, suggests that this is a rerun of an early stage of disease propagation. These observations are consistent with a role for infrequent stochastic generation of antibody species with aberrant signaling capacity, as initially envisaged (Edwards et al.,

1999). An important question is whether the continued production of autoanti-body from plasma cells may determine relapse by providing afferent immunoreg-ulatory signals favoring reemergence of a self-amplification loop of autoreactive B cell survival (Figure 21.1B). Agents targeting different windows of activity in B cell life history may help answer this question.

4.5. Why Are There Two Patterns of Relapse?

In RA there appear to be two distinct temporal patterns of relapse (Figure 21.3) immediately following B cell return or at random over a period of many months. The first pattern might reflect the resistance of some pathogenic memory B cell populations to rituximab due perhaps to their geographic situation, e.g., in the marginal zone of the spleen or in synovial tissue. The significance of survival of B cells in the synovium may be not so much that it is the target tissue but that it is a solid tissue that may be difficult to penetrate. Relapse at B cell return might therefore reflect involvement of "original" pathogenic clones in reestablishment of a vicious cycle and reattaining of clinically relevant levels of proinflammatory autoantibodies. Initial studies of bone marrow from RA patients and of peripheral blood in RA and SLE patients supports the idea that inadequate depletion of B cells may be associated both with early B cell return and an immediate autoanti-body rise associated with relapse, indicating that doses used for B cell depletion may at times be suboptimal.

In the delayed pattern of relapse, it is possible that qualitative as well as quantitative factors are involved. IgG RhF from (CD20-) plasma cells can, in theory, recruit newly emerging IgM-RhF B cells to enter into a vicious cycle (Figure 21.1B). Such B cells would then become the focus of increased and inappropriate T cell help, resulting in germinal center formation, affinity maturation, and class switching. The pattern of RhF return in patients would support this in that IgM-RhF is usually the first class to rise before relapse.

How To Bolster Your Immune System

How To Bolster Your Immune System

All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.

Get My Free Audio Book

Post a comment