Rituximab combination therapy in relapsing multiple sclerosis

Abstract

In multiple sclerosis (MS), the presence of B cells, plasma cells and excess immunoglobulins in central nervous system lesions and in the cerebrospinal fluid implicate the humoral immune system in disease pathogenesis. However, until the advent of specific B-cell-depleting therapies, the critical role of B cells and their products in MS was unproven. Rituximab, a monoclonal antibody that depletes B cells by targeting the CD20 molecule, has been shown to effectively reduce disease activity in patients with relapsing MS as a single agent. Our investigator-initiated phase II study is the only published clinical trial in which rituximab was used as an add-on therapy in patients with relapsing MS who had an inadequate response to standard injectable disease-modifying therapies (DMTs). The primary endpoint, magnetic resonance imaging (MRI) gadolinium-enhanced (GdE) lesion number before versus after rituximab, showed significant benefit of rituximab (74% of post-treatment MRI scans being free of GdE lesions compared with 26% free of GdE lesions at baseline; p < 0.0001). No differences were noted comparing patients on different DMTs. Several secondary clinical endpoints, safety and laboratory measurements (including B- and T-cell numbers in the blood and cerebrospinal fluid (CSF), serum and CSF chemokine levels, antibodies to myelin proteins) were assessed. Surprisingly, the decline in B-cell number was accompanied by a significant reduction in the number of T cells in both the peripheral blood and CSF. Rituximab therapy was associated with a significant decline of two lymphoid chemokines, CXCL13 and CCL19. No significant changes were observed in serum antibody levels against myelin proteins [myelin basic protein (MBP) and myelin/oligodendrocyte glycoprotein (MOG)] after treatment. These results suggest that B cells play a role in MS independent from antibody production and possibly related to their role in antigen presentation to T cells or to their chemokine/cytokine production.

Introduction

Multiple sclerosis (MS) is a presumed autoimmune disease that affects about 2.5 million people worldwide. Though typically manifesting as a relapsing-remitting disease initially, it often results in progressive functional deficits over time with or without discrete episodes of exacerbations. Based on clinical course, three main clinical subtypes of MS are defined: relapsing-remitting (RR), secondary progressive (SP) and primary progressive (PP). Several key findings implicate a pathogenic role of B cells and antibodies in MS. Intrathecal oligoclonal bands (OCBs) containing immunoglobulins (Igs) and seen following cerebrospinal fluid (CSF) electrophoresis are present in more than 90% of all patients with MS [Trotter and Rust, 1989]. Notably, higher concentrations of Igs in CSF correlate with more aggressive types of MS, and with episodes of MS worsening [Avasarala et al. 2001; Olsson and Link, 1973; Zeman et al. 1996]. Antibodies that may mediate demyelination have been identified in MS lesions by histopathology in a substantial proportion of patients with MS [Lucchinetti et al. 1996]. Clonally expanded B cells accumulate in chronic MS lesions and in the CSF of patients with MS [Baranzini et al. 1999; Colombo et al. 2000; Qin et al. 1998]. Ectopic lymphoid tissue containing B cells, a finding first reported in 1978 [Prineas and Wright, 1978], has been observed in the central nervous system (CNS) of patients with progressive MS [Magliozzi et al. 2007]. Because of all these observations, much prior research on MS in the latter part of the past century focused on the roles of B cells, plasma cells and the humoral immune system in MS pathogenesis. However, until the advent of specific B-cell depleting therapies, the critical role of B cells and their products in MS was unproven.

Rituximab

A little more than a decade ago the mouse-human chimeric monoclonal antibody to CD20, rituximab, was approved by the US Food and Drug Administration (FDA) for the treatment of some forms of lymphoma. Rituximab targets cells that express surface CD20, exclusively found on pre-B and mature B cells. Rituximab, a complement binding IgG1 κ monoclonal antibody, specifically lyses circulating B cells while sparing stem cells and mature plasma cells. Initially approved for therapy of non-Hodgkin’s lymphoma in the USA in 1997, its FDA approval has expanded to encompass chronic lymphocytic leukemia and, in 2006, rheumatoid arthritis. Most recently, in 2011 it was approved for Wegener’s granulomatosis and microscopic polyangiitis, two rare types of vasculitis [Langford, 2012]. The advent and availability of rituximab allowed for the first time the ability to determine the effects of B-cell depletion on the course of MS. The results have been beneficial in relapsing MS.

Two published studies have supported a therapeutic role of Rituximab in patients with relapsing MS as a single agent [Bar-Or et al. 2008; Hauser et al. 2008]. A phase II, double-blind clinical trial in 104 patients with RRMS randomized to receive a single course of rituximab or placebo showed that patients who received rituximab had reduced inflammatory brain lesions as evaluated by gadolinium-enhanced magnetic resonance imaging (MRI), and a lower number of clinical relapses that was sustained for 48 weeks after treatment [Hauser et al. 2008]. Another smaller open-label phase I study was performed in patients with RRMS. This was a 72-week clinical trial to evaluate as primary outcome the safety of two courses of rituximab (administered at week 0 and week 24). No serious adverse events were observed. The majority of adverse events were infusion-associated events known to be due to cytokine release during B-cell lysis. All infection-associated adverse events were mild to moderate. In this cohort of RRMS, B-cell depletion was accompanied by a sustained reduction of relapses and magnetic resonance activity throughout the 72-week study duration [Bar-Or et al. 2008].

Rituximab has also been tried as a single agent in PPMS [Hawker et al. 2009]. In a 96-week multicenter trial, 439 patients with PPMS received two 1000 mg infusions of rituximab or placebo every 24 weeks for four courses. The primary endpoint was time to confirmed disability progression based on sustained 12-week increase in the Expanded Disability Status Scale (EDSS) score. This study, termed OLYMPUS, did not achieve its primary endpoint, although at least one secondary MRI endpoint revealed a beneficial effect. That is, a diminished increase in T2 lesion volume was observed with rituximab therapy. Of note, pre-planned subgroup analyses showed that time to confirmed disability was significantly delayed by rituximab in patients under 51 years old and in those with gadolinium-enhanced (GdE) lesions, with the greatest effect in those who were aged under 51 years and had GdE lesions. Those with shorter disease duration also had a better response. This study had two important conclusions: first the study indicated a critical role for B cells in the disability progression of at least a subset of patients with PPMS; second that PPMS may be more heterogeneous and with greater inflammation (as evinced by GdE lesions) than previously realized.

However, trials of monoclonal antibodies targeting B-cell activating factor (BAFF) and a BAFF receptor (transmembrane activator and calcium modulator and cyclophylin ligand interactor) have been uniformly negative in patients with MS. Why that might be is not understood [Hartung and Kieseier, 2010].

Phase II trial of rituximab in combination with β interferon or glatiramer acetate

The bulk of this review will focus on our own studies using rituximab to deplete patients of circulating B cells in combination with other immunomodulatory medications in patients with relapsing MS. Thus far, ours has been the only published clinical study in which rituximab was used as an add-on therapy in relapsing MS [Naismith et al. 2010]. Our investigator-initiated phase II study was funded by the National MS Society, USA and the National Institutes of Health and began to enroll patients in spring 2002, prior to any published reports of use of rituximab in MS. This trial enrolled subjects with inadequate response to standard injectable disease-modifying therapies [DMTs: β interferons (β-IFNs) or glatiramer acetate], and had a preplanned primary endpoint to determine the effect of rituximab on GdE lesion numbers. Several secondary endpoints were also assessed, including effects of rituximab on the Multiple Sclerosis Functional Composite (MSFC), EDSS, and safety. Patients enrolled in the study were doing suboptimally on the DMTs, but still they may have derived some benefits from the treatment. When the study began in 2002, rituximab was not yet demonstrated to be an effective therapy for MS and there were some theoretical reasons that it might make patients worse. Therefore our study design included the addition of rituximab to patients’ existing therapy, as we did not want to take those enrolled off their DMTs for many months.

Enrollment criteria were relapsing MS with a significant clinical relapse within the prior 18 months, despite using an injectable DMT at the time. The prospective subject was required to have at least one GdE lesion on any of three monthly pretreatment brain MRIs. Sixty-nine patients met the clinical inclusion criteria, of which 50% (35/69) had at least one GdE lesion on brain MRI, qualifying them for enrollment. It took eight years to enroll the 30 planned subjects due to the fact that the study was conducted at a single site, and had rather stringent inclusion and exclusion criteria. Also, three of the 35 qualifying patients withdrew consent prior to receiving any rituximab, and two were removed from the study due to side effects during the initial dosing. The 30 remaining subjects received rituximab at the oncologic dosing regimen of 375 mg/m² weekly for 4 weeks. All 30 subjects completed the three post-rituximab brain MRIs and 52 weeks of clinical follow up as planned. Because this group of patients had failed standard agents, they tended to be older and with longer duration of disease than most studies in RRMS. In this trial, the median age was 43.5 years with disease duration of 7.5 years (Table 1). Eight subjects were men and 22 were women. Most of these patients were taking β-IFN therapy (n = 24) and six were taking glatiramer acetate. Each patient remained on the same DMT at the same dose for the entire study. Enrolled patients in this add-on trial were more severely affected by MS than subjects in most RRMS trials, with median EDSS of 4.0 (mean 4.7, range 2–6.5), mean 25 ft timed walk of 8.55 s (range 4.4–56.7 s) and mean MS severity score within the worst 30th percentile [Naismith et al. 2010; Roxburgh et al. 2005].

Table 1.

Characteristics of subjects enrolled in the combination trial.

Age (range)	43.5 years (20–53)
Women:men	22:8
Disease duration (range)	7.5 years (2–32)
MRI GdE number, median (mean ± SD)	1.0 (2.81 ± 0.41)
Disease-modifying treatment duration (range)	2.5 years (0.5–13)
MS treatments, n
Interferon β1a intramuscular	7
Interferon β1a subcutaneous	4
Interferon β1b subcutaneous	13
Glatiramer acetate	6

GdE, gadolinium-enhanced lesion; MRI, magnetic resonance imaging; MS, multiple sclerosis.

Results of primary endpoint: gadolinium-enhanced lesion number on brain magnetic resonance imaging

Three monthly post-treatment brain MRIs were obtained, beginning at week 12 after the initial rituximab infusion. Pretreatment and post-treatment MRIs were carefully examined in analyses that were blinded as to timing of the scans with regard to treatment. GdE lesions were reduced in number after treatment with rituximab, with 74% of post-treatment MRI scans being free of GdE lesions compared with 26% free of GdE lesions at baseline (p < 0.0001; Figure 1). Median and mean GdE lesion numbers were reduced from 1.0 to 0 and from 2.81 to 0.33 per month respectively. Prior to beginning the study, we had defined MRI ‘response’ as a 50% or greater reduction in GdE lesions. By this criterion, 25 of the 30 subjects were responders [Naismith et al. 2010].

Figure 1.

Magnetic resonance imaging (MRI) brain scans pre and post rituximab categorized by number of gadolinium-enhanced (GdE) lesions per scan. Each subject had three pretreatment (total 90) and three post-treatment (total 90) MRI scans. The figure shows the number of scans pretreatment (pink) and the number of scans post treatment (blue) with zero, one, two or more than two GdE lesions. Sixty-seven scans post treatment had zero enhanced lesions, and no scans post treatment had more than two GdE lesions.

Results of secondary endpoints: relapse rate and disability

Although the study had not been powered to examine relapse rate changes with B-cell depletion, this endpoint was explored post hoc. All 30 subjects had been patients at the John Trotter MS Center prior to enrolling in the trial, and thus background data were accessible in their clinical charts. Fifty-seven relapses had occurred during the 18 months prior to the study in the 30 subjects who completed the study. During the 52 weeks of observation in this study, only seven clinical relapses occurred in seven subjects. Thus, the annualized relapse rate declined from 1.27 before treatment to 0.23 after treatment.

EDSS remained stable in 21/30 subjects after treatment. Improvement following treatment was assessed. Sustained improvement was defined as a decrease in EDSS by 1.0 point from a starting EDSS of up to 5.5, or a 0.5 point change from EDSS 6.0 or higher, and had to be present at week 32 and sustained at week 52 compared with baseline. By these criteria, the EDSS improved in seven and worsened in two subjects after rituximab. Thus, during the study year, 28 of 30 subjects were stable or improved, and two worsened.

As a secondary study endpoint, the MSFC was performed three times pretreatment and three times after treatment (weeks 24, 28, 32). Means at each time point were calculated. Three MSFC practice runs were performed at least 1 week apart prior to baseline assessments that were to be used in analyses of endpoints. Notably, improvement in MSFC post treatment was observed. The MSFC improved by 0.093 z score (p < 0.02; 95% confidence interval 0.018–0.17). Improvement was largely driven by improved 3 s Paced Auditory Serial Addition Test (PASAT) scores. The median PASAT score improved from baseline (p = 0.05 between weeks 0 and 52). However, the 25 ft timed walk and 9-hole peg test did not improve or otherwise change significantly following treatment.

No effect of B-cell depletion on neutralizing antibodies to β interferons

Neutralizing antibodies to β-IFNs (NAbs) were detected in six subjects prior to rituximab administration. Of these six, two reverted to normal, two declined but not to zero, and two remained unchanged after rituximab. Interestingly, four subjects developed NAbs after rituximab treatment. Antibodies to the study drug, which is a mouse-human chimeric antibody, were tested prior to and at weeks 20 and 28 after treatment. Four subjects (13%) tested positive for anti-rituximab antibodies after the study drug was administered. All four demonstrated complete B-cell depletion after rituximab treatment; none were MRI nonresponders based on our predefined criteria of 50% reduction in GdE lesion numbers [Naismith et al. 2010].

Adverse events

Several adverse events were encountered. In two cases, each during the first rituximab dosing, the side effects were significant enough to remove the patient from the study. In one case, the patient had shortness of breath that abated when the infusion was stopped. The second developed uncomfortable muscle spasms which were treated with oral prednisone, and was removed from the study because corticosteroid treatment could confound the primary endpoint of number of GdE lesions. Eleven additional subjects of the 30 who completed the study had minor and temporary infusion reactions that consisted of any combination of fever, chills, flushing, itching and diarrhea. Additional adverse events from which patients recovered included four uncomplicated urinary tract infections, a self-limited upper respiratory tract infection and bronchitis.

Blood and spinal fluid laboratory studies

Additional aims of the study were to determine the effects of B-cell depletion on the presence of antibodies to myelin proteins, and on CSF B- and T-cell numbers, IgG concentration, IgG index, IgG synthesis rate and OCB numbers. For this purpose, most of the patients enrolled in the study (26 out of 30) underwent CSF and blood sampling 1 week before and 24 or more weeks after the initial dose of rituximab. Overall, IgG levels in the CSF, including IgG index, IgG concentration, IgG synthesis rate and OCBs, did not change significantly after treatment [Cross et al. 2006; Piccio et al. 2010]. Cells in the CSF and blood were analyzed and characterized by flow cytometry (within 5 h from collection) to identify T cells (identified by the CD3 surface marker), B cells (defined as surface CD19+) and plasma cells (surface CD138+). The different cell subsets were also examined for the concurrent expression of costimulatory molecules (CD80, CD86) or activation markers. As we had expected, we observed a significant reduction in B-cell numbers in the blood and in the CSF. CD19+ B cells in blood and spinal fluid were virtually absent after treatment. Surprisingly, the decline in B-cell number was accompanied by a significant reduction in the number of CD3+ T cells in both the peripheral blood and CSF. Overall, CSF T cells declined by more than 50% compared with pretreatment measurements. T cells in the blood declined by a mean of 12% after rituximab. With regard to costimulatory and other activation-related markers, CD25, CD38 and CXCR5 were all nonsignificantly decreased on CSF CD3+ T cells after treatment. The proportions of CD19+ B cells in CSF coexpressing CD80, CD86, CD138 (plasmablasts), CD25 and CD27 were all increased after rituximab treatment. However, these studies of CSF B cells after rituximab were hindered by the exceedingly low numbers of CD19+ cells that could be found post treatment. Even with such low numbers of cells for study, the increased proportion of CD19+ B cells coexpressing CD80 and CD86 after treatment were statistically significant (p = 0.002 and p = 0.005 respectively).

The decline in CSF T cells was unexpected, as rituximab targets CD20, which is restricted to B cells. This finding led us to further investigate the mechanisms leading to T-cell reduction. One possible reason for the reduced T-cell numbers was hypothesized to be through a reduced production of chemoattractant factors due directly or indirectly to the reduction in B cells. After a literature search, we identified 17 candidate chemokines and chemoattractant factors. We measured these individually in freshly thawed aliquots of CSF by enzyme-linked immunosorbent assay, but only nine were detectable at sufficient levels for accurate measurements. Two of the nine detectable chemokines, CXCL13 and CCL19, declined significantly in the CSF and in the serum following rituximab therapy. A mild correlation between the percentage decrease in CSF T-cell numbers and in CSF CXCL13 levels was observed, suggesting that the decrease in CXCL13 might be related to the decline in T-cell numbers. No differences were noted in B cells, T cells, CXCL13 and CCL19 changes before or after rituximab treatment between subjects taking β-IFN versus glatiramer acetate [Piccio et al. 2010].

Whole Ig and IgM levels to human recombinant MOG and human MBP were measured in serum samples from a subset of patients (16 for MOG and 15 for MBP) by ELISA at baseline, and again at 24 weeks post treatment. Overall no significant differences were observed in whole Ig and IgM levels against these myelin proteins after treatment [Cross et al. 2006].

Predictors of response to rituximab

Some patients receiving add-on rituximab responded better than others, both clinically and by imaging criteria. A post hoc investigation was therefore performed to identify possible baseline predictors of response to B-cell depletion with rituximab. For this post hoc study, ‘Ideal responders’ were defined as having zero GdE lesions and improved or stable clinically during the time they were depleted of circulating B cells. ‘Nonresponders’ were defined as those who had a relapse or clinical deterioration, or no reduction in number of GdEs after B-cell depletion. Among the 30 treated subjects there were 10 ‘ideal responders’ and 9 ‘nonresponders’. Eleven subjects were intermediate, not meeting criteria for either group. The categorization of individual subjects did not change after this point, as we subsequently compared the groups for response predictors.

Demographic data (age, duration of disease, gender), and blood and CSF laboratory results were then compiled and compared for the ideal responders versus nonresponders. These results have not been published previously. Phenotyping of blood and CSF cells, CSF and blood levels of chemokines and cytokines related to B-cell function, antibodies to recombinant human MOG in blood and CSF, serum IgG, IgM and IgA levels, CSF IgG index, IgG concentration, OCB numbers and IgG synthesis rate were all compared between the two response groups (Table 2). All 10 ideal responders were women, whereas the nonresponders comprised 6 women and 3 men. The mean ages of the responders versus nonresponders were 40.0 ± 9.9 versus 42.6 ± 7.3, not significantly different. Baseline serum IgG, IgM and IgA levels were also not different and thus not predictive of response to rituximab. Similarly, baseline CSF findings did not distinguish between the two response groups. The results showed no differences in presence of CSF-specific OCBs (86% versus 88% for responders versus nonresponders respectively), IgG concentration (5.5 ± 2.5 mg/dl versus 5.3 ± 3.9 mg/dl), IgG synthesis rate (11.4 ± 13.1 versus 9.7 ± 11.1), or CSF antibodies to recombinant human MOG (0.134 versus 0.221, average ELISA optical densities). CSF IgG index trended toward being higher in ideal responders (1.2 ± 0.3 versus 0.90 ± 0.4, p = 0.11). At baseline, the proportions of CSF B and T cells in CSF (CD19 B cells 4.1% versus 3.3% in ideal responders versus nonresponders) within the mononuclear cell forward versus side scatter gate were no different between groups. No significant differences were noted for baseline CXCL12, CXCL13, CCL19, IP10, CXCL16, tumor necrosis factor β (lymphotoxin), interleukin-16 (IL-16) or C3a in CSF. Perhaps of note, C3a decreased by more than 20% in CSF from pretreatment to post treatment in five of seven subjects in the ideal responder group, and with no increases in C3a in that group, whereas C3a in CSF decreased in none and increased in two in the nonresponder group. Also, IL-16 decreased by more than 20% in CSF following treatment in three of seven ideal responders, with no increases in IL-16 post treatment in that group, whereas in the nonresponder group no subjects showed a decrease in CSF IL-16 and one had an increase by more than 20%.

Table 2.

Comparison of baseline characteristics in responders versus nonresponders.

	Responders	Nonresponders	p value
Women:Men	10:0	6:3
Age	40.0	42.6	NS
IgG, serum	1024 ± 292	1121 ± 264	NS
IgM, serum	114 ± 36	123 ± 64	NS
IgA, serum	222 ± 76	213 ± 75	NS
%CD19, blood	11.7	16.2	NS
%CD19, CSF	4.14	3.34	NS
% CD3, blood	54.46	54.07	NS
%CD3 CD25, blood	41.2 ± 13.2	40.4 ± 12.6	NS
CXCL13, serum	65.8 ± 25	63.3 ± 16	NS
CXCL13, CSF	27.5 ± 38.4	13.2 ± 13.6	NS
IgG index, CSF	1.20 ± 0.29	0.90 ± 0.41	0.11, NS
IgG concentration, CSF	5.5 ± 2.5	5.3 ± 3.9	NS
CSF albumin index	5.05 ± 2.1	5.84 ± 1.7	NS
CSF IgG synthesis rate	11.43+13.1	9.66 ± 11.1	NS
CSF rMOG (ELISA OD)	0.1337 ± 0.19	0.2212 ± 0.18	NS
CSF CXCL12 CSF (pg/ml)	1065 ± 295	1190 ± 286	NS
CSF C3a CSF (ng/ml)	17.3 ± 5.0 (n = 7)	19.1 ± 8.9 (n=6)	NS
CSF CCL19	379 ± 217	338 ± 74	NS
CSF IP-10	324 ± 137	384 ± 230	NS
CSF CCL2 (MCP-1)	385.6 ± 152.7	517.8 ± 160.1	0.16, NS
CSF CXCL16	4.0 ± 0.9	4.4 ± 0.7	NS
CSF IL-16	9.7 ± 3.8	12.9 ± 3.3

CSF, cerebrospinal fluid; ELISA, enzyme-linked immunosorbent assay; Ig, immunoglobulin; IL, interleukin; IP-10, interferon γ induced protein 10; MCP-1, monocyte chemotactic protein 1; NS, nonsignificant; OD, optical density; rMOG, recombinant myelin oligodendrocyte glycoprotein.

Discussion

Two major findings have come from these studies in which B cells were depleted in patients with MS. The primary and groundbreaking new finding is that B cells are critically important in the pathophysiology of relapsing MS. Upon introduction of rituximab, the question of whether B cells are important could finally be tested directly in humans. In particular, it has been learned that B cells are important in the development of active GdE lesions, because in the placebo-controlled single-agent study [Hauser et al. 2008], and in our combination study [Naismith et al. 2010], the number of enhanced MS lesions on brain MRIs plummeted following peripheral B-cell depletion. A second important finding was that the improvements in MRI activity and relapse rate could not be attributed to a decline in Ig levels in either peripheral blood or CSF. That is, there was no significant decline in IgM, IgG or IgA, nor was there any decline in antibodies specific to human MOG, an antigenic component of CNS myelin, that might account for the dramatic beneficial effects. This implies that other functions of B cells, such as antigen presentation or cytokine production, must be involved in the dramatic responses that have been noted.

Our study is the first reported of rituximab used in combination with standard injectable DMTs. Subjects enrolled in our study were either on β-IFNs or glatiramer acetate throughout the study. A rationale for using different therapies in combination is to achieve additive or synergistic benefits by targeting different aspect of disease pathophysiology. We found no differences in clinical, MRI and laboratory outcomes related to the different DMTs the rituximab-treated patients were receiving.

Our studies did uncover significant reductions in two chemokines, CXCL13 and CCL19, in CSF and blood, following B-cell depletion with rituximab. Lymphoid tissue chemokines, especially CXCL13 and CCL19, have been identified in meninges and CNS parenchyma of MS [Krumbholz et al. 2006]. Both CXCL13 and CCL19 have been detected within active MS lesions and proffered as prognostic indicators of relapsing disease [Khademi et al. 2011; Krumbholz et al. 2007; Serafini et al. 2004]. Other data, however, suggest that these chemokines may be general determinants for the recruitment of CXCR5- and CCR7-expressing cells, including B or T lymphocytes and leukemic cells [Buonamici et al. 2009; Kowarik et al. 2012]. The constitutive expression of CCL19 by brain endothelium in animal models of MS supports its proposed role in CNS immune surveillance and in the retention of CCR7+ T cells at perivascular surfaces [Alt et al. 2002; Engelhardt, 2006; Lalor and Segal, 2010]. In our small combination study, we were unable to identify baseline characteristics that were clearly associated with a good response to rituximab, although a trend for a higher baseline IgG index in the good compared with poor responders was seen.

Future ongoing studies to deplete B cells in patients with MS will likely use monoclonal antibodies different from rituximab, particularly since rituximab is a chimeric molecule which can lead to anti-rituximab antibodies that can neutralize its function. Ocrelizumab is a fully humanized anti-CD20 monoclonal antibody that is based on rituximab, and has been used in several preliminary clinical studies. Recently, a phase II randomized, double-blinded 48-week treatment study of ocrelizumab in 220 patients with RRMS was reported [Kappos et al. 2011]. The results showed very significant reductions in MRI activity (89–96% depending upon dose) and annualized relapse rate (70–80%, depending upon dose), similar to those seen with rituximab. Being fully humanized rather than chimeric, ocrelizumab is expected to result in less neutralizing antibody formation, and may have a more favorable efficacy and side-effect profile than rituximab. Ofatumumab is another humanized monoclonal antibody to CD20, which is approved for refractory chronic lymphocytic leukemia in the USA and Europe [Cheson, 2010]. Ofatumumab targets a different epitope on CD20 than rituximab and ocrelizumab, and is also being studied in MS. As current and future trials using these and other anti-B cell therapies in patients with MS proceed, it will be important to examine factors predictive of good response. Identifying any such factors will not only be of great practical importance for clinicians, but may provide further insights into the underlying pathophysiology of MS.