Targeting B Cells to Modify MS, NMOSD, and MOGAD

Proof of principle: rituximab

The critical role of B cells in MS²⁰ and NMOSD²¹ was recently reviewed. It was the demonstration that rituximab is highly effective in MS that prompted a reappreciation of the contributions of B cells to MS pathogenesis (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 The central role of B cells in the immunopathogenesis of MS

B and T cells in the peripheral lymphoid tissues reciprocally activate each other. They migrate to the CNS passing through the blood-brain barrier. Most B cells locate to the perivascular space. Aggregates of B lymphocytes are observed in the pia mater overlying the cortex. In secondary progressive MS, a compartmentalized inflammation in an ectopic follicle-like lymphoid tissue is driven by B cells, plasma cells, T cells, and follicular dendritic cells. In the CSF, antibody-producing memory B cells, plasmablasts, and plasma cells give rise to oligoclonal bands. From ref. 18 with permission by Springer Nature.

In the first case report of a patient with aggressive relapsing MS disease stabilized with rituximab, B cells were depleted in CSF and peripheral blood.²² B-cell counts in patients with primary progressive MS were lowered more in peripheral blood than in CSF.²³ In a phase 2 trial of patients with relapsing-remitting MS (RRMS) receiving rituximab as add-on therapy, decreases of both B- and T-lymphocyte counts were observed in CSF.²⁴ Several case reports convincingly demonstrated that rituximab not only mitigated or arrested progression of a fulminant disease course but also led to clinical improvement.^22,25,26

The beneficial effects of B-cell depletion in NMOSD were first demonstrated in an open-label study of rituximab, published in 2005,²⁷ followed by a retrospective analysis of 25 patients with NMOSD in 2008²⁸ and a prospective long-term cohort study of 10 patients.²⁹

Mode of action of CD20 cell depletion in MS—evidence emphasizing the role of B cells in MS pathogenesis

Binding of monoclonal antibodies to CD20

CD20 is a member of the membrane-spanning 4A gene family encoded by the MS4A1 gene on chromosome 11.³⁰ The CD20-targeting monoclonal antibodies recognize shared or slightly different epitopes. Ublituximab (TG-1101) binds to a unique part of the CD20 domain^31,32 (figure 2). Recent work suggests a double-barrier dimer structure.³³ In general, binding of these antibodies induces the redistribution of CD20 into lipid rafts within the plasma membrane and may thus disrupt B-cell receptor signaling. Because of amino acid mutations in the Fc portion, rituximab and ofatumumab bind C1q more avidly than ocrelizumab and lead to complement activation and complement-dependent cytotoxicity. In contrast, ocrelizumab more potently promotes antibody-dependent cellular cytotoxicity and phagocytosis via its interaction of Fcγ receptors on natural killer cells, monocytes, macrophages, and neutrophils.^34,35 Based on these differential properties, anti-CD20 antibodies are termed type I (rituximab and ofatumumab) or type II (ocrelizumab).^36,37

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 Epitopes on CD20 recognized by anti-CD20 monoclonal antibodies

CD20-targeted monoclonal antibodies recognize epitopes either identical or spatially in close neighborhood. Only ublituximab binds to a unique epitope on the CD20 domain. From ref. 31 with permission by SAGE Publishers.

CD20 cell depletion had not only a therapeutic effect on RMS or RRMS but also, for the first time, on the primary progressive course of the disease (PPMS).^38,39 In the first trial using rituximab in PPMS, the primary end point was not reached. However, when stratified to subsets, the group (1) <50 years and (2) Gd + did better statistically than placebo.³⁸

One of the most fascinating aspects of CD20 depletion in MS is that although the target cell is known, it remains unclear which intercepted B-cell function is most relevant in this context. B cells may contribute to autoimmune disease via autoantibody-dependent and autoantibody-independent functions.^40,41

Figure 3 illustrates the specific B-cell developmental stages targeted by a therapy directed against the CD20 differentiation antigen. This is an effective therapeutic approach, which clearly underpins the current clinical developments and recent approvals granted to subsequent preparations with similar activity profiles (e.g., evobrutinib⁴² NCT04032158, ofatumumab⁴³ NCT02792218 or NCT02792231, ublituximab NCT03277261 or NCT03277248, and NCT04032171). Furthermore, the development of rituximab biosimilars like obinituzumab is in progress to reduce drug resistance.⁴⁴ It is interesting to note that low-dose intrathecal administration of rituximab leads to complete depletion of CD20 in peripheral blood⁴⁵ but not to complete CD20 depletion in the CNS.⁴⁶

• Download figure
• Open in new tab
• Download powerpoint

Figure 3 Cellular targets of CD19 and CD20 cell depletion therapies

A number of differentiation antigens are expressed on the cell surface during B-cell maturation and are subsequently recognized by the respective monoclonal antibodies used during therapy. Monoclonal antibody/antigen binding ultimately leads to depletion via antibody mediated or complement-dependent cytotoxic mechanisms. It is important that both early and late maturation stages are not depleted because they do not express CD19 or CD20. This means that the ability to repopulate B cells is preserved and humoral immune memory is not impaired. This maintains the function of natural defense mechanisms. CD19 in contrast to CD20 is expressed also on pro–B cells and plasma cells. Thus, CD19-directed monoclonal antibodies have a broader coverage of the B-cell lineage. A minor subpopulation of CD3 T lymphocytes, CD8 more than CD4 T cells, also display the CD20 antigen. They have been shown to be depleted following rituximab administration.^e25 It is controversial whether they are increased in activity exhibiting higher proinflammatory potential in blood and CSF of people with MS compared with healthy individuals. CD20-depleting therapy removes myelin-reactive CD8 T lymphocytes from the circulation of patients with MS.^e33

Rituximab in MS

A phase 1 trial evaluating 26 patients with RRMS, in the absence of a control group, demonstrated a good 72-week rituximab safety profile and showed that rituximab reduced the development of new lesions between week 4 and week 72, as well as flare-ups, compared with the year preceding the treatment.^e34 The double-blind, placebo-controlled phase 2 trial (Helping to Evaluate Rituxan in Relapsing-Remitting Multiple Sclerosis [HERMES], e-table 1, links.lww.com/NXI/A346) additionally confirmed that rituximab 1,000 mg administered on days 1 and 15 was more effective than placebo, not only in terms of the primary MRI end points but also for all secondary clinical end points examined.^e35 Pharmacodynamically, rituximab was associated with rapid almost complete depletion of CD19⁺ B cells from weeks 2 to 24. By week 48, CD19 cells had returned to 31% of baseline_.^e35 B-cell depletion resulted in markedly diminished proinflammatory Th1 and Th17 responses of CD4 and CD8 T lymphocytes.⁴⁸

Given the surprisingly good efficacy of CD20 cell depletion in patients with RRMS, it was hoped that this therapeutic approach might also prove to be effective in the primary progressive course of the disease. To this end, a randomized, double-blind, placebo-controlled phase 2/3 trial was conducted in patients with PPMS (OLYMPUS, e-table 1, links.lww.com/NXI/A346). Four hundred thirty-nine patients received 2 1,000 mg rituximab or placebo infusions every 24 weeks until week 96. This trial failed to meet the clinical primary end point, as there was no significant difference in confirmed disease progression. Despite this disappointing result, expectations for rituximab's efficacy remained, as the secondary radiologic end point of T2 lesion load was met and post hoc subgroup analyses indicated that younger patients (less than 50 years of age) with active inflammatory lesions, in particular, seemed to benefit from rituximab treatment.^38,e36

Furthermore, several retrospective analyses and a subgroup analysis provided evidence that rituximab can be effective both in aggressive RMS and progressive MS^e37–e43 and that it is also more effective than first-generation MS treatments (i.e., interferon-beta and glatiramer acetate).^e44,e45 In the large Swedish study, doses of rituximab above 750 mg/6 months were not more effective than those below.^e43 A retrospective Swiss study of 37 patients with RRMS and 22 patients with SPMS observed that lowering rituximab from 1,000 mg to 500 mg/6 months was safe and produced clinical and MRI stability. It also left serum neurofilament light chain levels unaffected. The authors suggested that with a lower dose one might avoid more marked hypogammaglobulinemia and thereby the risk of infectious complications^e46

Ocrelizumab: breakthrough in MS therapy

Subsequent studies, investigating the efficacy of CD20 depletion in patients with MS, used the humanized equivalent ocrelizumab instead of the chimeric monoclonal antibody rituximab (table 3). Ocrelizumab was administered initially in 2 induction doses of 300 mg IV at a 14-day interval, followed by a dose of 600 mg IV every 6 months over the course of the trial. In the initial randomized, double-blind, placebo-controlled phase 2 trial in RMS with a double-dummy design, interferon beta-1a administered IM served as the active comparator in RMS. Ocrelizumab was shown to be more effective than placebo and the active comparator in terms of the radiologically defined primary end point (number of gadolinium-positive T1 lesions in brain MRI). No significant differences were identified between individual groups in terms of side effects.^e47

Subsequent phase 3 trials led to the approval of ocrelizumab for the treatment of RMS and PPMS.^e48 Both ocrelizumab vs IM interferon beta-1a in RMS (OPERA I and II)^e49 and ocrelizumab vs placebo in early PPMS (A Study of Ocrelizumab in Participants With Primary Progressive Multiple Sclerosis [ORATORIO])³⁹ met the defined clinical primary end points. The RMS twin studies achieved the primary end point, which was defined as the annualized relapse rate, with an absolute risk reduction of 46% and 47%, respectively. All secondary end points were also met with the exception of the percentage change in brain volume in the OPERA II trial.

Following the double-blind phase, patients from both OPERA studies were maintained on ocrelizumab or switched from interferon-ß 1a for 3 years when clinical and MR outcomes and safety were assessed. Pooled analysis of the 88.6% of patients that completed 5 years showed maintenance of clinical (annualized relapse rate, 24-week confirmed disability progression and improvement) and MRI effects (total number of new or newly enlarged T2 lesions and T1 gadolinium-enhancing lesions assessed every 48 weeks).

Patients on continuous ocrelizumab treatment from beginning fared better than those with delayed treatment onset. Whole, gray and white matter brain volume loss was less marked in the group on continuous vs delayed treatment. No new safety signals emerged.^e50

In the double-blind period, NEDA (no evidence of clinical or MRI disease activity) was attained in 48.5% of the ocrelizumab and 27.8% in the IFNß-1a–treated patients. During the open-label extension, the proportion of patients with NEDA was 65.4% in patients continuously receiving ocrelizumab vs 55.1% in the switching population.^e51

In the PPMS trial, both the primary end point (disease progression confirmed after 12 weeks) and the secondary MRI end points were met. After 120 weeks, 32.9% of patients in the ocrelizumab and 39.3% in the placebo group had progressed (relative risk reduction of 24%), whereas absolute brain lesion volume decreased by 3.4% (ocrelizumab) and 7.4% (placebo), respectively. A recent post hoc analysis confirmed an improvement in arm function using the 9 Hole Peg Test. The proportion of patients with confirmed deterioration in the 9 Hole Peg Test was also lower in the ocrelizumab treatment group.^e52 A retrospective analysis of a large multicenter MS cohort revealed that 5% of all patients experienced confirmed disease progression, whereas the annualized relapse rate was 0.17.^e53

It is noteworthy that the PPMS trial patterned on the results of the phase 2 OLYMPUS trial of rituximab only included patients who had a relatively short duration of the disease, as defined by age (18–55 years) and symptoms <15 years in patients with an expanded disability status score of >5.0 or <10 years in those with an expanded disability status score of <5.0 at the time of screening. Both acute infusion reactions and respiratory tract and oral herpes infections emerged as significant adverse events. There was also a trend toward an increased risk of cancer (particularly of breast cancer) (2.3% vs 0.8%), although extended follow-up and postmarketing experience did not confirm this association. Besides, the incidence was in the range expected in the general population.^e54,e55

Repletion of B cells appears not to be uniform. A recent study of 74 patients with MS noted 41.8% who had their CD19 cells returning at 6 months, the so-called fast responders. Twenty-four percent had CD19 cells above 2%. Fast response was associated with a higher body mass index.^e56

B-cell repopulation has been associated with a rise in the relapse rate in the case of NMOSD, a relapsing-remitting chronic inflammatory CNS disease with concomitant astrocytopathy.^e57 In this context, CD19/CD27-positive B cells seem to be especially important.^e58 It remains unclear to what extent this correlation can be extrapolated to RMS. In the pivotal clinical trials, 20.7% of patients with RMS and 26.3% of patients with PPMS experienced a decrease in absolute lymphocyte counts below the lower limit of normal.^e59 The majority of patients developed grade 1 or 2 lymphopenia, the incidence of grade 3 lymphopenia was 1%, and no CD19-positive cells could be detected in peripheral blood as early as 2 weeks after the start of the treatment.^42,e49 The lymphocyte population was restored in 90% of the patients, after a 2.5-year (median 72 weeks) ocrelizumab treatment pause.^e59 In comparison, the lymphocyte population in the rituximab phase 2/3 trial (OLYMPUS) was restored in 35% of the patients, 48 weeks after discontinuing treatment.³⁸ Results from the surveillance studies after marketing authorization was granted in January 2018^e59 (for indications, refer to e-table 2, links.lww.com/NXI/A346) confirmed the benefits of ocrelizumab. Of note, 66.4% of patients with RMS receiving ocrelizumab vs 24.3% of interferon beta-1a patients exhibited no evidence of clinical or radiologic disease activity (NEDA).^e60 In the absence of direct comparative studies evaluating ocrelizumab against other MS therapies, a meta-analysis performed showed that ocrelizumab is a viable treatment option, particularly for patients with highly active RMS.^e61 Furthermore, several retrospective analyses and a subgroup analysis provided evidence that rituximab can be effective both in aggressive RMS and progressive MS^e37–e43 and that it is also more effective than first-generation MS treatments (i.e., interferon beta and glatiramer acetate).^e44,e45 To better characterize the response of patients with PPMS to immunotherapy, the composite outcome termed NEPAD—no evidence of progression or active disease—has recently been introduced. Operationally, it is defined as absence of 12-week confirmed disability progression; absence of 12-week confirmed progression of ≥20% on the Timed 25-Foot Walk test and 9-Hole Peg Test; no brain MRI activity (no new/enlarging T2 lesions and no T1 gadolinium-enhancing lesions); and no protocol-defined relapses.^e58 In the ORATORIO trial, ocrelizumab tripled the proportion of patients with PPMS with NEPAD after 120 weeks compared with placebo.^e62 An analysis of pooled phase 2 and 3 data suggests that ocrelizumab effectively suppresses disease activity detected by MRI within 4 weeks and suppresses clinical disease activity within 8 weeks.^e63 A post hoc study suggests that ocrelizumab is also effective in patients with MS with increased baseline disability.^e64 An MR spectroscopy study taking sequential measurements of markers of neuronal-myelin coupling over a 96-week period demonstrated that ocrelizumab reduces cerebral gliosis compared with patients who received interferon-beta.^e65 It also appears that progression in patients with PPMS is associated with chronic lesion activity in the white matter.^e66 Currently, multiple phase 3b studies are being performed to better characterize ocrelizumab in MS (table 2). ORATORIO HAND (NCT04035005) explores its effect on the function of the upper extremity in PPMS, CONSONANCE (NCT03523858) monitors disease activity in PPMS, and CHORDS (NCT02637856) and CASTING (NCT02861014) scrutinize safety and efficacy in RRMS after suboptimal response to a different DMT.

ENSEMBLE (NCT03085810) examines the effect of ocrelizumab in people with early RMS. A recently published study (ENSEMBLE PLUS) showed that shorter infusion times were well tolerated and not associated with more adverse events.^e67

A phase III b open-label study, VELOCE (NCT00676715), evaluated the effectiveness of common vaccinations in patients with relapsing MS treated with ocrelizumab. Responses to clinically relevant antigens (tetanus toxoid, pneumococcal antigens, and influenza or keyhole limpet hemocyanin as a neoantigen) were elicited but attenuated compared with controls on interferon-ß or no disease-modifying drug.^e68 OBOE (NCT02688985) investigates immunologic changes occurring during ocrelizumab treatment.

In the absence of a phase 3 trial, the issue of whether rituximab is an equivalent alternative to ocrelizumab remains open to debate.^e69–e72 Direct comparative studies between rituximab and other treatment options would be helpful to close the gaps in our current understanding.^e73

Ofatumumab

Ofatumumab is a human monoclonal antibody that binds to a small-loop epitope on CD20. It depletes B cells through complement-dependent cytotoxicity and antibody-dependent cell-mediated cytotoxicity with great efficiency even when CD20 expression is low.^32,43

A small phase 2 study examining ascending doses of 100, 300, and 700 mg IV ofatumumab showed almost complete reduction in new MRI lesion activity.^e74 These results prompted a larger-scale phase 2 dose-response study of subcutaneous ofatumumab in patients with relapsing-remitting MS.⁴² Two hundred thirty-two patients were randomized to receive 3, 30, or 60 mg every 12 weeks, ofatumumab 60 mg every 4 weeks, or placebo for 24 weeks. The primary end point was the number of cumulative new gadolinium-enhancing lesions on cerebral MRI. The cumulative number of gadolinium-enhancing lesions was reduced by 2 thirds in patients allocated to ofatumumab with a dose-dependent depletion of circulating CD19 B cells. Complete depletion apparently was not necessary for a treatment effect. Onset of action was observed by week 12. Furthermore, a conditioning dose added no benefit, and repletion occurred faster in all ofatumumab doses than previously reported with anti-CD20 therapy. Adverse events were mostly injection related and mild.⁴³ An overview of further trials of subcutaneous ofatumumab in MS is provided in table 3.

Two pivotal ofatumumab phase 3 clinical trials enrolling participants with typical relapsing (94%) and secondary progressive (5–6%) MS were conducted^e71,e72. The primary end point was annualized relapse rate, and pooled analysis was performed in a preplanned meta-analysis of disability worsening. Ofatumumab produced a relative risk reduction in the annualized relapse rate of between 50.5% (ASCLEPIOS I) and 58.5% (ASCLEPIOS II) compared with teriflunomide^e75,e76 (details in table 3). In the prespecified combined analysis of both trials, ofatumumab was superior to teriflunomide in cutting the risk of 3- and 6-month confirmed disability worsening by 34.4.% and 32.5%, respectively. These clinical results were corroborated by MRI evaluation of metrics of disease activity and burden with one exception. No difference was noted for brain volume loss. Serum neurofilament light chain levels were lowered more markedly in the ofatumumab than the terilunomide group. Injection site reactions were common, but overall, ofatumumab was well tolerated.^e76

Ofatumumab in August 2020 received approval by the FDA for relapsing forms of MS (CIS, RRMS, and active SPMS) (e-table 2, links.lww.com/NXI/A346). The efficacy of ofatumumab offers patients an alternative administration route. They could self-inject.

This could obviate the need to attend a medical enter and avoid exposure to infective agents. This may arguably reduce utilization of healthcare resources. It remains to be seen whether the less profound depletion and faster repletion of B cells achieved with ofatumumab will also translate into a more favorable safety profile.

Ongoing: ublituximab

Ublituximab (TG-1101) is a novel glycoengineered anti-CD20 chimeric IgG1 monoclonal antibody. Potential advantages over currently available CD20 directed antibodies encompass induction of a higher degree of antibody-dependent cellular cytotoxicity, activity in low CD20 epitope expressing cells as is characteristic in rituximab resistance, binding to a novel epitope of CD20, and the shorter infusion time of 1 hour. Results of a phase 2 placebo-controlled trial, highlighting both efficacy and safety data, have recently been published.³¹ CD19-positive cell depletion, the primary outcome, was achieved in >95% and NEDA in 74% of patients receiving ublituximab. Infusion-related reactions were the most common adverse events but mild in nature. Two phase 3 trials of ublituximab vs teriflunomide in patients with relapsing MS are currently ongoing.

Atacicept, an inhibitor of B-cell differentiation that failed in MS

Atacicept is a fully humanized recombinant fusion protein that interferes with B-cell differentiation, maturation, survival, and antibody production by binding to the cytokines BLyS (B-lymphocyte stimulator) and APRIL (also known as TNFSF13).^e77,e78 Animal studies suggested that its mode of action could produce therapeutic benefit in MS.^e79 Unfortunately, a phase 2 trial revealed that this promising^e80 mode of action leads rather to an exacerbation of disease, and the study had to be prematurely terminated.^e81 This finding underscores the complex involvement of B cells in MS.^e82 The failure of atacicept in MS may be associated with its shifting the balance of regulatory B cells^e83 and memory B cells.^e84–e86 As a result, pathogenic memory B cells were stimulated. In addition, atacicept failed in optic neuritis, rheumatoid arthritis, and systemic lupus erythematosus.^e87 Basic research on the B-cell survival factor suggests that the mode of modulation is crucial to achieve clinical efficacy.^e88

Perspectives: Bruton tyrosine kinase inhibitors

Kinase inhibitors have recently appeared on the horizon as potential immunotherapeutics for MS.^e89 The implications of kinase inhibition and transition from bench to bedside in oncological diseases have been reviewed.^e90–e92 The cytosolic Bruton tyrosine kinase is expressed exclusively on cells of the hematopoetic lineage and therefore affects the safety profile of this class of kinase inhibitors.

Evobrutinib, a selective, covalent, oral Bruton tyrosine kinase (BTK) inhibitor, blocks B-cell activation and cytokine release.^e93 The development of BTK inhibitors has generated promising agents.^e94 Ibrutinib was the first in class to be assessed in B-cell malignancy clinical trials.^e95 Preclinical characterization and phase 1 trials revealed that evobrutinib and branebrutinib are both well tolerated and potent inhibitors with high kinase selectivity.^e96–e99 The first kinase inhibitors are licensed for hematooncologic indications, including mantle cell lymphoma, chronic lymphocytic leukemia, and graft-versus-host disease.^e100 The BTK inhibitor acalabrutinib has been granted breakthrough designation by the US Food and Drug Administration for the treatment of chronic lymphocytic leukemia.^e101 Thus, in malignancy, BTK inhibition is a promising therapeutic approach in B-cell diseases.^e102,e103 In MS, a phase 2 study of evobrutinib met its primary end point by significantly reducing the risk of developing gadolinium-enhancing lesions.⁴² It was well tolerated. A reversible increase in liver function tests was noted. Currently, studies with the BTK inhibitors BTKi (`168) (SAR442168) and GDC-0853 (fenebrutinib) have been completed, are ongoing, or about to get started.^e104,e105 E-table 3, links.lww.com/NXI/A346, summarizes completed and planned phase 2 and 3 trials in MS.

Preclinical data from experimental autoimmune encephalomyelitis, EAE, suggest that BTK inhibitors may unfold their beneficial effect in MS via multiple mechanisms, i.e., modulation of the cytoplasmic tyrosine kinase and Toll-like receptor signaling.^e106 This may impair generation of (auto)antibodies and B-cell antigen-presenting function. Actions on myeloid cells including microglia may be particularly important given their presumed role in driving and maintaining an immunoinflammatory response with consequent neurodegeneration during the progressive stages of MS. As small molecules, BTK inhibitors may access the CNS easily and arrive in zones of subpial compartmentalized inflammation that are considered to significantly contribute to the pathobiology of progression. Given the medium effect size of, e.g., evobrutinib shown in phase 2 and the apparently benign safety profile, combination of oral BTK inhibitors with other high-efficacy drugs is being considered.