B-cell Depletion Therapy in Pediatric Neuroinflammatory Disease

Abstract

Purpose of review

B-cell depletion therapy, including anti-CD20 and anti-CD19 therapies, is increasingly used for a variety of autoimmune and conditions, including those affecting the central nervous system. However, B-cell depletion therapy use can be complicated by adverse effects associated with administration and immunosuppression. This review aims to summarize the application of anti-CD20 and anti-CD19 therapies for the pediatric neurologist and neuroimmunologist.

Recent findings

Most existing literature come from clinical trials with adult patients, although more recent studies are now capturing the effects of these therapies in children. The most common side effects include infusion related reactions and increased infection risk from immunosuppression. Several strategies can mitigate infusion related reactions. Increased infections due to persistent hypogammaglobulinemia can benefit from replacement immunoglobulin.

Summary

B-cell depletion therapies can be safe and effective in pediatric patients. Anticipation and mitigation of common adverse effects through primary prevention strategies, close monitoring, and appropriate symptomatic management can improve safety and tolerability.

Introduction

B-cell depletion is a common target for various autoimmune disorders, including those affecting the central nervous system (CNS). B-cell depleting therapies are used to treat autoimmune demyelinating disease such as multiple sclerosis and other neuroinflammatory conditions such as autoimmune encephalitis. Amongst available B-cell depletion therapies for treating autoimmune disorders affecting the CNS, anti-CD20 therapies include rituximab, ocrelizumab, ofatumumab, and ublituximab. Inebilizumab is an anti-CD19 therapy approved for use in certain primary CNS inflammatory conditions [1].

While these B-cell depletion therapies have dramatically changed the landscape of therapeutic options for CNS autoimmune conditions in the past 20 years, neurologists less familiar with these therapies may be hesitant to prescribe them due to concern for significant adverse effects related to both immediate infusion reactions and more long-term immunosuppression. In this review, we will summarize anti-CD20 and anti-CD19 therapy mechanisms and indications in CNS autoimmune disorders with a focus on pediatric patients and discuss management strategies for common and serious adverse effects through two illustrative cases.

Mechanisms of Anti-CD20 and Anti-CD19 Therapies

Anti-CD20 therapies cause rapid and sustained depletion of circulating CD20 lymphocytes in the blood through a combination of antibody-dependent cellular cytotoxicity and complement-dependent cytotoxic activation [2]. The CD20 lymphocytes being affected are predominantly naïve and memory B-lymphocytes while mature plasma cells are spared [3]. Rituximab administration causes near complete CD20 B-lymphocyte depletion (99.8%) as early as 2 weeks following administration [4]. Duration of CD20 lymphocyte depletion by rituximab can be variable based on the individual and dosing regimen, but is generally expected to last several months (typically 6–12 months) [4-6]. Children may experience earlier B-lymphocyte repopulation than adults, with earlier lymphocyte repopulation occurring in those receiving lower doses of the medication [7].

Other anti-CD20 therapies including ocrelizumab, ofatumumab, and ublituximab work similarly to rituximab by targeting and depleting CD20 lymphocytes. One of the primary differences between these therapies is their intended immunogenicity or tendency to provoke an undesired immune response. Rituximab is a chimeric human-murine antibody that is approximately 65% humanized. Ocrelizumab is designed to be less immunogenic by having a humanized variable region that produces an antibody that is more than 90% humanized. Ofatumumab is a fully humanized antibody that contains no murine components and thereby theoretically the least immunogenic [8]. Ublituximab is a chimeric human-murine antibody similar to rituximab, but is glycoengineered have low fucose sugar molecules on the antibody, resulting in more efficient B-cell depletion and reduced infusion time [9]. This may be particularly advantageous for children who may not tolerate long infusions.

These four anti-CD20 agents also differ in their targeted CD20 epitope. Rituximab and ocrelizumab share the same core epitope on the large extracellular loop of CD20 ligand [10]. Ofatumumab targets a different unique epitope that contains portions of both the large and small CD20 extracellular loops [11]. Ublituximab targets a unique epitope on the large CD20 extracellular loop that overlaps with ofatumumab, rituximab, and ocrelizumab [12].

Inebilizumab is an afucosylated humanized anti-CD19 antibody that targets and efficiently depletes CD19 expressing B lymphocytes [13]. The primary advantage of targeting CD19 over CD20 antigen is the broader depletion of B lymphocytes, spanning nearly its entire lineage. While CD20 is primarily expressed on naïve and memory B-cells in peripheral circulation, CD19 is additionally expressed in pro-/pre-B cells in the bone marrow and majority of plasma cells that can be found in peripheral circulation, secondary lymphoid organs, and bone marrow [3, 13] (Fig. 1). Plasma cells are the primary cellular source of highly target-specific antibodies, and its depletion can potentially remove the population of plasma cells that produce deleterious autoreactive antibodies, such as aquaporin-4 (AQP4) immunoglobulin (Ig) G in the case of NMOSD [3, 14].

Fig. 1.

Overview of CD19 and CD20 expression during B-lymphocyte development. CD19 expression begins in the pro-B cell stage. Both CD19 and CD20 are expressed in mature B-lymphocytes in the peripheral circulation. CD19 expression persists on plasmablasts and majority of plasma cells, even after CD20 expression has been lost

Overview of B-cell Therapies

Among the currently available anti-CD20 therapies, off-label use of rituximab is likely the most frequently encountered option in the pediatric population. Approved for medical use by the FDA in 1997, rituximab is used in a large variety of CNS inflammatory conditions such as multiple sclerosis (MS) [4, 15], autoimmune encephalitis such as anti-NMDA receptor encephalitis (NMDARE), neuromyelitis optica spectrum disorder (NMOSD), myelin oligodendrocyte glycoprotein antibody associated disease (MOGAD), opsoclonus myoclonus ataxia syndrome (OMAS), neuropsychiatric systemic lupus erythematosus, cerebral vasculitis, and others [16]. It also has the largest body of evidence to support its efficacy and safety for use in the pediatric population [6, 7, 16-19]. A recent multicenter retrospective cohort study of participants with pediatric-onset multiple sclerosis (POMS) in Europe [17] reported decreased annualized relapse rate (ARR) following rituximab initiation (ARR: 0.6 pre-treatment, 0.03 during treatment, p < 0.001). Additionally, 90% of POMS participants remained relapse-free during rituximab treatment [17]. In a separate multicenter retrospective cohort of 16 children with pediatric NMOSD, Nosadini et al. also reported a decrease in ARR following rituximab therapy (ARR: 2.2 pre-treatment, 0.6 during treatment, p = 0.003) [20]. There is less pediatric-specific safety and efficacy data available for newer therapies such as ocrelizumab and ofatumumab, which only received US Food and Drug Administration (FDA) approval in 2017 and 2020, respectively for adult patients. However, several clinical trials for pediatric patients are ongoing for some of these therapies.

OPERETTA 2 (NCT05123703) is a phase 3 multi-center randomized controlled trial studying the safety and efficacy of ocrelizumab compared to fingolimod in children and adolescents living with multiple sclerosis with an anticipated study completion date in late-2029 [21]. Encouragingly, early prospective data using ocrelizumab from 3 tertiary pediatric neuroscience centers in United Kingdom reported no evidence of disease activity (NEDA-3) in 93% of patients at 12-month follow-up with a safety profile comparable to adult literature [22]. The NEOS trial (NCT04926818) is a phase 3 trial studying the safety and efficacy of ofatumumab and siponimod compared to fingolimod in pediatric multiple sclerosis, with an anticipated completion date in mid-2029 [23]. Several case reports have described the successful use of ofatumumab in pediatric patients who were allergic to rituximab for treatment of nephrotic syndrome and opsoclonus myoclonus ataxia syndrome [24, 25]. At the time of this writing, there are no anticipated or ongoing trials involving ublituximab in the pediatric population registered through ClinicalTrials.gov despite being FDA approved for treating adults with multiple sclerosis in 2022.

Inebilizumab is the only anti-CD19 agent currently approved for treating a primary CNS inflammatory condition, specifically NMOSD. The N-MOmentum study (NCT02200770) was a phase 2/3 multi-center randomized placebo-controlled trial studying inebilizumab use for treatment of NMOSD in adults [26]. The positive findings from the study lead to its FDA approval in 2020. There is currently an ongoing phase 2 open-label multicenter study evaluating the pharmacokinetics, pharmacodynamics, and safety of inebilizumab in pediatric patients with NMOSD with an estimated study completion in 2027 (NCT05549258) [27]. The ExTINGUISH trial (NCT04372615), is an ongoing phase 2 multicenter randomized controlled trial evaluating the safety and activity of inebilizumab in NMDARE amongst adolescents and adults with an estimated completion in 2026 [28].

Safety Overview

Overall, anti-CD20 therapies such as rituximab are considered safe and effective based on studies that followed patients typically over a period of approximately 1–2 years [15, 29, 30]. Two large scale pediatric retrospective observational studies on effect of rituximab therapy [16, 18] have reported safety outcomes that are comparable to what has been observed in adults. Amongst pediatric patients, the most common immediate adverse effects included infusion related reactions (IRRs) (12–15%) and anaphylaxis (2–4%). Neutropenia (< 1%) and hypogammaglobulinemia (22–41%) can also be observed with variable timing of onset and without clear relationship to dose. Infections (8–46%) are considered one of the more common long term adverse effects with respiratory tract and urinary tract infections being most common. Severe to lethal infections have been rarely reported. [16, 18] These adverse effects along with several rare but potentially severe adverse effects such as progressive multifocal leukoencephalopathy (PML) will be discussed in more depth in the following sections.

For comparison, two large randomized controlled trials (RTCs) studying the effect of rituximab in adult MS patients reported IRRs in 67% (placebo 23%) [29] and 78% (placebo 40%) [15] of patients following the first infusion, although this number decreased (20–23%) by the second infusion [15, 29]. Two other MS studies in adult patients, a cross-sectional study and a retrospective cohort study, reported IRRs of 19% [31] and 26% [32], respectively. Rate of reported hypogammaglobulinemia associated with rituximab was more variable, anywhere from 3% in a large retrospective multicenter study of 822 MS patients [33] and up to 21% in a more recent retrospective cohort of 134 patients receiving rituximab for NMOSD and MOGAD. A smaller study of 15 NMOSD patients treated with rituximab reported low IgG levels in 11 out of 15 patients (73%). Overall infection rate from the RTCs of rituximab in adult MS patients reported infections (respiratory tract and urinary tract infections being most common) in almost 70% of study participants and there was no clear difference from placebo [15, 29]. A meta-analysis on safety and efficacy of rituximab in primarily adult MS patients found 75 of 253 patients (29.6%) had adverse events with rituximab and infections occurred in 60 patients (23.7%) [34].

For inebilizumab, the N-MOmentum trial [35] showed similar rates of adverse events in both the treatment and placebo groups in the original 28-week study period as well as the subsequent open label period at the 2 year interim analysis [35, 36]. The most common adverse reactions in participants treated with inebilizumab was urinary tract infection (11%), arthralgia (10%) and IRRs (9%). Participants in the inebilizumab groups had lower lymphocyte counts on laboratory testing compared to placebo, consistent with its mechanism of action in depleting CD20 lymphocytes. Grade 3 neutropenia occurred in 2% of the inebilizumab-treated group but was generally transient and not associated with serious infection. The open-label extension evaluation of 75 N-MOmentum participants treated with inebilizumab for ≥ 4 years reported IRRs and infections in 14.7% and 78.7% of participant, respectively [37]. IgG level was noted to decrease by 4% over the initial 28-week study period of N-Momentum trial while a similar decrease was not observed in the placebo group. For participants on inebilizumab for ≥ 4 years, immunoglobulin levels (IgG, IgM, IgA, and IgE) were observed to decrease over time [37]. However, most participants (76%) were observed to have normal IgG levels (≥ 700 mg/dL) and low IgG level was not significantly correlated with increased infection risk [37].

Case 1: Acute Infusion Reactions with Rituximab

A previously healthy 5-year-old boy developed episodes of behavioral outbursts accompanied by regression of language skills and right sided stiffening and posturing over the course of two weeks. He was brought to the hospital following an episode of unresponsiveness with cyanosis and empirically received a loading dose of levetiracetam and started on lacosamide. He required mechanical ventilation for airway protection and subsequent screening routine MRI of the brain acquired with and without contrast was unremarkable. Electroencephalogram (EEG) was only notable for bilateral background slowing with superimposed diffuse beta activity, consistent with encephalopathy and benzodiazepine use. No epileptiform discharges were observed and lacosamide was discontinued on hospital day 2, partly due to concern for exacerbating bradycardia while on dexmedetomidine for sedation. Diagnostic lumbar puncture obtained during admission showed elevated white blood cells (WBC) (38 corrected [38] cells/mm³) with lymphocytic predominance (91%) and mildly elevated protein (48 mg/dL) in the cerebrospinal fluid (CSF). His infectious workup was unremarkable. Patient’s clinical symptoms and CSF findings were consistent with probable autoimmune encephalitis, and he was treated empirically with high dose intravenous methylprednisolone (IVMP) and intravenous immunoglobulin (IVIG). Antibody testing of the CSF for autoimmune encephalitis (Mayo Clinic Laboratories, Pediatric Autoimmune Encephalopathy/CNS Disorder Evaluation, PCDEC) later returned with positive for NMDA receptor antibodies with a 1:64 titer, confirming a diagnosis of definite anti-NMDA receptor encephalitis (NMDARE) [39].

Following the initial course of IVMP and IVIG, the patient demonstrated some clinical improvement and no longer required mechanical ventilation. However, because of minimal improvement in speech, decision was made to start rituximab (750mg/m²) on hospital day 11. He received acetaminophen and diphenhydramine as pre-treatment medications 20–30 min prior to starting the rituximab infusion and completed 4 out of the 5.5 h of infusion before he developed a seizure characterized by left head and gaze deviation, unresponsiveness to noxious stimulation, and oxygen desaturation. Seizure was aborted with intranasal midazolam. Rituximab infusion was discontinued at the time of seizure. Approximately 4 h following discontinuation of rituximab, he was noted to be transiently febrile to 103 °F. He remained somnolent for at least 24 h following his seizure. A follow-up MRI brain without contrast was obtained within 12 h following the seizure. Unfortunately, interpretation of the MRI study was limited by severe motion artifact, although the non-motion degraded portions of the study were reassuring against significant abnormalities such as hemorrhage, mass, and large territory infarcts.

Due to concern for worsening NMDARE disease activity, a 3-day course of IVMP was repeated with clinical improvement. The patient was discharged on hospital day 15 with planned 6–12 month course of monthly IVIG accompanied and IVMP 30 mg/kg. Despite his seizure and fever associated with the first rituximab infusion, a second infusion of rituximab 375mg/m² with the addition of IVMP (2 mg/kg) to his pre-treatment medication regimen was given in the outpatient setting two weeks after discharge following joint decision-making with the family. He tolerated this infusion well.

Case 1: Discussion

Rituximab is not commonly associated with the adverse effect of seizures and is typically thought to improve acute symptomatic seizure burden in patients with suspected CNS inflammatory conditions by treating the underlying inflammation. There are rare case reports of intravenous rituximab infusions being associated with posterior reversible encephalopathy syndrome (PRES) or coagulopathy resulting in seizures, although neither of these complications involved patients with primary CNS autoimmune disorders [40, 41]. For our patient, his active NMDARE likely caused his seizure and subsequent fever (a dysautonomia symptom). The uneventful completion of his outpatient rituximab infusion two weeks following hospital discharge, when he was more clinically stable, further supports this notion. Alternatively, his seizure may have been provoked by an acute infusion reaction associated with rituximab (which can include fever) that lowered his seizure threshold. Regardless of the exact mechanism, rituximab was unlikely to have caused the seizure directly. In a multicenter retrospective study of 144 pediatric patients who received rituximab treatment for CNS inflammatory disease patients by Dale et al. reports a single incidence of infusion-related fever that created a transient exacerbation of seizures, although the patient’s diagnosis was not reported [16].

Acute Infusion Related Reactions

Infusion related reactions are one of the most common adverse reactions to rituximab and can occur in about 12–15% of pediatric patients [16, 18]. Infusion reactions can have a wide spectrum of presentations and may include symptoms such as rash, swelling, malaise, fevers and chills, hypotension, hypertension, tachycardia, bradycardia, headaches, nausea, chest pain, coughing, wheezing, and throat itching. Pre-treatment with a combination of antihistamine, corticosteroids, analgesics, antipyretics, and anti-emetics can improve tolerability [42]. The authors recommend a combination of acetaminophen (antipyretic), diphenhydramine (antihistamine), and corticosteroids as part of our infusion pre-treatment regimen to be given approximately 30 min prior to infusion (Table 1). Very rarely does the infusion need to be aborted. The infusion can typically be paused for symptomatic treatment and restarted later at a slower infusion rate once the symptoms have improved or resolved. Most patients (nearly 88%) who experienced infusion reactions are still able to complete the infusion and receive further doses without recurrence of infusion reactions [18]. Adding corticosteroids to the infusion pre-treatment medications have been shown to decrease the incidence of IRRs in oncology patients even during the first infusion [43, 44]. Further adjustments to the pre-medication regimen to include other medications such as cetirizine, montelukast, famotidine, or a 3-day oral prednisone pulse prior to the infusion can often be helpful in subsequent infusion attempts.

Table 1.

Suggested rituximab infusion plan for treatment of primary CNS inflammatory conditions in pediatric patients. ^aRequired for inebilizumab, otherwise recommended for patients with high endemic risk for tuberculosis. ^bCD20 count is recommended when using inebilizumab. Abbreviations: BSA, body surface area; IgA, immunoglobulin A; IgG, immunoglobulin G; IgM, immunoglobulin M; VZV, varicella zoster virus

Initial screens	Required Hepatitis B screen (hepatitis B surface antigen, hepatitis B core antibody) Complete all required vaccination prior to initial therapy whenever possible Live and live-attenuated vaccine must be given at least 4 weeks prior to starting therapy and should be avoided during therapy Serum immunoglobulin levels (IgA, IgG, IgM) Tuberculosis screening (TB QuantiFERON Gold)a Highly recommended Complete blood cell count with differential Comprehensive metabolic panel Hepatitis C screen (hepatitis C antibody) Varicella zoster virus IgG (VZV IgG) Lymphocyte subsets Pregnancy testing in individuals of child bearing potential
Pre-treatment medication	Administer approximately 30 min prior to infusion: Acetaminophen, oral: 15 mg/kg up to 650 mg or per institution preference Diphenhydramine, oral: 1 mg/kg up to 50 mg Methylprednisolone, intravenous: 1 mg/kg
Rituximab initial dosing	Fixed dosing regimen: 1000 mg for 2 doses, 14 days apart In children < 12 years and < 40 kg: 500 mg for 2 doses, 14 days apart BSA dosing regimen: 750mg/m2 (maximum 1000 mg) for 2 doses, 14 days apart For very young patients: 375mg/m2 for 4 doses, 7 days apart
Rituximab maintenance dosing	Fixed dosing regimen: 1000 mg for 1 dose, every 6 months In children < 12 years and < 40 kg: 500 mg for 1 dose, every 6 months BSA dosing regimen: 750mg/m2 (maximum 1000 mg) for 1 dose, every 6 months For very young patients: 375mg/m2 for 1 dose every 6 months
Rituximab initial infusion rate	Start at 0.5 mg/kg/hr (up to 50 mg/hr) Increase by 0.5 mg/kg/hr (up to 50 mg/hr) every 30 min as tolerated up to Maximum of 4 mg/kg/hr (up to 400 mg/hr)
Rituximab maintenance infusion rate	Start at 1 mg/kg/hr (up to 100 mg/hr) Increase by 1 mg/kg/hr (up to 100 mg/hr) every 30 min as tolerate up to Maximum of 4 mg/kg/hr (up to 400 mg/hr)
Monitoring	Follow-up clinic visits every 3–6 months during the first year and can extend to every 6–12 months once stable Monitoring labs: Complete cell count with differential, comprehensive metabolic panel, and IgG level every 3–6 months CD19b count every 2–3 months until consistent suppression is achieved Repeat MRI brain, C-spine, T-spine with and without contrast 6 months after initiating therapy to obtain new baseline

Ofatumumab is unique amongst anti-CD20 therapy due to its dosing as a monthly self-administered injection following the initial loading doses rather than being administered as an intravenous infusion. However, ofatumumab can still be associated with injection related reactions seen in up to approximately 25% of patients [45] and can include systemic symptoms such as fever, headache, myalgia, chills, and fatigue as well as local injection site reactions such as erythema, swelling, itching, and pain [46, 47]. These reactions are most common with the first dose of ofatumumab and decrease with subsequent injections [45]. The majority of these reactions are also classified as mild to moderate [48] based on Common Terminology Criteria for Adverse Events (CTCAE) grading [49] and can be managed symptomatically if they occur [47]. Pre-treatment is not recommended due to lack of observed efficacy in clinical trials [48]. Furthermore, there is also potential for steroid pre-treatment to increase occurrence of flushing, chest discomfort, hypertension, tachycardia, and abdominal pain [50], which may potentially mimic or exacerbate injection related reactions.

Anaphylaxis and Desensitization

Rarely, anaphylaxis to B-cell therapy can occur (2–4%), typically with the first dose and more rarely with the second dose [16, 18]. Anaphylaxis, as defined by the National Institute of Allergy and Infectious Diseases and Food Allergy and Anaphylaxis Network, can present with symptoms involving multiple organ systems, and often includes a combination of skin manifestations, gastrointestinal symptoms, respiratory compromise, or cardiovascular compromise [51]. However, it can be challenging sometimes to distinguish an infusion related reaction to anaphylaxis. Epinephrine should be readily available for administration as first line treatment for anaphylaxis and should be included as part of every patient’s infusion protocol in the event of a potential anaphylactic reaction.

Patients who develop an anaphylactic reaction to rituximab should ideally be switched to a different therapy, such as a more humanized anti-CD20 therapy (e.g. ofatumumab) or a different drug class altogether. However, if that is not feasible, then rituximab desensitization therapy [52, 53] may be considered in patients with confirmed IgE-mediated anaphylaxis in collaboration with allergy and immunology. Elevated tryptase and histamine, which are released by degranulating mast cells, can be helpful in confirming an IgE-mediated anaphylaxis. Serum tryptase and histamine ideally should be obtained immediately following the reaction and within 1–4 h of symptom onset with a subsequent baseline tryptase level for comparison. It is important to note that desensitization therapy should not be applied to patients who developed delayed-type hypersensitivity reactions such as skin blistering or desquamation (e.g. Steven-Johnson syndrome), or serum sickness type symptoms (e.g. Type III hypersensitivity reaction). Patients with these delayed-type hypersensitivities should avoid the provoking medication altogether.

Rituximab Dosing and Infusion Rates

In the pediatric population, there is no standardized dosing and infusion regimen for rituximab. The most frequently encountered regimens for initial rituximab dosing for pediatric patients with CNS inflammatory conditions consist of either a fixed-dose regimen or body surface area (BSA) based dosing regimen. The fixed dose regimen typically consisting of 1000 mg (or 500 mg for children younger than 12 years and less than 40 kg) dosed twice, 14 days apart, followed by additional doses as needed every 6 months [42]. The BSA-based dosing regimen is typically 750mg/m² (max 1000 mg) every 14 days for 2 doses for followed by 750mg/m² every 6 months thereafter as needed [54]. Alternatively, for very young patients 375mg/m² given every 7 days for 4 consecutive doses at onset followed by 375mg/m² every 6 months thereafter as needed [42]. Table 1 summarizes suggested dosing strategies. Unfortunately, there have been no systematic studies to help determine which dosing regimen provides superior efficacy. The decision for which regimen to prescribe often is based on anticipated patient tolerability as well as feasibility of families to return for additional infusions.

Every 6 month maintenance dosing interval for rituximab is largely based off studies in adults, whereas children have been observed to show earlier lymphocyte repopulation in a dose-dependent fashion, particularly in the beginning of the treatment course [7]. Monitoring of CD19 lymphocyte counts in the beginning of the treatment course can be helpful in identifying patients who are prone to early repopulation. The authors recommend checking CD19 lymphocyte counts every 2–3 months after starting rituximab until the patient demonstrates consistent lymphocyte suppression. Of note, peripheral CD-20 B lymphocyte counts are usually determined by using CD19 cell count, which has an overlapping distribution with circulating CD20 B-lymphocytes [55] while minimizing potential antibody interference from the presence of anti-CD20 antibodies [2]. For patients who demonstrate early repopulation prior to the expected 6 months, increasing the dose up to 1000 mg every 6 months or scheduling more frequent infusions based on the timing of CD19 lymphocyte repopulation is recommended.

With regards to infusion rates, Ghezzi et al. [42]. recommends the first infusion to be started at a rate of 50 mg/hr and increased by 50 mg/hr every 30 min to a maximum of 400 mg/hr if there are no evidence of adverse reactions. If initial infusion is well tolerated, subsequent infusions can start at a faster rate and titrated to a maximum infusion rate of 400 mg/hr if there is no evidence of adverse reactions. The authors use a similar weight-based approach (Table 1). In the event of an infusion reaction, the authors recommend pausing the infusion for 30 min, before resuming at half the previous rate and re-starting the titration (Table 2). Usually, slowing down the infusion rate is sufficient to prevent infusion reactions.

Table 2.

Summary of key management and monitoring recommendations. Abbreviations: IRRs, infusion related reactions; IgG, immunoglobulin G; IVIG, intravenous immunoglobulins; SCIG, subcutaneous immunoglobulins

IRRs	Administer pre-treatment medications approximately 30 min prior to receiving B-cell depletion therapy to minimize IRRs Slower infusion rate during first 1–2 infusions is helpful in preventing IRRs Incidence of IRRs typically improve on subsequent infusions If IRRs occur, pause the infusion, give appropriate first line symptomatic treatment (e.g. additional diphenhydramine), and restart at half the prior rate and re-titrate once IRRs have resolved
Immunosuppression	Consider adjusting or discontinuing therapy if persistent lymphopenia (< 500 cells/mm3) during while on therapy Replacement immunoglobulins (IVIG or SCIG) can be considered for patients with frequent infections and persistent hypogammaglobulinemia (IgG < 300 mg/dL)
Vaccinations	All age appropriate and seasonal vaccinations are encouraged for primary prevention Live vaccines are contraindicated while receiving B-cell depletion therapy

Case 2: Subacute to Chronic Immunosuppression

All indicated live vaccines should be administered at least 4 weeks prior to starting B-cell depletion therapy.

A previously healthy 16-year-old girl was admitted following two episodes of focal paresthesia and allodynia over a one-month period. Her MRI studies during her admission showed multiple enhancing and non-enhancing T2/FLAIR hyperintense lesions involving her brain stem, corpus callosum, periventricular area, and spinal cord. Serum antibody testing for MOG and AQP4 antibodies (Mayo Clinic Laboratories, CNS Demyelinating Disease Evaluation, CDS1) were negative. Clinical symptoms and MRI findings were consistent with a diagnosis of pediatric onset multiple sclerosis (POMS) based on the 2017 McDonald criteria [56]. She received a 5-day course of IVMP (1 g/day) with symptomatic improvement. She later received two initial doses of ocrelizumab (300 mg, 14 days apart) within 1 month of her hospital discharge. Her first infusion was incomplete due to concern for bronchospasm a little over two hours into her infusion, which is thought to be an infusion related reaction despite having received acetaminophen, diphenhydramine, and IVMP (1 mg/kg) as her pre-infusion medications given approximately 30 min prior to receiving ocrelizumab. Her symptoms resolved with administration of rescue famotidine and an additional dose of diphenhydramine. After adding cetirizine and montelukast to her pre-infusion medications, she tolerated her second infusion without issue.

Less than one month following her initial incomplete ocrelizumab infusion, she experienced an early relapse consisting of truncal paresthesia with new enhancing lower cervical and high thoracic cord lesions on MRI. CD19 lymphocyte counts were undetectable. She received a 3-day course of IVMP (1 g/day). Incidentally, she was found to have leukopenia with nadir 1.37 × 10³ cells/mm³ and severe neutropenia with absolute neutrophil count (ANC) of 20 cells/mm³ at its nadir 22 days following her last ocrelizumab infusion and 36 days following her first ocrelizumab infusion. Leukopenia and neutropenia both normalized on repeat labs one week later without requiring granulocyte colony stimulating factor (G-CSF).

While we were suspicious that her ocrelizumab could have caused her neutropenia, she was later evaluated by hematology, who felt that her neutropenia was less likely to be the direct effect of ocrelizumab, since the neutropenia resolved quickly following treatment with corticosteroids. They felt the neutropenia was more likely to be due to another idiopathic autoimmune etiology with peripheral neutrophil destruction but otherwise intact bone marrow hematopoiesis. The patient was re-challenged with ocrelizumab with close monitoring of blood cell counts after joint decision making that included family and hematology. She tolerated two subsequent maintenance infusions of ocrelizumab without further evidence of leukopenia, neutropenia, or MS relapse.

Case 2: Discussion

Neutropenia, defined as ANC < 1.5 × 10³ cells/mm³, following anti-CD20 therapy can be broadly split into two categories: early onset neutropenia (EON) which occurs less than 4 weeks after receiving last treatment and late onset neutropenia (LON) which occurs greater than 4 weeks after receiving last treatment. Our patient developed EON following her last dose of ocrelizumab 22 days prior. EON is typically regarded as more rare compared to LON (11% EON vs. 89% LON [57]), although EON is likely under-reported due to lack of close monitoring immediately following therapy, especially if neutropenia is transient and is not associated with significant symptoms or illness [58]. In a retrospective study of over 200 patients with MS or NMOSD who have received at least one rituximab infusion with the aim of evaluating for non-B-cell effects of rituximab over time, Graves et al. reported significant increases in neutrophil in the 0–3 months after the first and third cycle (typically two infusions two weeks apart) of rituximab [59].

Mechanism for both LON and EON remain poorly understood. Direct cytotoxic effects are thought to be unlikely particularly since neutrophils are not known to have significant CD20 antigen expression [57]. Other proposed mechanisms have included antineutrophil autoantibody production following B-cell depletion, and hematopoietic lineage competition due to excessive promotion of B-cell lymphopoiesis over granulopoiesis, amongst others [60-62].

While LON has been described in up to 27% of patients in oncology literature, incidence of both LON and EON are more variable in autoimmune literature [60]. Amongst pediatric patients, Dale et al. noted a single patient who developed neutropenia following rituximab treatment in their retrospective cohort of 144 patients and McAtee et al. reported 3 cases from their retrospective cohort of 468 patients [16, 18].

Patients often can recover from neutropenia with little or no intervention [57]. While administration of G-CSF can speed up the resolution of neutropenia, its potential benefit also needs to be weighed against risk for potential relapse of demyelinating disease relapse [63]. Therefore, the decision to treat neutropenia with G-CSF should be determined on an individual level. The recurrence rate of neutropenia after being re-challenged with anti-CD20 therapy is only about 20% [57], so continuation of anti-CD20 therapy is likely reasonable to consider, particularly in milder cases. Switching between rituximab and ocrelizumab also does not appear to be helpful in preventing recurrence of neutropenia [57]. Neutropenia has also been reported with newer agents such as ofatumumab and ublituximab as well [9, 64]. Interestingly, neutropenia has also been described in with inebilizumab [35], which targets CD19 rather than CD20. This suggests that B-cell depleting agents may share a similar underlying mechanism for neutropenia [62].

Infection Risk

Anti-CD20 therapy is commonly associated with an increased risk of infections in all age groups. In children, most frequent infections include upper and lower respiratory tract infections, urinary tract infections, and skin or soft tissue infections [18]. Rarely, serious infections that can be life threatening or result in death are reported (1.4–3.6%) [16, 18].

Immediately following treatment with anti-CD20 therapy, there is an increase in infection risk within the first 6 months, with the most severe infections occurring within the first month [18]. This transient increased risk for infection is likely mediated by transient leukopenia that have been observed to occur in CD4 + and CD8 + T-cells [59]. Long term infection risks include increased susceptibility to infections and longer recovery time from infections due to lymphopenia (41%) and hypogammaglobulinemia (22%) [16]. Low levels of IgM and IgG immunoglobulin isotypes can also be seen [16, 18, 65-67]. Lymphopenia (< 500 cells/mm3) and low IgG levels (< 500 mg/dL) are associated with increased infection risk [68]. Persistent lymphopenia < 500 cells/mm³ while on anti-CD20 therapy may be a sufficient reason to adjust or discontinue therapy.

The authors recommend replacement immunoglobulin therapy, such as intravenous immunoglobulin (IVIG) or subcutaneous immunoglobulins (SCIG) for patients with severe or recurrent moderate to severe infections (e.g. requiring in-hospital or surgical interventions, repeated prolonged antibiotic course) and hypogammaglobulinemia (Table 2) [69-71]. Hypogammaglobulinemia is generally considered moderate if IgG level is between 200 and 399 mg/dL, and severe if between 0 and 199 mg/dL [69, 70]. Serious consideration for replacement immunoglobulin therapy, usually in close collaboration with immunology, is encouraged by the authors if IgG is persistently less than 300 mg/dL, although this recommendation is largely derived from rheumatology literature [72]. Dose reduction or switching to alternative therapies can also be helpful in reducing the number of infections [71]. More recent data from ofatumumab long-term safety study suggests that ofatumumab is less likely to cause significant hypogammaglobulinemia with stable IgG and decreased IgM, although IgM continues to remain above lower limit of normal in most patients [45].

Unmasking of Primary Immunodeficiencies

Rarely, the use of an anti-CD20 agent can lead to the unmasking of a primary immunodeficiency that was not present prior to use of immunotherapy. While hypogammaglobulinemia can be expected in upwards of 40% of patients, these patients are generally expected to recover in the absence of repeat treatment [73]. However, a small sub-population of patients develop persistent hypogammaglobulinemia or prolonged B-cell depletion following rituximab therapy. Ottaviano et al. studied a series of 53 children who received treatment for autoimmune cytopenias (i.e. autoimmune hemolytic anemia, Evan’s syndrome, immune thrombocytopenia), and found that patients with persistent hypogammaglobulinemia 12 months after receiving rituximab had a higher risk for being diagnosed with a primary immune deficiency [74]. It is hypothesized that patients that develop persistent infections and hypogammaglobulinemia after rituximab may have a pre-existing humoral dysfunction that is subsequently unmasked and accelerated by anti-CD20 therapy administration [73].

In a recent case report by Yang et al., a 21-year-old patient was admitted for new onset refractory status epilepticus in the setting of preceding illness, consistent with febrile infection-related epilepsy syndrome (FIRES). She received rituximab as part of her treatment for presumed autoimmune encephalitis and subsequently developed panhypogammaglobulinemia. Her true baseline antibody levels were not available, but testing showed low IgA despite treatment with IVIG. Genetic testing later revealed a pathogenic heterozygous variant in TNFRSF13B gene, which has been associated with common variable immunodeficiency (CVID) and IgA deficiency. This patient likely had an underlying genetic immunodeficiency may have predisposed her to significant immune dysregulation following a provoking illness [75].

Given the potential for immunosuppressive medication to unmask other immunodeficiencies, baseline immunoglobulin and lymphocyte subset testing prior to starting immunotherapy becomes very important. Not only are these tests invaluable as a baseline reference, but they may also be diagnostic for pre-existing immunodeficiencies and can help guide treatment selection and management strategies. A low threshold to pursue additional immunologic workup, preferably in collaboration with immunology, is also recommended in patients who develop persistent hypogammaglobulinemia, despite cessation of immunotherapy.

Considerations for Therapy Initiation

Due to the risk for significant adverse effects associated with B-cell depletion therapy, screening and continued monitoring for safety and tolerability are important. Manufacturer prescribing information for ocrelizumab, ofatumumab, and ublituximab all require the following: hepatitis B virus screening due to high risk of hepatitis B reactivation with anti-CD20 therapy, and testing for baseline serum immunoglobulin levels to identify any primary hypogammaglobulinemias present prior to therapy initiation, and completing recommended vaccinations several weeks prior to starting treatment (4 weeks for live or live-attenuated vaccines and 2 weeks prior to non-live vaccines) as live vaccinations are not recommended during treatment (Table 1) [47, 76, 77]. For hepatitis B virus screening, the authors recommend testing for hepatitis B surface antigen and hepatitis B core antibody. Hepatitis B surface antibody is expected to be positive in patients with prior hepatitis B vaccinations and therefore less informative.

Manufacturer prescribing information for inebilizumab similarly recommends screening for hepatitis B virus, baseline serum immunoglobulins, and tuberculosis screening. Live or live attenuated vaccines are also not recommended during treatment. Similarly, any recommended live or live-attenuated vaccine should be administered at least 4 weeks prior to initiation of inebilizumab.

In addition to the required screening assessments above, the authors recommend the following additional tests prior to starting B cell depleting therapy: complete metabolic panel, complete blood count with differential, hepatitis C screen (hepatitis C antibody), lymphocytes subsets, varicella-zoster virus antibody screen (VZV IgG), and tuberculosis screen (TB QuantiFERON Gold) in high endemic areas (Table 1) [78].

Therapy Monitoring

As mentioned previously, total CD19 lymphocyte count is most helpful in determining if effective lymphocyte depletion is occurring during the early phases of anti-CD20 therapy initiation, until consistent lymphocyte suppression is observed. Particularly in children, there is some evidence to suggest that B-cell repopulation occurs earlier compared to adults, possibly due to variable dosing regimens or more rapid immune recovery. More frequent initial monitoring of CD19 lymphocyte counts (the authors recommend every 2–3 months) can help guide an appropriate dosing regimen for pediatric patients on anti-CD20 therapy [7, 42]. Just as CD19 lymphocyte count is the preferred method of measuring effectiveness of anti-CD20 therapy B-cell depletion to minimize potential interference from the presence of administered anti-CD20 antibodies, CD20 lymphocyte counts can similarly be used to monitor anti-CD19 therapy. In addition to monitoring for effective lymphocyte depletion, complete metabolic panel, complete blood count, and IgG immunoglobulin levels are also recommended at 3-6-month intervals early in the treatment course and can be increased to 6–12 month intervals once clinical stability is demonstrated. Repeat MRI brain, cervical spine, and thoracic spine obtained with and without contrast should be obtained 6 months following initiation of B-cell therapy to establishing a new imaging baseline in demyelinating disorders (Table 1).

Recent studies on dynamics of B-cell repopulation using extended rituximab dosing regimens on its effect on MS relapse rates, including pediatric onset MS, suggest that anti-CD20 therapies may remain effective even after repopulation of circulating B lymphocytes [79, 80], which allows for the possibility of extending dosing regimens in the future. Continued routine monitoring of CD19 lymphocyte is likely unnecessary once an effective dosing regimen has been established, but they can be helpful in cases where there is concern for medication adherence or if there is concern for disease relapse. Unexpected early B-lymphocyte repopulation in the setting of disease relapse can be concerning for the formation of anti-therapy antibodies, particularly with less humanized agents such as rituximab, and further testing for the presence of anti-rituximab antibodies can be useful.

Vaccinations

Live vaccines are not recommended during treatment due to potential risks associated with live or live-attenuated vaccines in immunosuppressed patients. Consequently, effort should be made to ensure that children received their age-appropriate vaccines, particularly live vaccines such as measles, mumps, rubella, and varicella vaccines per Centers for Disease Control and Prevention (CDC) guidelines at least 4 weeks prior to therapy initiation (Table 2) [81]. Patients who test negative for VZV IgG should also receive varicella vaccination at least 4 weeks prior to initiation of therapy if possible (Table 2). Administration of indicated non-live vaccines (e.g. influenza, COVID-19 vaccines) can continue during anti-CD20 therapy. These non-live vaccinations are recommended to given approximately 2 weeks prior to their next schedule treatment to theoretically maximize effectiveness of the vaccine. Other special vaccine considerations for immunocompromised individuals receiving B-cell depletion therapy include 3-dose series for human papillomavirus (HPV) vaccine rather than the typical 2-dose series [82] and receiving recombinant herpes zoster (Shingrix) vaccine early starting at 19 years of age instead of at age 50 [83]. The General Best Practice Guideline for Immunization for those with altered immunocompetence by the CDC Advisory Committee on Immunization Practices (ACIP) [84] can be a helpful resource for immunotherapy prescribers, as vaccine recommendations can vary with different classes of immunosuppression that confer different infection susceptibilities.

Pregnancy Considerations

Family planning is an important topic with patients of childbearing potential receiving immunotherapy. Proactive pre-conception counselling should ideally start from time of initial diagnosis with the immunotherapy selection process [85] and be an ongoing part of routine follow-up visits. The goal of such discussions is to optimize balance between risks associated with suspending immunotherapy and avoiding adverse effects to the fetus [86], and should be tailored to each patient’s disease activity and preferences. Regulatory pregnancy labeling is frequently conservative as well as restrictive [86]. For example, the prescribing information for anti-CD20 and anti-CD19 therapies recommend individuals of childbearing potential to use effective contraception while receiving treatment for at least 6 months following the last dose [47, 76, 77, 87] or 12 months for rituximab [88]. However, there is recent reassuring safety data suggesting that B-cell depleting therapies such as rituximab and ocrelizumab can be used safely 3-6 months before conception [89-91].

B-cell depletion therapies such as rituximab, ocrelizumab, ofatumumab, inebilizumab are designed to mimic human IgG antibodies both in molecular size and structure. Maternal transfer of IgG antibodies are minimal during the first trimester and increases during the second and third trimester, which suggests that their effects on fetal organogenesis during the first trimester should also be minimal [91]. From a pharmacokinetic perspective, B-cell depleting therapies have half-lives ranging from 16 to 26 days and can be expected to be eliminated from the body within about 3-6 months after exposure [92]. There is significant expert opinion consensus that pregnancy attempts can start 1-3 months after the last infusion [85, 86]. Pregnancy tests are recommended before each treatment in patients attempting to conceive so therapy may be stopped if pregnancy is achieved [85]. In the case of ofatumumab, expert consensus suggests that injections can be stopped when trying to conceive or if patient wishes to continue treatment until confirmed pregnancy, then the monthly injections can be timed with menses to decrease chance of exposure [86]. If conception is delayed, monitoring labs to assess peripheral B-cell reconstitution (CD19 or CD20) can start 6 months following last treatment and therapy can be re-dosed if patient is not pregnant after 6-9 months or if there is evidence of peripheral B-cell reconstitution [85, 92].

In general, continuation of anti-CD20 therapy during pregnancy is not recommended due to risk for B-cell depletion in newborns, but may potentially be considered for those at high risk for disease relapse during pregnancy [91, 93, 94]. Reassuringly, B-cell depletion in infants typically normalized within 6 months [91]. Otherwise, no significant fetal or maternal harm has been associated with exposure to rituximab [91], ocrelizumab [95, 96], or ofatumumab [97] immediately prior to or during pregnancy based on currently available data, obtained primarily through ongoing pregnancy exposure registries. The SOPRANINO study (NCT04998851), which seeks to evaluate B-cell levels in infants of lactating women receiving ocrelizumab, is scheduled to complete in 2025.

Of note, ublituximab is the only anti-CD20 therapy for which the prescribing information recommends pregnancy testing prior to each infusion due to early animal studies with ublituximab that resulted in both maternal and fetal harm.

Other Adverse Events

Other rare but potentially life-threatening adverse effects include immune mediated colitis and PML. Immune mediated colitis has been reported in several patients receiving ocrelizumab in the post-marketing setting, leading to an amendment in the prescribing information for ocrelizumab. A recently published case series identified 38 cases of inflammatory colitis associated with ocrelizumab with a median time to onset of 8 months (median age 43, age range 26–68) [98]. Presenting symptoms included diarrhea, abdominal pain and discomfort, blood in stool, and weight loss. Majority of the cases (36 of 38) reported serious outcomes including disability and need for hospitalization. Twenty cases reported receiving treatment, which included one or more of the following: corticosteroids, surgery, immunosuppressants, and biologics indicated for inflammatory bowel disease. Ocrelizumab was discontinued for most of the affected patients [98]. Similar reports of de novo colitis symptoms following treatment have also been reported with rituximab [99, 100].

Immunosuppressive treatments can increase the risk for PML, a rare but serious neurological disorder. Patients receiving multiple sclerosis disease modifying therapy (DMT) are at risk for PML. Symptoms of PML can vary significantly based on the regions of the brain being affected, but generally presents with acute to subacute onset cognitive and behavioral changes, ataxia, language disturbances, visual deficits, weakness, and seizures [101, 102]. PML is caused by reactivation of a dormant systemic John Cunningham Virus (JCV), which is estimated to be present between 50 and 90% of the general population [103]. Historically, the association between PML and rituximab was first noted and studied in patients with hematologic malignancies receiving rituximab as part of their treatment regimen [104-106]. As awareness of rituximab associated PML grew, findings from large epidemiologic studies evaluating the risk of PML in patients with both oncologic and autoimmune disorders lead to the inclusion of a black box warning for PML on the rituximab label within Europe (2006) and the United States (2008).

Compared to other DMTs such as natalizumab and fingolimod, which have higher incidence of PML, anti-CD20 therapies are generally considered relatively safer with lower risk for PML. A descriptive observational study published in 2019 using the FDA Adverse Event Reporting System (FAERS) database identified 786 patients who developed PML out of 100,921 MS patients (mean age 48.9 years), of which 10 patients were receiving rituximab [107]. Only a small handful of cases of PML attributed to other anti-CD20 therapies have been reported in patients not exposed to other medications with a higher risk for PML such as natalizumab [102, 108]. At the time of this writing, there has been no reported cases of PML in patients receiving inebilizumab [87]. There are also no reported cases of PML in pediatric patients on anti-CD20 therapy for the treatment of autoimmune disease [16, 18]. However, given the high risk for mortality and disability associated with PML [109] and lack of effective treatment options, clinicians should remain vigilant in considering PML on the differential when a patient is having clinical or radiographic worsening of symptoms while on B-cell depletion therapy.

Conclusion

B-cell depletion therapies, such as anti-CD20 and anti-CD19 agents, have dramatically changed the treatment landscape of CNS neuroinflammatory conditions, including but not limited to MS, MOGAD, NMOSD, autoimmune encephalitis, autoimmune epilepsies, and OMAS. As these medications continue to gain wider use in children, the general child neurologist and neuroimmunologist will also likely encounter these medications more frequently, leading to an increased need for a more nuanced understanding of their application and associated adverse effects.

In pediatric patients, off-label rituximab remains the most prescribed anti-CD20 agent due to the lack of FDA approved anti-CD20 therapies for use in children. Several pediatric trials are underway that may pave the way for FDA approval for therapies such as ocrelizumab and ofatumumab for use in POMS as well as inebilizumab for use in NMOSD and NMDARE. As new therapies become available, future studies are needed to optimize dosing regimen in children for maximum therapeutic benefit while minimizing adverse effects.

Abbreviations

ACIP

Advisory Committee on Immunization Practices

ANC

absolute neutrophil count

ARR

annualized relapse rate

AQP4

aquaporin-4

BSA

body surface area

CDC

Centers for Disease Control and Prevention

CNS

central nervous system

CSF

cerebrospinal fluid

CTCAE

Common Terminology Criteria for Adverse Events

CVID

common variable immunodeficiency

EEG

electroencephalogram

EON

early onset neutropenia

FAERS

FDA Adverse Event Reporting System

FDA

US Food and Drug Administration

FIRES

febrile infection-related epilepsy syndrome

G-CSF

granulocyte colony stimulating factor

IgA

immunoglobulin A

IgE

immunoglobulin E

IgG

immunoglobulin G

IgM

immunoglobulin M

IRR

infusion related reaction

IVIG

intravenous immunoglobulins

IVMP

intravenous methylprednisolone

JCV

John Cunningham virus

LON

late onset neutropenia

MOG

myelin oligodendrocyte glycoprotein

MOGAD

myelin oligodendrocyte glycoprotein antibody associated disease

MS

multiple sclerosis

NMDARE

anti-NMDA receptor encephalitis

NMOSD

neuromyelitis optica spectrum disorder

OMAS

opsoclonus myoclonus ataxia syndrome

POMS

pediatric onset multiple sclerosis

PML

progressive multifocal leukoencephalopathy

RBC

red blood cells

RTC

randomized controlled trial

SCIG

subcutaneous immunoglobulins

WBC

white blood cells

Data Availability

No datasets were generated or analysed during the current study.