Reimbursement Review

Zilucoplan (Zilbrysq)

Sponsor: UCB Canada Inc.

Therapeutic area: Generalized myasthenia gravis

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Clinical Review

Abbreviations

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Table 1: Background Information of Application Submitted for Review

AChR = acetylcholine receptor; gMG = generalized myasthenia gravis; NOC = Notice of Compliance; SC = subcutaneous.

Source: Product monograph for zilucoplan.¹

Introduction

Myasthenia gravis (MG) is a rare, chronic, autoimmune neuromuscular disease in which antibodies against the neuromuscular junction disrupt nerve impulse conduction, resulting in localized or generalized skeletal muscle weakness.^2,3 In most patients, MG initially affects the extraocular muscles (ocular MG) and then progresses to other muscle groups, including the bulbar and proximal limb skeletal muscles. When the disease progresses to other muscle groups, it is referred to as generalized myasthenia gravis (gMG).^4-6 The diagnosis of gMG is based on clinical presentation; serological tests to detect antibodies against anti–acetylcholine receptor (AChR) (which 80% to 90% of patients with gMG have), MuSK, and LRP4; and electrodiagnostic tests for the evaluation of neuromuscular transmission.^6,7 Approximately 80% of all patients with MG have gMG.⁶ In Canada, the incidence and prevalence of gMG are estimated at 23 per 1 million person-years and 32 per 100,000, respectively.⁸ The Myasthenia Gravis Foundation of America (MGFA) classification system groups patients with MG, according to the severity and localization of symptoms, into 5 functional classes: I (ocular manifestations only), II (mild), III (moderate), IV (severe generalized), and V (intubation or myasthenic crisis).⁹ Other validated scales used in MG to determine disease and symptom severity include the Myasthenia Gravis Activities of Daily Living (MG-ADL) scale, Quantitative Myasthenia Gravis (QMG) scale, the Myasthenia Gravis Composite (MGC) scale, and the Myasthenia Gravis Quality of Life 15-item (MG-QoL15) scale.¹⁰ Because the symptoms and disease course of gMG are highly variable and heterogenous among patients, it has been referred to as a “snowflake disease.”¹¹ Patients experience a variety of symptoms, including fatigue, droopy eyelids, diplopia, neck weakness, difficulty swallowing or chewing, speech disturbances, difficulty breathing, and upper and/or lower limb weakness.¹¹ The symptoms of gMG occur unpredictably and fluctuate in nature, intensity, and severity on a day-to-day basis and throughout a patient’s life.² Patients can also experience exacerbation, defined by 3 criteria, which necessitates inpatient treatment: temporal (a rapid, progressive, and unpredictable worsening of symptoms); subjective (deterioration of bulbo-pharyngeal function, restriction of strength for head and neck muscles or extremity muscles, which affects patients’ everyday life, or the start of weakness in breathing with reduced cough impulse); and objective (at least a 5-point improvement in QMG score with no more than 5 points from ocular MG).¹² Furthermore, MG exacerbations can deteriorate into a myasthenic crisis, in which patients experience sudden respiratory failure that requires emergency intubation or ventilation.¹² Almost 15% of patients with gMG have been classified as having refractory gMG.¹³ According to the clinical expert consulted for this review, the goals of treatment for patients with MG are to reduce MG-related morbidity and mortality, minimize treatment-associated morbidity and mortality, reduce weakness (ocular, bulbar, respiratory, axial, and extremity) associated with MG and prevent disease exacerbations or an MG crisis, and improve health-related quality of life (HRQoL).

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of zilucoplan 0.3 mg/kg administered as a subcutaneous (SC) injection once daily for the treatment of gMG in adult patients who have anti-AChR antibody–positive disease. Patients continued to receive standard therapy throughout the pivotal trial. Per sponsor request, this review focuses on the use of zilucoplan “as an add-on therapy for the treatment of adult patients with AChR antibody positive refractory gMG, defined as not achieving symptom control after:

• Treatment for at least 1 year with 2 or more of the following therapies: prednisone, azathioprine, mycophenolate cyclosporine, cyclophosphamide, methotrexate, tacrolimus, other corticosteroids for gMG, other immunosuppressive therapies (ISTs), OR

• History of treatment with at least one of these therapies for 1 year or more and required chronic plasma exchange (PLEX), intravenous immunoglobulin (IVIg), or subcutaneous immunoglobulin (SCIg) at least every 3 months for the 12 months prior to treatment with zilucoplan.”

The focus will be on comparing zilucoplan to relevant comparators and identifying gaps in the current evidence.

Perspectives of Patients, Clinicians, and Drug Programs

The information in this section is a summary of input provided by the patient and clinician groups who responded to Canada's Drug Agency (CDA-AMC) call for input and from clinical expert(s) consulted by CDA-AMC for the purpose of this review.

Patient Input

CDA-AMC received 1 patient group submission from Muscular Dystrophy Canada (MDC), a health charity that supports people affected by muscular dystrophies and related muscle diseases (neuromuscular disorders) in Canada, which included a survey (conducted with e-blasts, personalized invitations, and online patient groups) and semi-structured virtual interviews with adults living with MG. MDC also conducted an MG Journey Mapping project for adults living with MG in Canada, using virtual interviews, round table sessions, surveys, and HRQoL measures (EQ visual analogue scale, EQ-5D, MG-ADL, MG-QoL). A total of 127 participants (84 females and 43 males) aged from 22 years to 78 years from all provinces in Canada contributed to the MDC’s group submission, the majority of whom reported having gMG. Additionally, 47 people living in Canada (33 females, 14 males) with MG provided input on their hopes and expectations for zilucoplan and their everyday experiences with MG. None of the respondents included in the MDC’s patient input had experience with zilucoplan.

The respondents indicated that MG has a significant impact on productivity, fatigue and energy levels, quality of sleep, respiratory health, mobility, strength, independence, relationships and social participation, eyes and vision, speech, and swallowing. In addition to physical and mental health, quality of life, and the well-being of their families, the respondents reported that available treatments for MG have a positive impact on health outcomes but can also have negative effects (adverse events [AEs] with steroids, slow onset of medication effects, and a feeling of trial and error with medications). Patients had concerns about the long-term and sustained benefits of supportive treatments.

Patients with MG sought improved outcomes with new treatments, including decreased intensity of exacerbations and side effects, maintenance of independence, and fewer hospital admissions for serious MG-related circumstances. Patients were willing to tolerate the side effects of medications if they improved MG outcomes. In addition, respondents stated that although current medications appeared to decrease the number of exacerbations, they did not have an impact on patients’ overall quality of life. Moreover, MDC input noted that patients, families, and caregivers value the following factors when evaluating MG therapies: treatment method and delivery (e.g., invasiveness, duration, frequency of administration), potential side effects (e.g., low risk of side effects, number of side effects), HRQoL, convenience of treatment (e.g., administration at home or community centre, perceived control and flexibility, time to travel to clinic, access to parking for clinic visits), financial impact (e.g., treatment coverage by public or private insurance), and access to treatment. According to MDC input, HRQoL was noted as a key priority, and was rated higher than the convenience of a drug. MDC input stated that beyond the accessibility of treatments, patients with MG need improved treatment options with enhanced effectiveness and tolerance over the long-term.

Patients and caregivers reported varied experiences in diagnostic testing for MG. Although some respondents experienced minimal problems with testing and diagnosis, especially in cases of MG crisis or hospitalization due to MG, the majority reported significant difficulties getting diagnosed, which included a lengthy process with many missed opportunities, delayed diagnosis, misdiagnosis (such as stroke or Bell palsy), and costs incurred. According to early findings of the MG Journey Mapping project, the time from the first bothersome symptom to diagnosis ranged from 7 years to 23 years. All the respondents underwent diagnostic blood testing, and many underwent single-fibre electromyography to confirm the diagnosis.

Clinician Input

Input From Clinical Expert Consulted by CDA-AMC

The following input was provided by 1 clinical specialist with expertise in the diagnosis and management of MG.

The clinical expert noted that although there has been significant improvement in the management of patients with MG over the past few decades, there remain significant treatment gaps for patients with gMG. Given that patients with gMG experience disease morbidity while awaiting the beneficial effects of prolonged treatments that often have intolerable side effects, the expert highlighted the need for new treatments that have a faster onset of action (particularly among patients with any bulbar or respiratory involvement), are more effective, have fewer significant adverse effects, and have greater durability than current treatment options. Additionally, the clinical expert noted that patients would benefit from new treatments that are the least invasive (e.g., oral preferred over SC or IV route of administration), are less frequent (e.g., weekly or monthly dosing preferred over daily administration), allow for reduced doses of other immunosuppressive drugs, and may be used in combination with existing or future treatments that have differing mechanisms of action (e.g., the combination of a peptide complement inhibitor plus a neonatal fragment crystallizable receptor [FcR] inhibitor in patients with severe or refractory MG). Overall, the expert described the goals of treatment for patients with MG, which are to reduce MG-related morbidity and mortality, minimize treatment-associated morbidity and mortality, reduce weakness (ocular, bulbar, respiratory, axial, and extremity) associated with MG, prevent disease exacerbations and MG crises, and improve HRQoL.

The expert noted that, depending on how refractory disease is defined, approximately 10% to 15% of patients do not respond to conventional treatment. These patients require more aggressive treatments, including IVIg or PLEX, and are considered to be the target population for complement inhibitors such as zilucoplan. The expert explained that patients who were identified as refractory in the RAISE trial were in alignment with the way refractory disease would be defined in clinical practice, particularly with reference to an adequate trial of prednisone in addition to another immunosuppressive therapy (IST).

The clinical expert noted that patients are initially identified as having MG based on clinical suspicion; the diagnosis is confirmed with electrophysiology (i.e., repetitive nerve stimulation for assessment of decrement) and serology (e.g., confirmation of antibodies for anti-AChR, MuSK, and LRP4). The presence of anti-AChR antibodies is a reliable diagnostic finding with a high specificity, so the identification of patients with MG using anti-AChR antibody testing is straightforward. Nevertheless, given the rarity of the disease and the fact that the initial identification of patients with MG is based on clinical suspicion, MG is likely underdiagnosed in the population, according to the clinical expert. The expert noted that the availability and timeliness of anti-AChR antibody assays varies across Canada.

According to the expert consulted, patients who are most likely to benefit from treatment with zilucoplan are generally aligned with patients who were enrolled in the RAISE trial, who met the following criteria: anti-AChR antibody–positive gMG, MGFA disease class II to class IV, and with an MG-ADL score of 6 or greater. The exception was the criterion of a QMG score of 12 or greater in the RAISE study, which is reportedly not used commonly in clinical practice. The expert pointed out that patients who have had an adequate trial (both dose and duration) of both prednisone and at least 1 nonsteroidal immunosuppressive therapy (NSIST), or those who demonstrated intolerance to the combination of prednisone and NSIST, should be eligible for treatment with zilucoplan; such criteria limit the inclusion to patients who have been on conventional treatment for at least 6 months (and commonly 12 months), reflecting the duration needed for corticosteroids (3 to 6 months) and mycophenolate mofetil and azathioprine (likely 12 to 18 months) to produce an optimal benefit.

The MG-ADL scale is a patient-reported outcome deemed by the expert consulted to be clinically relevant (a 2-point change is deemed a minimum clinically meaningful improvement) for evaluating response to treatment. According to the clinical expert for this review, the MG-ADL can be self-administered within minutes (by patients, with supervision of a neuromuscular neurologist) and should be used as an eligibility criterion for treatment with zilucoplan and as monitoring of efficacy throughout treatment. Reasons to discontinue treatment with zilucoplan include a clear lack of response to treatment with zilucoplan (i.e., no reduction in MG-ADL after about 6 months), intolerance due to significant AEs, and the requirement for additional ongoing treatments with IVIg or PLEX despite an adequate trial of zilucoplan, according to the clinical expert.

The clinical expert explained that because MG is a rare disorder that requires nuanced management, patients with gMG should be diagnosed, treated, and monitored by a neuromuscular neurologist with experience in gMG. This is especially important with more advanced treatment options, given the resources required to use them in the management of gMG.

Clinician Group Input

CDA-AMC received 1 clinician group submission from the Neuromuscular Disease Network for Canada (NMD4C), which comprised the experience of 8 clinicians who have treated patients with gMG. The clinician group agreed with the clinical expert on unmet treatment needs, the goals of treatment, treatment response evaluations, and care management for patients with MG. NMD4C identified emergency department visits, hospitalizations, and intensive care unit admissions as additional treatment outcomes. The ability to self-administer zilucoplan at home was seen as providing patients with greater autonomy in their care management. Both the clinician group and the clinical expert consulted for this review indicated that patients with gMG who have anti-AChR antibody–positive disease would most likely benefit from treatment from zilucoplan; evidence on the efficacy of zilucoplan has not been confirmed for patients with MG who are seronegative, the clinician group noted.

Drug Program Input

Input was obtained from the drug programs that participate in the reimbursement review process. The following were identified as key factors that could potentially affect the implementation of a recommendation for zilucoplan: relevant comparators, considerations for the initiation of therapy, considerations for the continuation or renewal of therapy, considerations for the discontinuation of therapy, considerations for the prescribing of therapy, and system and economic issues. The clinical expert consulted for this review provided advice on potential implementation issues raised by the drug programs. Refer to Table 4 for more details.

Clinical Evidence

Systematic Review

Description of Studies

One phase III, multicentre, double-blind trial (RAISE, previously known as MG0010) compared the efficacy and safety of zilucoplan with placebo. The RAISE trial enrolled 174 patients, aged 18 years to 74 years, with anti-AChR antibody–positive gMG. The primary objective was to evaluate the change from baseline to week 12 in MG-ADL score. Key secondary end points included changes from baseline to week 12 in QMG score, MGC score, and Myasthenia Gravis Quality of Life 15-item revised (MG-QoL15r) score. Additional secondary end points included the achievement of minimal symptom expression (MSE) at week 12 without rescue therapy, an MG-ADL responder rate at week 12 without rescue therapy, and a QMG responder rate at week 12 without rescue therapy. The achievement of minimal manifestation status per Myasthenia Gravis Foundation of America Post-Intervention Status (MGFA-PIS) at week 12 without rescue therapy was an exploratory end point in the RAISE study. Patients with refractory anti-AChR antibody–positive gMG were included in exploratory subgroup analyses. Outcomes reported for patients with refractory gMG included change from baseline to week 12 in MG-ADL score, QMG score, MGC score, and MG-QoL15r score; the achievement of MSE at week 12 without rescue therapy; an MG-ADL responder rate at week 12 without rescue therapy; and a QMG responder rate at week 12 without rescue therapy.

This review presents data for patients in the overall trial population, which aligns with the Health Canada indication. The Clinical Study Report, with a data cut-off date of December 30, 2021, was the primary data source for the RAISE study. Per sponsor request, this review focuses on the sponsor’s reimbursement request for patients with refractory gMG, which is similar to the reimbursement criteria for the comparator therapy, eculizumab. Eculizumab received a positive final recommendation in 2020 from CADTH for patients with refractory gMG, which is defined as not achieving symptom control after:

• “an adequate trial of two or more immunosuppressive therapies (ISTs), either in combination or as monotherapy in the previous 12 months, OR

• an adequate trial of at least one IST and chronic plasmapheresis or plasma exchange or intravenous immunoglobulin at least four times (every three months) in the previous 12 months.”

In the overall population, 56.9% of patients with gMG were female, 43.1% were male, and the mean age was 53.0 years (standard deviation [SD] = 15.1 years). At screening, most patients had a diagnosis of MGFA class III disease (67.2%), generalized symptoms at disease onset (64.4%), disease onset at a mean of 43.8 years (SD = 18.0 years), and a mean disease duration of 9.2 years (SD = 9.9 years). More patients had higher scores in the stratified randomization of MG-ADL scores (62.1% of patients had a score ≥ 10) and QMG scores (56.3% of patients had a score ≥ 18). Most disease characteristics and patient histories were similar in the 2 treatment groups, but a higher proportion of patients had undergone thymectomy in the zilucoplan group (52.3%) than in the placebo group (42.0%). Patients had similar between-group mean baseline scores on the MG-ADL and QMG scales. Patients were considered to have refractory gMG in the RAISE trial if they had received treatment for at least 1 year with 2 or more of the following therapies: prednisone, azathioprine, mycophenolate mofetil, cyclosporine, cyclophosphamide, methotrexate, tacrolimus, rituximab, eculizumab, other corticosteroids for gMG, or other ISTs; or had a history of treatment with at least 1 of the aforementioned therapies for 1 year or more and required chronic PLEX, IVIg, or SCIg at least every 3 months during the 12 months before enrolment. No patients in the RAISE trial had received eculizumab as a past or baseline gMG therapy, and 1 patient had received rituximab as a past therapy, but the trial excluded patients who had received rituximab in the 12 months before baseline and, therefore, no patients were receiving rituximab at baseline.

Efficacy Results

Efficacy results were summarized using the data cut-off date of December 30, 2021, for the following outcomes: changes from baseline to week 12 in MG-ADL score, QMG score, and MG-QoL15r score; achievement of MSE at week 12 without rescue therapy; an MG-ADL responder rate at week 12 without rescue therapy; a QMG responder rate at week 12 without rescue therapy; and minimal manifestation status per MGFA-PIS at week 12 without rescue therapy.

Change From Baseline to Week 12 in the MG-ADL Score

In the overall population, the least squares (LS) mean change from baseline to week 12 in the MG-ADL score was –4.39 (standard error [SE] = 0.45) in the zilucoplan group and –2.30 (SE = 0.44) in the placebo group (LS mean difference of –2.09; 95% confidence interval [CI], –3.24 to –0.95; P < 0.001), which favoured zilucoplan. In the refractory subpopulation, the LS mean change from baseline to week 12 in the MG-ADL score was –4.72 (SE = 0.58) in the zilucoplan group and –1.62 (SE = 0.58) in the placebo group (LS mean difference of –3.11; 95% CI, –4.69 to –1.52; nominal P < 0.001).

Change From Baseline to Week 12 in the QMG Score

In the overall population, the LS mean change from baseline to week 12 in the QMG score was −6.19 (SE = 0.56) in the zilucoplan group and −3.25 (SE = 0.55) in the placebo group (LS mean difference of –2.94; 95% CI, –4.39 to –1.49; P < 0.001), which favoured zilucoplan. In the refractory subpopulation, the LS mean change from baseline to week 12 in the QMG score was –6.08 (SE = 0.76) in the zilucoplan group and –2.76 (SE = 0.75) in the placebo group (LS mean difference of –3.32; 95% CI, –5.42 to –1.23; nominal P < 0.001).

Change From Baseline to Week 12 in the MGC Score

In the overall population, the LS mean change from baseline to week 12 in the MGC score was –8.62 (SE = 0.81) in the zilucoplan group and –5.42 (SE = 0.79) in the placebo group (LS mean difference of –3.20; 95% CI, –5.24 to –1.16; P = 0.0023), which favoured zilucoplan. In the refractory subpopulation, the LS mean change from baseline to week 12 in the MGC score was –7.85 (SE = 1.09) in the zilucoplan group and –4.17 (SE = 1.07) in the placebo group (LS mean difference of –3.68; 95% CI, –6.65 to –0.72; nominal P = 0.0156).

Change From Baseline to Week 12 in the MG-QoL15r Score

In the overall population, the LS mean change from baseline to week 12 in the MG-QoL15r score was –5.65 (SE = 0.77) in the zilucoplan group and –3.16 (SE = 0.76) in the placebo group (LS mean difference of –2.49; 95% CI, –4.45 to –0.54; P = 0.0128), which favoured zilucoplan. In the refractory subpopulation, the LS mean change from baseline to week 12 in the MG-QoL15r score was –5.63 (SE = 0.96) in the zilucoplan group and –2.36 (SE = 0.95) in the placebo group (LS mean difference of –3.28; 95% CI, –5.89 to –0.67; nominal P = 0.0145).

Achievement of MSE at Week 12 Without Rescue Therapy

In the overall population, the percentage of patients who achieved MSE (an MG-ADL score of 0 or 1) at week 12 without rescue therapy was 14.0% in the zilucoplan group and 5.8% in the placebo group (between-group difference of 8.2%; 95% CI, –0.% to 17.0%). In the refractory subpopulation, the number of patients who achieved MSE at week 12 without rescue therapy was 2 of 44 patients (4.5%) in the zilucoplan group and 0 of 43 patients in the placebo group.

Achievement of at Least a 3-Point Reduction in MG-ADL Score at Week 12 Without Rescue Therapy

In the overall population, the percentage of patients who were MG-ADL responders (at least a 3-point decrease in MG-ADL score) at week 12 without rescue therapy was 73.1% in the zilucoplan group and 46.1% in the placebo group (between-group difference of 27.0%; 95% CI, 12.9% to 41.1%). In the refractory subpopulation, the number of patients who were MG-ADL responders at week 12 without rescue therapy was 33 of 44 patients (75.0%) the zilucoplan group and 17 of 42 patients (40.5%) in the placebo group.

Achievement of at Least a 5-Point Reduction in QMG Score at Week 12 Without Rescue Therapy

In the overall population, the percentage of patients who were QMG responders (at least a 5-point decrease in QMG score) at week 12 without rescue therapy was 58.0% in the zilucoplan group and 33.0% in the placebo group (between-group difference of 25.0%; 95% CI, 10.5% to 39.5%). In the refractory subpopulation, the number of patients who were QMG responders at week 12 without rescue therapy was 24 of 43 patients (55.8%) in the zilucoplan group and 11 of 41 patients (26.8%) in the placebo group.

Minimal Manifestation Status per MGFA-PIS at Week 12 Without Rescue Therapy

In the overall population, the number of patients who achieved minimal manifestation status per MGFA-PIS at week 12 without rescue therapy was 22 of 78 (28.2%) in the zilucoplan group and 16 of 83 (19.3%) in the placebo group (between-group difference of 10.9%; 95% CI, –1.5% to 23.2%). The number of patients who achieved minimal manifestation status per MGFA-PIS at week 12 without rescue therapy was not reported in the refractory subpopulation.

Harms Results

The analysis population for harms included all patients who received at least 1 dose of study drug. Patients were grouped according to the treatment received, using data from the December 30, 2021, data cut-off date.

The number of patients in the overall gMG population with at least 1 treatment-emergent adverse event (TEAE) was 66 of 86 patients (76.7%) in the zilucoplan group and 62 of 88 patients (70.5%) in the placebo group. The most common TEAEs occurring in greater than 5% of patients in either the zilucoplan group or the placebo group, respectively, were headache (15.1% versus 15.9%), injection-site bruising (16.3% versus 9.1%), MG (10.5% versus 9.1%), diarrhea (10.5% versus 2.3%), injection-site pain (9.3% versus 3.4%), urinary tract infection (8.1% versus 4.5%), contusion (8.1% versus 3.4%), increased lipase (8.1% versus 1.1%), nasopharyngitis (5.8% versus 3.4%), vomiting (3.5% versus 5.7%), rash (3.5% versus 5.7%), and increased amylase (5.8% versus 2.3%).

In the refractory subpopulation, the number of patients with at least 1 TEAE was 39 of 44 patients (88.6%) in the zilucoplan group and 34 of 44 patients (77.3%) in the placebo group. The most common TEAEs occurring in greater than 10% of patients in either the zilucoplan group or the placebo group, respectively, were headache (20.5% versus 15.9%), MG (13.6% versus 13.6%), injection-site bruising (15.9% versus 11.4%), diarrhea (15.9% versus 2.3%), and vomiting (4.5% versus 11.4%).

The number of patients in the overall gMG population with at least 1 serious adverse event (SAE) was 11 of 86 patients (12.8%) in the zilucoplan group and 13 of 88 patients (14.8%) in the placebo group. The most common SAEs reported in at least 2% of patients in either the zilucoplan group or the placebo group, respectively, were MG (2.3% versus 5.7%), COVID-19 (1.2% versus 2.3%), and COVID-19 pneumonia (1.2% versus 2.3%).

In the refractory subpopulation, the number of patients with at least 1 SAE was 6 of 44 patients (13.6%) in the zilucoplan group and 8 of 44 (18.2%) in the placebo group. SAEs specified by system organ class were not reported for the refractory subpopulation.

In the overall gMG population, 4 of 86 patients (4.7%) in the zilucoplan group stopped the study treatment due to AEs, as did 2 of 88 patients (2.3%) in the placebo group. Withdrawals due to AEs in the zilucoplan group were due to (1 patient [1.2%] each) aphthous ulcer, mouth ulceration, COVID-19, and increased hepatic enzyme; the TEAE of aphthous ulcer and the TEAE of COVID-19 were considered serious, with the latter having a fatal outcome. Withdrawals due to AEs in the placebo group were due to (1 patient [1.1%] each) cerebral hemorrhage and hyperemesis gravidarum; both TEAEs were considered serious and the TEAE of cerebral hemorrhage had a fatal outcome. In the overall gMG population, the 1 patient (1.2%) in the zilucoplan group who died experienced an SAE leading to death due to COVID-19 and COVID-19 pneumonia, and the 1 patient (1.1%) in the placebo group who died experienced an SAE leading to death due to cerebral hemorrhage.

In the refractory subpopulation, 1 of 44 patients (2.3%) stopped the study treatment in the zilucoplan group, and no patients stopped the study treatment in the placebo group. The reasons for withdrawals due to AEs were not reported for the refractory subpopulation.

Notable Harms

Infections occurred in 23 of 86 patients (26.7%) in the zilucoplan group and 16 of 88 patients (18.2%) in the placebo group (between-group difference of 8.6%; 95% CI, –3.8% to 20.9%). Of these, 4 patients (4.7%) in the zilucoplan group and 4 patients (4.5%) in the placebo group had serious infections. AEs of special interest were not reported for the refractory subpopulation.

Critical Appraisal

Randomization appeared to be adequate in the RAISE trial. Treatment groups were balanced overall on demographic and disease characteristics, indicating that randomization was likely successful, and the risk of selection bias was low. Randomization was stratified by baseline MG-ADL score (≤ 9 versus ≥ 10), QMG score (≤ 17 versus ≥ 18), and geographic region (East Asia, Europe, and North America). The instruments used to evaluate the primary and secondary efficacy outcomes (MG-ADL, QMG, MGC, MG-QoL15r) were appropriate, and their psychometric properties have been investigated in patients with MG, although no minimal important differences (MIDs) have been estimated for the MG-QoL15r. Minimal manifestation status per MGFA-PIS without rescue therapy was based on clinician-assessed patient symptoms of MG after the initiation of MG-specific therapy, and was intended to capture patients who may not meet the definition of complete stable remission or pharmacologic remission but who have muscle weakness based on careful examination; this was an exploratory end point and no MID has been validated in the indicated population. Results for minimal manifestation status per MGFA-PIS at week 12 were not reported for the refractory subpopulation. There was low risk of bias for allocation concealment because patients and study staff were blinded to treatment assignment, both treatments were identical, and the unblinding of treatment assignment was not permitted before the initiation of rescue therapy.

Patients with refractory disease represented about one-half of the enrolled patients with gMG. The selection criteria for the refractory subgroup were specified a priori and were similar to the criteria used to define patients with refractory gMG in other randomized clinical trials (RCTs) (e.g., the REGAIN trial for eculizumab). Baseline characteristics were, overall, similar between treatment groups, so concerns regarding the prognostic balance of the refractory subpopulation were low. The population included in the refractory subgroup is adequately reflective of patients with refractory gMG in the clinical setting in Canada, according to the clinical expert consulted by the review team. Although subgroup analyses were not adjusted for multiplicity and not powered to detect differences among strata, results from the refractory subgroup can be compared for consistency with analyses of the overall trial population in the RAISE trial, and they share the same limitations as those analyses. Results for the refractory subgroup showed consistency with the overall trial population across all outcomes.

There was a notable proportion of patients with important protocol deviations in both the zilucoplan group (38.4%) and the placebo group (36.4%). Deviations included prohibited concomitant medication use (changes to gMG conventional medications, use of prohibited concomitant medications, and changes to cholinesterase inhibitor dosing less than 10 hours before evaluation), which affected 24.7% of patients, and changes related to the inclusion criteria (not withholding acetylcholinesterase inhibitor therapy for at least 10 hours before QMG assessment, changes in corticosteroids in the 30 days before baseline, and receiving inpatient treatment with IVIg that was not reported as rescue therapy), which affected 7.5% of patients. The proportion of patients with deviations during the study were balanced between groups. Multiple imputation (MI) methods were used to account for missing data in the primary and secondary end points, based on assumptions of missing not at random (MNAR), which assumed missingness to be related to the study drug or to the receipt of rescue medication, and missing at random (MAR). Based on the specified approach, each of these were imputed with either their baseline value or the last observed value (whichever was worse). Because the rate of intercurrent event (ICE) 1 (receipt of rescue therapy with IVIg, PLEX, or eculizumab) was higher in the placebo group (11%) than in the zilucoplan group (5%), this appeared to be an overly pessimistic approach that could possibly introduce bias in favour of zilucoplan; however, the imbalance was not large enough to raise serious concerns about biased treatment effects. Sensitivity analyses to account for the censoring of patients who experienced treatment failure were also conducted. Although the supplemental analyses and sensitivity analyses of the primary analysis did not adequately assess the potential bias related to missing data, concerns about losses to follow-up were low, because approximately 95% of patients with gMG in the overall population and the refractory subpopulation completed the RAISE study, with balanced proportions between treatment arms.

External Validity

Per sponsor request, the focus of this review was on the sponsor’s reimbursement request, which was narrower than the Health Canada indication. The reimbursement request aligned with the criteria for the refractory subgroup of the RAISE trial (i.e., zilucoplan as add-on therapy for the treatment of adult patients with anti-AChR antibody–positive refractory gMG, defined as not achieving symptom control after treatment for at least 1 year with 2 or more of the following therapies: prednisone, azathioprine, mycophenolate cyclosporine, cyclophosphamide, methotrexate, tacrolimus, other corticosteroids for gMG, other ISTs; or a history of treatment with at least 1 of these therapies for 1 year or more and required chronic PLEX, IVIg, or SCIg at least every 3 months for the 12 months before treatment with zilucoplan). The requested reimbursement criteria also aligned with the reimbursement criteria for the comparator therapy, eculizumab, which received a positive final recommendation in 2020 from CADTH.

The 49% of patients in the RAISE trial whose disease was not refractory were not included in the reimbursement request. Although the clinical expert consulted for this review agreed that there is a current unmet need in patients who are nonrefractory and have responded inadequately to the existing, standard gMG therapies, that population was not the focus of this review.

According to the expert, the stratification of patients during randomization by baseline scores on the MG-ADL and the QMG scales appeared to be appropriate for ensuring an equal distribution of patients by disease severity. There were no specific patient populations missing from eligibility in the trial who might otherwise be considered eligible in clinical practice, the expert noted. In line with patients who were not eligible for the RAISE trial, the expert emphasized that patients with MuSK-positive MG should not be treated with a complement inhibitor, such as zilucoplan, because of its mechanism of action; rather, patients with MG and MuSK-positive serology would benefit from treatment with a neonatal FcR inhibitor (e.g., efgartigimod alfa).

Looking at the types and duration of prior conventional treatments received by patients in the RAISE trial, the clinical expert agreed that all patients in the RAISE trial were adequately managed on conventional therapy for gMG at time of enrolment and were reflective of patients who experience an unmet need in clinical settings in Canada.

Moreover, the RAISE trial did not provide evidence for the comparisons between zilucoplan and other currently available active treatments for gMG. The most relevant comparators for zilucoplan among patients with refractory gMG include other complement inhibitors (i.e., eculizumab, ravulizumab), neonatal FcR inhibitors (e.g., efgartigimod alfa), chronic IVIg, and chronic PLEX, according to the expert. Rituximab as a comparator is less applicable because access for patients with gMG in Canada is limited, its use off-label, there is a lack of rigorous clinical trial evidence for patients with anti-AChR antibody–positive MG, and there is some evidence of improved benefit among patients with MuSK-positive MG.

The expert agreed that the primary end point of change from baseline in MG-ADL score is an important outcome for evaluating treatment response and is aligned with clinical practice, including the threshold used in the RAISE trial. MIDs used for the QMG and MGC scores were also noted by the expert to align with the literature for thresholds validated in patients with MG. According to the expert, most patients with gMG would be assessed at approximately 12 weeks to evaluate treatment response, as was done for all end points in the RAISE trial, with additional assessments at the 3-month or 4-month time point to assess responsiveness or maintenance of response.

GRADE Summary of Findings and Certainty of the Evidence

The selection of outcomes for the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach was based on the sponsor’s Summary of Clinical Evidence, consultation with clinical experts, and input received from patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with expert committee members: activities of daily living (MG-ADL score), disease severity (QMG score), treatment response (MGC score), HRQoL (MG-QoL15r score), and harms (infections).

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. The target of the certainty of evidence assessment was the presence or absence of a clinically important change from baseline to week 12 scores in the MG-ADL, the QMG, and the MGC, based on thresholds identified in the literature. The certainty of evidence assessments for change from baseline to week 12 in the MG-QoL15r score, number of patients achieving MSE at week 12 without rescue therapy, MG-ADL responder rate at week 12 without rescue therapy, QMG responder rate at week 12 without rescue therapy, number of patients with minimal manifestation status per MGFA-PIS at week 12 without rescue therapy, and infections were based on the presence or absence of any (nonnull) effect.

Results of GRADE Assessments

Table 2 presents the GRADE summary of findings for zilucoplan versus placebo in patients with anti-AChR antibody–positive gMG.

Table 2: Summary of Findings for Zilucoplan Versus Placebo for Adult Patients With Anti-AChR Antibody–Positive gMG

AChR = acetylcholine receptor; CI = confidence interval; gMG = generalized myasthenia gravis; LSM = least squares mean; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA-PIS = Myasthenia Gravis Foundation of America Post-Intervention Status; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item – Revised; MID = minimal important difference; mITT = modified intention to treat; MSE = minimal symptom expression; NR = not reported; OR = odds ratio; QMG = Quantitative Myasthenia Gravis; RCT = randomized controlled trial.

Note: Study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the table footnotes.

^aRated down 1 level for serious imprecision. Based on a 2-point MID identified in the literature, the 95% CI included the possibility of little-to-no difference and clinically important benefit. The 2-point MID has been estimated for change within an individual patient and was applied in the absence of an estimate of a between-group MID.

^bBased on the testing procedure used for the secondary end points, the CI was 98.75% for MG-ADL responders, 98.3% for QMG responders, and 97.5% for the achievement of MSE in the statistical hierarchy, using the Holm’s procedure. The associated CI was generated using a post hoc analysis upon request by the review team.

^cRisk difference (95% CI) was not included in the sponsor’s planned analyses; the absolute risk difference was requested by the review team for interpretation purposes.

^dRated down 2 levels for very serious imprecision. In the absence of an identified threshold in the literature or by the clinical expert consulted by CDA-AMC for the review, the null was used as the threshold. The 95% CI included the null. The CDA-AMC review team judged the effect estimate to be large and the number of events to be small, raising concerns about prognostic balance and potential overestimation of the true effect.

^eRated down 1 level for serious imprecision. In the absence of an identified threshold in the literature or by the clinical expert consulted by CDA-AMC for the review, the null was used as the threshold. The point estimate and the entire CI suggested potential benefit. The CDA-AMC review team judged the effect estimate to be large and the sample size to be small, raising concerns about prognostic balance and potential overestimation of the true effect.

^fRated down 1 level for serious imprecision. Based on an MID that ranged from 2 to 3 identified in the literature, the 95% CI included the possibility of little-to-no difference and clinically important benefit. This MID has been estimated for change within an individual patient and was applied in the absence of an estimate of a between-group MID.

^gRated down 1 level for serious imprecision. In the absence of an identified threshold in the literature or by the clinical expert consulted by CDA-AMC for the review, the null was used as the threshold. The CDA-AMC review team judged the effect estimate to be large and the sample size to be small, raising concerns about prognostic balance and potential overestimation of the true effect.

^hRated down 1 level for serious imprecision. Based on a 3-point MID identified in the literature, the 95% CI included the possibility of little-to-no difference and clinically important benefit. This MID has been estimated for change within an individual patient and was applied in the absence of an estimate of a between-group MID

ⁱRated down 2 levels for very serious imprecision. In the absence of an identified threshold in the literature or by the clinical expert consulted by CDA-AMC for the review, the null was used as the threshold. The 95% CI included the null. The effect estimate was based on a small sample size. This analysis was not adjusted for multiplicity and the results should be considered as supportive evidence.

^jRated down 1 level for serious imprecision. In the absence of an identified threshold in the literature or by the clinical expert consulted by CDA-AMC for the review, the null was used as the threshold. The effect estimate was based on a small sample size.

^kRated down 2 levels for very serious imprecision. In the absence of an identified threshold in the literature or by the clinical expert consulted by CDA-AMC for the review, the null was used as the threshold. The 95% CI included the null. The effect estimate was based on a small sample size.

Source: RAISE Clinical Study Report.¹⁴

Long-Term Extension Studies

Description of Studies

The sponsor included 1 ongoing, phase III, multicentre, open-label extension study (RAISE-XT) of adult patients with anti-AChR antibody–positive gMG (the overall gMG population and the refractory subpopulation) who had previously participated in a double-blind trial of zilucoplan 0.3 mg/kg or placebo to evaluate the long-term efficacy and safety of zilucoplan. The primary outcome of the RAISE-XT study was the incidence of TEAEs, defined as an AE starting on or after the time of first administration of the study drug and up to and including 40 days after the final dose (or last contact). Secondary efficacy outcomes included change from baseline to RAISE-XT extension (E) week 12 (E12) in the MG-ADL score, the QMG score, the MGC score, and the MG-QoL15r score. Exploratory outcomes in the RAISE-XT study included minimal manifestation status per MGFA-PIS at week E12 without rescue therapy, responder rates for the MG-ADL, QMG, and MGC scores at week E12 without rescue therapy, and the achievement of MSE (an MG-ADL score of 0 or 1) at week E12 without rescue therapy. In the overall gMG population, 93 patients in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 90 patients the placebo and zilucoplan 0.3 mg/kg group were similar in mean age (53 years and 54 years), the proportion of female (56% and 53%) and male patients (44% and 47%), and the proportion of patients enrolled from North America (57% and 54%).

Efficacy Results

Change From Parent-Study Baseline to Week E12 in the MG-ADL Score

In the overall population, the LS mean change from parent-study baseline to week E12 in the MG-ADL score was –5.90 (SE = 0.47) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –6.17 (SE = 0.59) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the mean change from parent-study baseline to week E12 in the MG-ADL score was –6.61 (SE = 0.63) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –6.24 (SE = 0.71) in the placebo and zilucoplan 0.3 mg/kg group.

Change From Parent-Study Baseline to Week E12 in the QMG Score

In the overall population, the LS mean change from parent-study baseline to week E12 in the QMG score was –8.78 (SE = 0.66) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –8.53 (SE = 0.79) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the mean change from parent-study baseline to week E12 in the QMG score was –8.18 (SE = 0.71) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –8.34 (SE = 1.09) in the placebo and zilucoplan 0.3 mg/kg group.

Change From Parent-Study Baseline to Week E12 in the MGC Score

In the overall population, the LS mean change from parent-study baseline to week E12 in the MGC score was –11.77 (SE = 0.86) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –12.30 (SE = 1.12) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the mean change from parent-study baseline to week E12 in the MGC score was –11.83 (SE = 1.17) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –13.34 (SE = 1.35) in the placebo and zilucoplan 0.3 mg/kg group.

Change From Parent-Study Baseline to Week E12 in the MG-QoL15r Score

In the overall population, the LS mean change from parent-study baseline to week E12 in the MG-QoL15r score was –9.92 (SE = 0.95) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –8.07 (SE = 1.08) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the mean change from parent-study baseline to week E12 in the MG-QoL15r score was –9.46 (SE = 1.15) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –9.34 (SE = 1.32) in the placebo and zilucoplan 0.3 mg/kg group.

Exploratory End Points

The number of patients who achieved MSE (an MG-ADL score of 0 or 1) at week E12 without rescue therapy, were responders on the MG-ADL or the QMG at week E12 without rescue therapy, and who achieved minimal manifestation status per MGFA-PIS at week E12 without rescue therapy were numerically higher in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group than in the placebo and zilucoplan 0.3 mg/kg group.

Harms Results

The number of patients in the overall population who experienced at least 1 TEAE was 89 of 93 patients (95.7%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 86 of 90 patients (95.6%) in the placebo and zilucoplan 0.3 mg/kg group. The most common TEAEs occurring in 10% of patients or greater in any group (the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo and zilucoplan 0.3 mg/kg group, respectively) were MG (29% and 29%), COVID-19 (39% and 31%), headache (19% and 22%), nasopharyngitis (23% and 17%), arthralgia (20% and 13%), diarrhea (19% and 13%), fatigue (18% and 13%), nausea (15% and 17%), upper respiratory tract infection (14% and 18%), urinary tract infection (16% and 14%), pain in extremity (16% and 10%), cough (12% and 12%), fall (11% and 10%), back pain (11% and 10%), rash (10% and 10%), vomiting (11% and 6%), injection-site bruising (4% and 11%), and oropharyngeal pain (1% and 10%). The number of patients in the refractory subpopulation who experienced at least 1 TEAE was 82 of 85 patients (96.5%) in the zilucoplan 0.3 mg/kg group (zilucoplan 0.3 mg/kg and 0.3 mg/kg and placebo and zilucoplan 0.3 mg/kg groups combined). The most common TEAEs occurring in 10% of patients or greater in the refractory subpopulation in any group (the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo and zilucoplan 0.3 mg/kg group, respectively) were MG (42% and 38%), COVID-19 (44% and 24%), headache (21% and 21%), arthralgia (26% and 10%), nasopharyngitis (16% and 19%), urinary tract infection (19% and 12%), upper respiratory tract infection (19% and 12%), diarrhea (16% and 12%), nausea (12% and 17%), fall (14% and 12%), back pain (14% and 12%), and pain in extremity (16% and 10%).

The number of patients in the overall population who experienced at least 1 serious TEAE was 35 patients (37.6%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 29 patients (32.2%) in the placebo and zilucoplan 0.3 mg/kg group. Serious TEAEs occurring in 2% of patients or greater in any group (the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo and zilucoplan 0.3 mg/kg group, respectively) were MG (9.7% and 11.1%), COVID-19 pneumonia (4.3% and 1.1%), myocardial infarction (4.3% and 0.0%), pneumonia (3.2% and 1.1%), cholecystitis (2.2% and 1.1%), Staphylococcus bacteremia (2.2% and 0.0%), atrial fibrillation (2.2% and 0.0%), cardiac arrest (2.2% and 0.0%), cellulitis (2.2% and 2.2%), and large intestine polyp (0.0% and 2.2%). The number of patients in the refractory subpopulation who experienced at least 1 serious TEAE was 20 of 43 patients (46.5%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 15 of 42 patients (35.7%) in the placebo and zilucoplan 0.3 mg/kg group.

TEAEs resulting in permanent withdrawal of the study drug in the overall population were reported in 21 of 183 patients (11.4%), with 9 of 93 patients (9.7%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 12 of 90 patients (13.3%) in the placebo and zilucoplan 0.3 mg/kg group. The most common TEAE resulting in permanent withdrawal of the study drug was MG (6 patients [3.3%]). The number of patients in the refractory subpopulation who experienced at least 1 serious TEAE was 11 of 85 patients (12.9%), 2 of 43 (4.7%) of whom were in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 9 of 42 (21.4%) of whom were in the placebo and zilucoplan 0.3 mg/kg group.

A total of 6 patients died in the RAISE-XT study. The number of patients who experienced a TEAE leading to death was 3 patients (3.2%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 1 patient (1.1%) in the placebo and zilucoplan 0.3 mg/kg group. In the zilucoplan 0.3 mg/kg and 0.3 mg/kg group, the TEAEs leading to death were cardiac arrest (2 patients [2.2%]) and head injury (1 patient [1.1%]). In the placebo and zilucoplan 0.3 mg/kg group, the TEAE was death (1 patient [1.1%]). All 4 TEAEs leading to death resulted in permanent withdrawal of the study drug. Two patients in the refractory subpopulation died during the study; 1 patient in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group had a TEAE of death. No details were reported for the deaths in the refractory subpopulation.

The number of patients who experienced any infections was 67 patients (72.0%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 65 patients (72.2%) in the placebo and zilucoplan 0.3 mg/kg group. Of these, the number of patients who experienced serious infections was 16 (17.2%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 11 (12.2%) in the placebo and zilucoplan 0.3 mg/kg group. The number of patients in the refractory subpopulation who experienced any infections was 34 of 43 (79.1%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 33 of 42 (78.6%) in the placebo and zilucoplan 0.3 mg/kg group.

Critical Appraisal

The RAISE-XT study was an open-label, noncomparative extension of the RAISE parent study. The key limitation related to the absence of a comparator group is that patients were not randomized to treatment groups (whereas patients in the parent study were randomized to zilucoplan or placebo, all patients in the extension study received zilucoplan), which precludes inferences that any observed differences are due to treatments received. Importantly, treatment efficacy and harms for patients with longer follow-up should be interpreted cautiously, as it cannot be determined whether the findings are due to the natural history of the disease, study treatments including concomitant therapies, or other unknown factors. The number of patients with missing data for the secondary end point of change from parent-study baseline to week E12 of the RAISE-XT study was similar in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo group, and the number of patients with missing data was low, overall, for scores on the MG-ADL (4%), the QMG and the MGC (6%), and the MG-QoL15r (7%), despite a lack of imputation. The proportions of patients with missing data for change from parent-study baseline to week E12 in the refractory subpopulation were also similar in the 2 treatment groups, at 7% for MG-ADL score, 8% for QMG and MGC scores, and 9% for MG-QoL15r score. A greater proportion of patients had missing data for the exploratory end points of the RAISE-XT study, including the achievement of MSE (9%), the MG-ADL responder rate (9%), the QMG responder rate (10%), and the achievement of minimal manifestation status per MGFA-PIS (15%), which were similar in the 2 treatment groups but may increase concerns related to the interpretation of the findings that may not be generalizable to the full population.

Zilucoplan will likely be used as long as the patient is responding to treatment and any AEs are manageable, according to the clinical expert consulted for the review, so it would be reasonable for patients to continue treatment for at least the 12 weeks, as was evaluated in the RAISE-XT study. No concerns were raised by the expert based on the AEs observed with longer treatment with zilucoplan.

Indirect Comparisons

Description of Studies

The sponsor submitted an indirect comparison that compared the short-term efficacy of zilucoplan with that of other treatments used for the management of patients with anti-AChR antibody–positive gMG in the overall gMG population, as well as those with refractory gMG (subgroup analysis).¹⁵ The indirect comparison was based on a systematic literature review, and used Bayesian network meta-analysis (NMA) methods to estimate the comparative efficacy in the proportion of patients who met the MG-ADL response criteria at the end of the primary studies, and the change from baseline in the MG-ADL score at 12 weeks (± 2 weeks). The treatments included in the NMAs were zilucoplan, eculizumab, efgartigimod alfa, IVIg, PLEX, rituximab, ravulizumab, and rozanolixizumab.

A total of 12 double-blind, placebo-controlled RCTs were included in the NMA, but the primary analyses were based on phase III studies only (5 RCTs).

Efficacy Results

In the overall gMG population, the proportion of patients with a 3-point or greater (or 2-point or greater) improvement in MG-ADL score at the end of the studies (week 6 to week 26) was analyzed based on data from 5 RCTs (793 patients). The odds ratio (OR) for the proportion of responders was 1.37 (95% credible interval [CrI], 0.51 to 3.67) for zilucoplan versus eculizumab and 0.65 (95% CrI, 0.24 to 1.79) for zilucoplan versus efgartigimod alfa.

The subgroup analysis of patients with refractory gMG reported ORs of 2.00 (95% CrI, 0.63 to 6.63) and 0.95 (95% CrI, 0.29 to 3.21) for zilucoplan versus eculizumab and zilucoplan versus efgartigimod alfa, respectively. Of note, only 2 of 5 studies in this sensitivity analysis were exclusively patients with refractory gMG. The other studies included either a mixed population (in 1 study, 63% of patients had refractory disease) or an unknown proportion of patients with refractory disease (2 studies).

The primary analysis for the change from baseline in MG-ADL score included 5 studies with 755 patients and was based on outcomes reported at 10 weeks (2 studies) or 12 weeks (3 studies). The mean difference in the change from baseline in the MG-ADL score was −0.38 points (95% CrI, −2.21 to 1.41 points) for zilucoplan versus eculizumab and −1.50 points (95% CrI, −3.58 to 0.59 points) for zilucoplan versus efgartigimod alfa. For the refractory subgroup, the mean difference in the change from baseline in the MG-ADL score was −1.39 points (95% CrI, −3.49 to 0.68 points) for zilucoplan versus eculizumab and −2.51 points (95% CrI, −4.83 to −0.18 points) for zilucoplan versus efgartigimod alfa.

The sensitivity analysis that included 6 to 12 phase II and III studies (depending on the analysis) showed similar results for zilucoplan versus eculizumab and zilucoplan versus efgartigimod alfa in the overall and refractory populations, compared with the primary analyses in these populations.

Harms Results

No safety outcomes were included in the NMA.

Critical Appraisal

The indirect treatment comparison (ITC) report provided insufficient detail on the methods used to select studies for inclusion, so it is unclear if all potentially relevant studies were considered. In addition, the ITC report did not describe the findings from the feasibility assessment that was used to inform the conduct of the NMA. No information was provided on the characteristics of the studies included in the analyses, which was a major limitation.

Based on the data available, several important sources of heterogeneity were identified, including differences in outcome definitions and the timing of assessments, the placebo MG-ADL response rate, and the terms of disease severity and MG treatment history of the patients enrolled. Specifically, the trials had different proportions of patients rated as having mild, moderate, and severe and/or refractory gMG. The clinical expert stated that patients with refractory gMG or more severe disease may be less likely to respond to therapy, so differences in the distribution of these patients across trials may bias the findings. Both the overall population analyses and the refractory subgroup analyses contain a varied mix of patients with and without refractory gMG, which is a key source of heterogeneity. Moreover, the refractory subgroup analyses cannot be considered a true comparative assessment of patients with refractory gMG because not all studies were limited to patients with refractory disease.

A major limitation of the ITC methods was related to the timing of outcomes. The responder analyses used each study’s primary outcome time point, which varied from 6 weeks to 26 weeks. The change from baseline analyses assessed outcomes at week 10 or week 12 in the primary analyses but included other time points in the sensitivity analyses (from 4 weeks to 52 weeks). The CDA-AMC reviewer considered the differences in the timing of outcomes to be a significant source of heterogeneity that was not controlled for in the analyses. Restricting the outcome to those reported at 10 weeks or 12 weeks also had serious limitations, related in part to the differences in dosing schedules (intermittent versus continuous), which could bias the results. The response definition also varied, with 4 of the 5 studies in the primary analysis reporting the proportion of patients with at least a 3-point improvement in the MG-ADL score, and 1 of the key studies using a 2-point threshold.

Overall, the evidence networks were sparse, with the primary analysis based on 1 trial per comparator. Generally, the duration of follow-up in the source studies was limited (up to 26 weeks for the primary analyses), so comparative estimates were based on short-term data. The NMA results lacked precision, as shown by the wide 95% CrI. Considering the heterogeneity in the patient and study characteristics that was identified after a limited assessment, there is likely substantial risk of bias for the comparisons in the network. As a result, no conclusions could be drawn on the comparative efficacy of zilucoplan. No harms outcomes were assessed in the NMA, so the comparative safety of zilucoplan is unknown.

Studies Addressing Gaps in the Evidence From the Systematic Review

No additional studies addressing important gaps in the systematic review were identified.

Conclusions

Patients and clinicians identified a need for new treatments for gMG that reduce gMG-related and treatment-related morbidity and mortality, decrease the number and intensity of exacerbations, provide durable benefit, improve quality of life, and decrease the risk of side effects. Evidence from the randomized, phase III, double-blind RAISE trial of 174 adult patients with anti-AChR antibody–positive gMG (including 88 with refractory anti-AChR antibody–positive gMG) demonstrated that treatment with zilucoplan is likely superior to placebo in providing improvement in signs and symptoms of disease activity at week 12, as measured by the MG-ADL score, the primary outcome, in the overall trial population, which included patients with refractory and nonrefractory disease. Zilucoplan also likely results in a clinically meaningful benefit in the key secondary outcomes of QMC score and MGC score at week 12, compared with placebo in the overall trial population. Zilucoplan likely results in improvements in HRQoL compared to placebo, based on results from the MG-QoL 15r instrument. For other secondary outcomes (i.e., at least a 3-point response in MG-ADL score and at least a 5-point response in QMG score), zilucoplan is also likely to show improvements at week 12 compared with placebo in the overall trial population. MSE score at week 12 favoured zilucoplan numerically; although it did not reach statistical significance in the overall trial population, it may result in a benefit compared to placebo. Zilucoplan may result in an increase in the number of patients with minimal manifestation status per MGFA-PIS at week 12, an exploratory outcome, compared to placebo. Note that confidence in the between-group differences for efficacy outcomes in the overall trial population were limited because of imprecision (indicated by the associated CIs that included small effects close to the null or crossed the null) and relatively small sample sizes.

The prespecified subgroup of patients with refractory disease (50.6%) aligned with the sponsor’s reimbursement request under review for zilucoplan. Results of the subgroup analyses were subject to limitations, such as a reduced sample size and a lack of formal statistical approaches to control for an inflated type I error rate in multiple comparisons. However, the clinical benefit observed in the refractory subgroup was in all cases consistent with or tended to be slightly larger than the results shown in the overall trial population and was considered to be clinically meaningful by the clinical expert consulted for this review. Results for minimal manifestation status per MGFA-PIS at week 12 were not reported for the refractory subpopulation. The safety profile of zilucoplan in the refractory subgroup was consistent with that reported in the overall trial population, with no unexpected safety signals, and was considered manageable by the clinical expert consulted for this review. The evidence is very uncertain about the effect of zilucoplan on the number of infections experienced by patients, compared to placebo, at week 12 (very low certainty) in the overall trial population.

One open-label extension study of patients with gMG, including patients with refractory gMG, demonstrated findings supportive to those in the RAISE trial, which involved an additional 84 weeks of treatment with zilucoplan; however, the open-label nature of the study design limits the ability to distinguish between the effects of treatment and the natural history of the disease and between study treatments, including concomitant therapies, and other unknown factors.

A sponsor-submitted NMA comparing the efficacy of zilucoplan with other treatments for gMG, including eculizumab, efgartigimod alfa, IVIg, PLEX, and rituximab, had several limitations (sparse evidence networks, short-term data, imprecision of estimates, and heterogeneity in patient and study characteristics), which preclude any conclusions from being drawn on the comparative efficacy of zilucoplan. Because harms outcomes were not assessed in the NMA, the comparative safety of zilucoplan is unknown.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of zilucoplan 40 mg/mL SC injection in the treatment of adult patients with anti-AChR antibody–positive MG.

Disease Background

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the CDA-AMC review team.

MG is a rare, chronic, autoimmune, neuromuscular disease in which antibodies against the neuromuscular junction disrupt nerve impulse conduction, resulting in localized or generalized skeletal muscle weakness.^2,3 In most patients, MG initially affects the extraocular muscles (ocular MG) and then progresses to other muscle groups, including the bulbar and proximal limb skeletal muscles. When the disease progresses to other muscle groups, it is referred to as gMG.^4-6 Approximately 80% of all patients with MG have gMG.⁶ The majority of patients with gMG (80% to 90%) have antibodies against anti-AChR; however, antibodies against MuSK and LRP4 are also present in some patients with MG.^2,6,7,16 A small number of patients do not have any measurable antibodies and are seronegative.^2,6

The MGFA classification system groups patients with MG into 5 functional classes, based on the severity and localization of symptoms.⁹ The classification ranges from class I (ocular manifestations only) to class V (intubation or myasthenic crisis).⁹ Class II, class III, and class IV represent patients with mild MG, moderate MG, and severe gMG, respectively. Other validated scales used to determine disease and symptom severity in patients with MG include the MG-ADL, the QMG, the MGC, and the MG-QoL15.¹⁰

gMG is a disease of fluctuating muscle weakness that often gets worse after exercise or later in the day.² In patients with MG, involuntary or smooth muscles are not affected.¹¹ The symptoms and disease course of gMG are highly variable and heterogenous among patients. It has been referred to as a snowflake disease, as it affects each patient differently.¹¹ Patients experience a variety of symptoms, including fatigue, droopy eyelids, diplopia, neck weakness, difficulty swallowing or chewing, speech disturbances, difficulty breathing, and upper and/or lower limb weakness.¹¹ The symptoms of gMG occur unpredictably and fluctuate in nature, intensity, and severity on a day-to-day basis and throughout a patient’s life.² Patients can also experience exacerbation, which by definition comprises 3 criteria: temporal (a rapid, progressive, and unpredictable worsening of symptoms); subjective (deterioration of bulbo-pharyngeal function, restriction of strength for head and neck muscles or extremity muscles that affects a patient’s everyday life or the start of weakness in breathing with reduced cough impulse); and objective (at least a 5-point change in QMG score with no more than 5 points from ocular MG). Exacerbation requires inpatient treatment.¹² Furthermore, MG exacerbations can deteriorate into a myasthenic crisis, in which patients experience sudden respiratory failure that requires emergency intubation or ventilation.¹² In the literature, it is noted that patients with gMG experience a significant disease burden as a result of the disease symptoms and the side effects of treatments that reduce their quality of life.¹⁷

In Canada, the incidence and prevalence of gMG are estimated at 23 per 1 million person-years and 32 per 100,000, respectively.⁸ The mortality rate of MG has been reported to be between 0.06 and 0.89 per million person-years.¹⁸ In 2013, there were 3,611 prevalent cases in Ontario, and the crude prevalence rate was 32.0 per 100,000 population. In 2021, the incidence of MG in the US estimated to be 3.2 per 100,000 and the total prevalence was estimated to be 37.0 per 100,000, using population estimates from the US Census. Almost 15% of patients with gMG have been classified as having refractory gMG.¹³ The diagnosis of gMG is based on clinical presentation; serological tests that detect antibodies against anti-AChR, MuSK, and LRP4; and electrodiagnostic tests for the evaluation of neuromuscular transmission.^6,7

Standards of Therapy

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the CDA-AMC review team.

Four classes of therapies are currently available to treat MG: symptomatic treatment (i.e., acetylcholinesterase inhibition), chronic immunotherapies (i.e., glucocorticoids, nonsteroidal immunosuppressive, and immunomodulatory drugs), rapid but short-acting immunomodulating treatments (i.e., IVIg and therapeutic PLEX), and surgical treatment (i.e., thymectomy).¹⁹ The choice of therapies is largely dependent on the time of onset of clinical effect, in addition to the disease pace and severity.¹⁹

Initial symptomatic treatment for most patients with mild or moderate MG is an oral acetylcholinesterase inhibitor (usually pyridostigmine), which is aimed at improving clinical symptoms and neurologic deficits over the course of treatment.¹⁹ Glucocorticoids (e.g., oral prednisone) are commonly added to pyridostigmine initially.²⁰ Patients who do not respond to or remain significantly symptomatic after pyridostigmine combined with prednisone are usually offered NSISTs (e.g., azathioprine, mycophenolate mofetil).^19,20 Oral methotrexate may be considered as a steroid-sparing drug in patients with gMG who experience a poor response or intolerance to other steroid-sparing drugs.²¹ Bridge therapy for patients who are intolerant to glucocorticoids (e.g., those with poorly controlled diabetes) include IVIg, PLEX, complement inhibitors (e.g., eculizumab, ravulizumab), neonatal FcR inhibitors (efgartigimod alfa), and rozanolixizumab.²⁰ Specifically, biologics (efgartigimod alfa, rozanolixizumab, ravulizumab, and zilucoplan) are used as chronic immunotherapy for patients with anti-AChR antibody–positive gMG.¹⁹

Patients with MG who experience a worsening of myasthenic symptoms are offered treatment individualized to the nature of their exacerbations, including the underlying cause (e.g., concurrent infection, surgery, pregnancy, medications, spontaneous), severity (e.g., life-threatening respiratory failure, such as myasthenic crisis), presence of functional limitations (e.g., dysphagia, dyspnea), rate of onset of potential treatments (e.g., hours for pyridostigmine, days for IVIg or PLEX, and weeks for immunotherapy or glucocorticoids), and treatment setting (e.g., outpatient, inpatient, intensive care unit).¹⁹ Therapeutic PLEX (plasmapheresis) and IVIg have a rapid onset (days) of action and are used in acute exacerbations, preoperatively (e.g., before thymectomy), or as a bridge to slower-acting immunotherapies (particularly among patients avoiding or minimizing glucocorticoids).¹⁹ In addition to its role in exacerbations of MG, IVIg is used as an alternative to plasmapheresis or several immunotherapy drugs in patients with refractory MG.¹⁹

Thymectomy (i.e., surgical section of the thymus) should be considered for adult patients with anti-AChR antibody–positive gMG to minimize the need for immunotherapy or as an alternative to immunotherapy (among patients who experience response failure to an initial adequate trial of immunotherapy or intolerable side effects from immunotherapy), and to reduce the risk of hospitalization for disease exacerbations.²¹ Thymectomy may be used in parallel with pyridostigmine and immunotherapy in select patients, including patients with thymoma (approximately 10% to 15% of patients with MG), and without thymoma (in patients with anti-AChR antibody–positive gMG who are younger than 50 years).¹⁹ Treatment decisions for thymectomy are age-stratified for older adults with nonthymomatous anti-AChR antibody–positive gMG; patients aged 51 to 65 years may be offered thymectomy on the basis of symptom severity and operative risks, but thymectomy is not indicated for patients older than 65 years.^21,22 According to the clinical expert consulted for this review, thymectomy, although not curative, can produce a long-lasting benefit or disease remission in eligible patients with MG.

Patients with refractory disease are limited by the toxicities of immunosuppressive and immunomodulatory drugs and may require glucocorticoids in high doses, IVIg, rituximab, C5 complement inhibitor (e.g., eculizumab, ravulizumab), and cyclophosphamide.^19,20 Treatment for patients with refractory disease lacks evidence-based guidelines; overall, treatment is more aggressive and highly individualized.²⁰ The International Consensus Guidelines recommend rituximab as an option for patients with refractory anti-AChR antibody–positive MG who experience response failure or intolerable side effects with ISTs.^21,23 Eculizumab should be considered for patients with severe, refractory anti-AChR antibody–positive gMG; its duration of treatment and role in the treatment of other MG populations and stages of disease may evolve with future research.^21,24

Drug Under Review

The key characteristics of zilucoplan are summarized in Table 3, along with other treatments available for adult patients with anti-AChR antibody–positive gMG.

Zilucoplan is an immunosuppressant and C5 inhibitor that inhibits the effects of C5 through a dual mechanism of action. It specifically binds to C5, thereby inhibiting its cleavage by the C5 convertase to C5a and C5b, which results in a downregulation of the assembly and cytolytic activity of the membrane attack complex. Additionally, by binding to the C5b moiety of C5, zilucoplan sterically hinders the binding of C5b to C6, which prevents the subsequent assembly and activity of the membrane attack complex, should any C5b be formed.

The recommended dose of zilucoplan for adult patients with gMG is 0.3 mg/kg as SC injection once daily, administered about the same time every day, according to the patient’s body weight (i.e., a 16.6 mg total daily dose for patients with a body weight of < 56 kg, a 23.0 mg total daily dose for those with a body weight of ≥ 56 to < 77 kg, and a 32.4 mg total daily dose for patients with a body weight of ≥ 77 kg). Zilucoplan received Notice of Compliance from Health Canada on July 11, 2024, “for the treatment of generalized myasthenia gravis (gMG) in adult patients who are anti-acetylcholine receptor (AChR) antibody positive. Patients continued to receive standard therapy throughout the pivotal trial.” Zilucoplan has not been previously reviewed by CDA-AMC.

The reimbursement request is “as an add-on therapy for the treatment of adult patients with AChR antibody positive refractory gMG, defined as not achieving symptom control after:

• Treatment for at least 1 year with 2 or more of the following therapies: prednisone, azathioprine, mycophenolate cyclosporine, cyclophosphamide, methotrexate, tacrolimus, other corticosteroids for gMG, other immunosuppressive therapies (ISTs), OR

• History of treatment with at least one of these therapies for 1 year or more and required chronic plasma exchange (PLEX), intravenous immunoglobulin (IVIg), or subcutaneous immunoglobulin (SCIg) at least every 3 months for the 12 months prior to treatment with zilucoplan.” Per sponsor request, this review focuses on the indication for patients with refractory gMG.

Zilucoplan has FDA approval “for the treatment of generalized myasthenia gravis in adult patients who are anti-acetylcholine receptor (AChR) antibody positive” and is authorized by the European Medicines Agency “as an add-on to standard therapy for the treatment of generalised myasthenia gravis (gMG) in adult patients who are anti-acetylcholine receptor (AChR) antibody positive.”

Table 3: Key Characteristics of Zilucoplan, Efgartigimod Alfa, Eculizumab, Immunoglobulins, PLEX, and Rituximab

AChR = acetylcholine receptor, FcR = fragment crystallizable receptor; gMG = generalized myasthenia gravis; Ig = immunoglobulin; IgG1 = immunoglobulin G1; PLEX = plasma exchange; SC = subcutaneous.

^aHealth Canada indication, as applicable.

Source: Product monograph for zilucoplan.¹

Perspectives of Patients, Clinicians, and Drug Programs

Patient Group Input

This section was prepared by the CDA-AMC review team based on the input provided by patient groups. The full original patient group input received by CDA-AMC has been included in this section of the report.

CDA-AMC received 1 patient group submission from the MDC. MDC is a health charity that supports people affected by muscular dystrophies and related muscle diseases (neuromuscular disorders) in Canada. MDC’s mission is to enhance the lives of those impacted by neuromuscular disorders by continually working to provide ongoing support and resources while relentlessly searching for a cure through well-funded research.

MDC identified and contacted adults living with MG to participate in a survey and semi-structured virtual interviews. Surveys were shared with members through e-blasts, personalized invitations, and online patient groups. MDC also conducted an MG Journey Mapping project for adults living with MG in Canada, using virtual interviews, round table sessions, surveys, and HRQoL measures (EQ visual analogue scale, EQ-5D, MG-ADL, MG-QoL). MDC’s patient group submission to CDA-AMC included information from 127 individuals (84 females and 43 males aged 22 years to 78 years from all provinces in Canada) the majority of whom reported having gMG. Additionally, 47 people living in Canada (33 females, 14 males) with MG provided input on their hopes and expectations for zilucoplan and their everyday experiences with MG. None of the respondents included in the MDC’s patient input had experience with zilucoplan.

Respondents indicated that MG has a significant impact on productivity, fatigue and energy levels, quality of sleep, respiratory health, mobility, strength, independence, relationships and social participation, eyes and vision, speech, and swallowing. Respondents also noted MG impacts beyond physical symptoms, including mental health, quality of life, and the well-being of their families.

Respondents reported that although available treatments for MG had a positive impact on health outcomes, overall, they had negative experiences with steroids, disliked the slow onset of medication effects, and had a feeling of trial and error with medications. Patients also had concerns about the long-term and sustained benefits of supportive treatments.

Patients with MG are looking for improved outcomes with new treatments, including a decreased intensity of exacerbations and side effects, the maintenance of independence, and fewer hospital admissions for serious MG disease-related circumstances. Patients are willing to tolerate side effects of medications if they lead to improved MG outcomes. In addition, respondents stated that although current medications appear to decrease the number of exacerbations, they do not have an impact on patients’ overall quality of life. Moreover, MDC input noted that patients, families, and caregivers consider the following factors when evaluating MG therapies: treatment method and delivery (e.g., invasiveness, duration, frequency of administration), potential side effects (e.g., low risk of side effects, number of side effects), HRQoL, convenience of treatment (e.g., administration at home or in a community centre, perceived control and flexibility, time to travel to clinic, access to parking for clinic visits), financial impact (e.g., treatment coverage by public or private insurance), and access to treatment. According to the MDC input, HRQoL outweighs the convenience of a drug as a key priority. MDC input stated that in addition to accessibility to treatments, patients with MG need improved treatment options that enhance effectiveness and tolerance over the long-term.

Patients and caregivers reported varied experiences in diagnostic testing for MG. Although some respondents experienced minimal problems with testing and diagnosis, especially in cases of MG crisis or hospitalization due to MG, the majority reported significant difficulties getting a diagnosis, which included a lengthy process with many missed opportunities, delayed diagnosis, misdiagnosis (such as stroke or Bell palsy), and costs incurred. According to early findings of the MG Journey Mapping project, time from the first bothersome symptom to diagnosis ranged from 7 years to 23 years. All respondents underwent diagnostic blood testing, and many underwent single-fibre electromyography to confirm diagnosis.

Clinician Input

Input From the Clinical Expert Consulted for This Review

All CDA-AMC review teams include at least 1 clinical specialist with expertise in the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 1 clinical specialist with expertise in the diagnosis and management of MG.

Unmet Needs

The clinical expert explained that although there has been significant improvement in the management of patients with MG over the past few decades, there remain significant treatment gaps for patients with gMG. Given that patients with gMG experience disease morbidity while awaiting the beneficial effects of prolonged treatments that often have intolerable side effects, the expert highlighted the need for new treatments that have a faster onset of action (particularly for patients with any bulbar or respiratory involvement), are more effective, reduce significant AEs, and have better durability than current treatment options. Additionally, the clinical expert noted that patients would benefit from new treatments that are less invasive (e.g., administered orally rather than SC or IV route of administration), are administered less frequently (e.g., weekly or monthly rather than daily), allow for reduced doses of other immunosuppressive drugs, and may be used in combination with existing or future treatments that have different mechanisms of action (e.g., a combination of a peptide complement inhibitor plus a neonatal FcR inhibitor for patients with severe or refractory MG). Overall, the expert outlined that goals of treatment for patients with MG, which are to reduce MG-related morbidity and mortality, minimize treatment-associated morbidity and mortality, reduce weakness (ocular, bulbar, respiratory, axial, and extremity) associated with MG and prevent disease exacerbations or an MG crisis, and improve HRQoL.

Place in Therapy

The expert noted that, depending on how refractory disease is defined, approximately 10% to 15% of patients do not respond to conventional treatment and, therefore, require more aggressive treatments, including IVIg or PLEX; these patients are considered to be the target population of complement inhibitors such as zilucoplan. The expert noted that the patient population identified as refractory in the RAISE trial is aligned with the way refractory disease is defined in clinical practice, particularly with reference to an adequate trial of prednisone in addition to another IST.

Patient Population

Patients are initially identified as having MG through clinical suspicion, and diagnosis is confirmed with electrophysiology (i.e., repetitive nerve stimulation for assessment of decrements) and serology (e.g., confirmation of antibodies for anti-AChR, MuSK, and LRP4), according to the clinical expert, who added that the presence of anti-AChR antibodies is a reliable diagnostic finding with high specificity, so the identification of patients with MG using anti-AChR antibody testing is straightforward. Nevertheless, given the rarity of disease and the fact that the initial identification of MG is based on clinical suspicion, the clinical expert noted that MG is likely underdiagnosed in the population. The expert also noted that the availability and timeliness of anti-AChR antibody assays varies across Canada.

According to the expert consulted, the patients who are most likely to benefit from treatment with zilucoplan are generally aligned with patients who were enrolled in the RAISE trial, who had anti-AChR antibody–positive gMG, MGFA disease class II to class IV, and an MG-ADL score of 6 or greater; the exception was the QMG score, which was 12 or greater in study participants but is reportedly not used commonly in clinical practice. The expert pointed out that patients who have had an adequate trial (in both dose and duration) of prednisone and at least 1 NSIST or who demonstrate intolerance to the combination of prednisone and NSISTs should be eligible for treatment with zilucoplan, but these criteria limit eligibility to patients who have been on conventional treatment for at least 6 months (and commonly for 12 months), reflecting the duration needed for corticosteroids (3 to 6 months) and mycophenolate mofetil and azathioprine (likely 12 to 18 months) to produce an optimal benefit.

Assessing the Response Treatment

The MG-ADL scale is a patient-reported outcome deemed by the expert to be clinically relevant (a 2-point change is considered the minimum clinically meaningful improvement) to the evaluation of response to treatment. According to the clinical expert consulted for this review, the MG-ADL can be self-administered in minutes (with the supervision of a neuromuscular neurologist) and should be used as an eligibility criterion for treatment with zilucoplan and to monitor efficacy throughout treatment.

Discontinuing Treatment

A clear lack of response (i.e., no reduction in MG-ADL score after about 6 months) to treatment with zilucoplan, intolerance due to significant AEs that require the discontinuation of treatment, and the requirement for additional ongoing treatments with IVIg or PLEX, despite an adequate trial of zilucoplan, were reasons to discontinue treatment with zilucoplan, according to the clinical expert.

Prescribing Considerations

The clinical expert said that because MG is a rare disorder that requires nuanced management, patients with gMG should be diagnosed, treated, and monitored by a neuromuscular neurologist with experience in gMG.

Clinician Group Input

This section was prepared by the CDA-AMC review team based on the input provided by clinician groups. The full original clinician group input received by CDA-AMC have been included in the Perspectives of Patients, Clinicians, and Drug Programs section of this report.

CDA-AMC received 1 clinician group submission from the NMD4C, which comprised the input of 8 clinicians with experience treating patients with gMG. The clinician group agreed with the clinical expert on the unmet treatment needs, goals of treatment, treatment response evaluations, and care management for patients with MG. NMD4C identified the additional treatment outcomes of emergency department visits, hospitalizations, and intensive care unit admissions. The ability to self-administer zilucoplan at home provides patients with greater autonomy in their care management, according to the clinician group input. Both the clinician group and the clinical expert consulted for this review indicated that patients with anti-AChR antibody–positive gMG would most likely benefit from treatment from zilucoplan, although the clinician group noted that evidence on the efficacy of zilucoplan has not been confirmed for patients with MG who are seronegative.

Drug Program Input

The drug programs provide input on each drug being reviewed through the CDA-AMC reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical expert consulted by CDA-AMC are summarized in Table 4.

Table 4: Summary of Drug Plan Input and Clinical Expert Response

AChE = acetylcholinesterase; AChR = acetylcholine receptor; CDEC = Canadian Drug Expert Committee; FcR = fragment crystallizable receptor; gMG = generalized myasthenia gravis; Ig = immunoglobulin; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGFA = Myasthenia Gravis Foundation of America; NSIST = nonsteroidal immunosuppressive therapy; pCPA = pan-Canadian Pharmaceutical Alliance; PLEX = plasma exchange; SC = subcutaneous.

Clinical Evidence

The objective of the CDA-AMC Clinical Review report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of zilucoplan, administered as an SC injection, once daily, according to body weight (16.6 mg for a body weight of < 56 kg, 23 mg for a body weight of ≥ 56 kg to < 77 kg, and 32.4 mg for a body weight of ≥ 77 kg), for the treatment of adults with anti-AChR antibody–positive gMG. Patients continued to receive standard therapy throughout the pivotal RAISE trial. The focus will be placed on comparing zilucoplan to relevant comparators and identifying gaps in the current evidence. Zilucoplan received a Notice of Compliance from Health Canada on July 11, 2024. The reimbursement request is for the refractory population (adults with refractory gMG), based on the sponsor’s consultation with clinical experts on the current treatment paradigm, patients’ unmet needs, and the anticipated use of zilucoplan for gMG in Canada.²⁵

A summary of the clinical evidence included by the sponsor in the review of zilucoplan is presented in 3 sections, with the CDA-AMC critical appraisal of the evidence included at the end of each section. The systematic review includes pivotal studies and RCTs that were selected in accordance with the sponsor’s systematic review protocol. The CDA-AMC assessment of the certainty of the evidence in this first section, which uses the GRADE approach, follows the critical appraisal of the evidence. The second section includes sponsor-submitted long-term extension studies. The third section includes indirect evidence from the sponsor. The sponsor did not include any additional studies that addressed important gaps in the systematic review evidence.

Included Studies

Clinical evidence from the following are included in the CDA-AMC review and appraised in this document:

• 1 pivotal, phase III, randomized trial identified in the Systematic Review section

• 1 long-term extension study

• 1 ITC.

Systematic Review

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the CDA-AMC review team.

Description of Study

Characteristics of the included study are summarized in Table 5.

Table 5: Details of the Study Included in the Systematic Review

AChR = acetylcholine receptor; gMG = generalized myasthenia gravis; Ig = immunoglobulin; IMP = investigational medicinal product; IST = immunosuppressive therapy; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA = Myasthenia Gravis Foundation of America; MGFA-PIS = Myasthenia Gravis Foundation of America Post-Intervention Status; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item revised; MSE = minimal symptom expression; Neuro-QoL = Quality of Life in Neurologic Disorders; PLEX = plasma exchange; QMG = quantitative myasthenia gravis; SC = subcutaneous; TEAE = treatment-emergent adverse events; VAS = visual analogue scale; WPAI:SHP = Work Productivity and Activity Impairment Questionnaire: Specific Health Problem.

Sources: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

RAISE was a multicentre, randomized, phase III, double-blind, placebo-controlled study designed to evaluate the efficacy and safety of zilucoplan in patients with gMG (overall trial population) and in a prespecified subgroup of patients with refractory gMG (refractory subpopulation). After completion of the 12-week treatment period in the RAISE study, eligible patients had the option to enrol in the RAISE-XT long-term extension study. Details of the RAISE-XT study is summarized in the Long-Term Extension Studies section. The overall study designs of the RAISE and RAISE-XT studies are depicted in Figure 1.

In the RAISE study, 174 patients with anti-AChR antibody–positive gMG (88 of whom made up the refractory subpopulation) were enrolled from September 2019 to December 2021 at 75 sites in 10 countries, with 2 sites (3 patients) in Canada. Eligible patients were randomized in a 1:1 ratio to receive zilucoplan 0.3 mg/kg once daily plus standard of care or to receive placebo plus standard of care. Randomization was stratified by baseline MG-ADL score (≤ 9 versus ≥ 10), QMG score (≤ 17 versus ≥ 18), and geographic region (East Asia, Europe, and North America). The primary objective of the RAISE study was to evaluate the change from baseline to week 12 in the MG-ADL score. Key secondary objectives included change from baseline to week 12 in the QMG score, MGC score, and MG-QoL15r score, and the frequency of TEAEs. Additional secondary objectives included the achievement of MSE (defined as an MG-ADL score of 0 or 1) at week 12 without rescue therapy, an MG-ADL responder rate (defined as a ≥ 3-point decrease in MG-ADL score) at week 12 without rescue therapy, and a QMG responder rate (defined as a ≥ 5-point decrease in QMG score) at week 12 without rescue therapy. Achievement of minimal manifestation status per MGFA-PIS at week 12 without rescue therapy was an exploratory objective. Data for the primary and key secondary outcomes were available for both the overall population and the refractory subpopulation. Data for the additional secondary outcomes were available for observed cases in both the overall population and the refractory subpopulation, and for imputed cases in the overall population. Data on the achievement of minimal manifestation status per MGFA-PIS at week 12 without rescue therapy were available for the overall population only. The Clinical Study Report, which had a data cut-off date of December 30, 2021, was the primary source for the RAISE study.

Figure 1: Study Design of the Pivotal Study (RAISE) and the Extension Study (RAISE-XT)

SC = subcutaneous.

Source: RAISE Clinical Study Report.¹⁴

Populations

Inclusion and Exclusion Criteria

The key inclusion criteria were adults aged 18 years to 74 years, with a diagnosis of gMG (MGFA disease class II to class IV at screening), and positive serology for anti-AChR-binding autoantibodies. At screening and baseline, eligible patients had to have an MG-ADL score of 6 or greater, a QMG score of 12 or greater, and a QMG score of 2 or greater on 4 or more items. Key exclusion criteria were known disease characteristics (positive serology for muscle-specific kinase, investigator-determined minimal manifestation status of gMG, investigator-determined fixed weakness or burned out gMG), medical history (abnormal thyroid function, history of meningococcal disease, current or recent systemic infection, active malignancy), and prior treatment (thymectomy; recent surgery requiring general anesthesia; and/or prior treatment with a complement inhibitor, rituximab, IVIg, SCIg, PLEX, or an experimental drug). Patients were considered to have refractory gMG if they had received treatment for at least 1 year with 2 or more of the following therapies: prednisone, azathioprine, mycophenolate mofetil, cyclosporine, cyclophosphamide, methotrexate, tacrolimus, rituximab, eculizumab, other corticosteroids for gMG, or other ISTs; or had a history of treatment with at least 1 of the aforementioned therapies for 1 year or more and required chronic PLEX, IVIg, or SCIg at least every 3 months for the 12 months before enrolment.

Interventions

Patients were randomized to receive zilucoplan 0.3 mg/kg or placebo administered by SC injection. Patients must have received a quadrivalent or serotype B meningococcal vaccine, or both, at least 14 days before the first dose of the study drug. At the day 1 visit, doses were administered by the patient into the abdomen (preferred site), thigh, or upper arm, under the supervision of site staff. After in-clinic education and training to ensure that the administration technique for self-injection was adequate and appropriate, all patients self-injected daily SC doses of the blinded drug (zilucoplan or placebo), according to randomized treatment allocation, for the subsequent 12 weeks. Single-use prefilled syringes in injection devices were provided for use during the study, and weight-bracketed dosing was used (i.e., patients were provided with prefilled syringes that contained fixed amounts of zilucoplan based on their weight, and each fixed amount covered a range of study patient weights). This weight-bracketed dosing strategy had the potential for a range of doses to be received, from a minimum dose of zilucoplan 0.22 mg/kg per day to a maximum dose of zilucoplan 0.42 mg/kg per day (refer to Table 6). Patients who presented with a higher body weight (> 150 kg) or a lower body weight (< 43 kg) were accommodated on a case-by-case basis.

Table 6: Zilucoplan Dose Presentations by Weight Bracket

Source: RAISE Clinical Study Report.¹⁴

Study periods in the RAISE study included screening (up to 4 weeks) and treatment (12 weeks), for a total study duration of 16 weeks. During the treatment period, patients were evaluated during clinic visits at week 1, week 2, and monthly visits at week 4, week 8, and week 12.

Patients were permitted to receive prescriptions and over-the-counter medications in the 30 days before screening or as concomitant medications (ongoing or initiated after the first dose of the study drug on day 1) up to the last study visit. Patients were expected to remain on stable doses of standard-of-care treatments for gMG, including corticosteroids and IST drugs, unless medically indicated. Pyridostigmine was to be withheld for a minimum of 10 hours before clinical evaluation at every study visit. The pyridostigmine dose could be reduced if judged by the investigator to be related to intolerable side effects. Doses of standard-of-care treatments were not to be increased during the study; however, patients were permitted to receive rescue therapy (IVIg, PLEX, or eculizumab), with the type, frequency, and duration at the discretion of the investigator. Patients were prohibited from receiving the concomitant medications of rituximab, aminoglycosides, colomycin, polymyxin, telithromycin, injectable cyclin, macrolides, fluoroquinolones, quinines, quinidine, hydroxychloroquine, procainamide, beta-blockers, diphenylhydantoin, dantrolene, D-penicillamine, and magnesium.

Dosing of a study drug could be interrupted or permanently discontinued at the discretion of the investigator and/or the sponsor. Patients who permanently discontinued the study drug underwent a safety evaluation (ongoing AEs or new SAEs that developed after the last study visit) 40 days after the last dose of the study drug. Patients diagnosed with meningococcal disease were to discontinue the study drug permanently.

Outcomes

A list of efficacy end points assessed in this Clinical Review report is provided in Table 7, followed by descriptions of the outcome measures (Table 8). Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence, as well as any outcomes identified as important to this review by the clinical expert consulted by CDA-AMC and input from patient and clinician groups and public drug plans. The CDA-AMC review team selected end points that were considered to be most relevant to CDA-AMC expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. All summarized efficacy end points were assessed using GRADE. Select notable harms outcomes considered important to CDA-AMC expert committee deliberations were also assessed using GRADE.

Table 7: Outcomes Summarized From the Study Included in the Systematic Review

MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA-PIS = Myasthenia Gravis Foundation of America Post-Intervention Status; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item revised; MSE = minimal symptom expression; QMG = quantitative myasthenia gravis; TEAE = treatment-emergent adverse event.

^aStatistical testing for these end points was adjusted for multiple comparisons (e.g., hierarchical testing).

Source: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

Activities of Daily Living

MG-ADL Scale

The MG-ADL scale is a brief, 8-item, interviewer-administered, patient-reported outcome designed to evaluate MG symptom severity.²⁸ Two questions pertain to ocular functions, 3 to oropharyngeal functions, 1 to respiratory function, and 2 to extremity functions. The 8 items are each scored on a scale from 0 to 3, and the total score is the sum of the 8 individual scores, and ranges from 0 to 24. Higher scores indicate more severe symptoms of MG.²⁸ A 2-point change in MG-ADL score is considered clinically meaningful.²⁹ This 2-point change threshold was based on a receiver operator characteristic curve method used in an observational study of patients with MG managed at the discretion of the treating physician.²⁹ In the RAISE study, the MG-ADL assessment was conducted at screening to assess patient eligibility and at each study visit.

Disease Severity

QMG Scale

The QMG scale is a standardized and validated quantitative strength scoring system that was developed specifically for MG.³⁰ The scoring system consists of 13 individual assessments; each scored on a scale from 0 to 3. The total score is the sum of the individual scores, and ranges from 0 to 39. Higher scores represent more severe impairment.³⁰ A change in the QMG score of 3 points or more may be considered clinically meaningful in a typical clinical study population of patients with MG.^30,31

Treatment Response

MGC Scale

MGC is a 10-item scale that has been used to measure the clinical status of patients with MG, both in the practice setting and in clinical studies, and evaluate treatment response. The 10 items and corresponding response scale scores are weighted and totalled. The total score is the sum of the 10 individual scores, and ranges from 0 to 50. Higher scores indicate more severe impairment due to the disease.^32,33 A study of patients with MG who are under routine care showed that a 3-point change in this assessment is considered clinically meaningful and is associated with the best sensitivity and specificity, based on a receiver operator characteristic curve method.³⁴

Additional Efficacy End Points

Minimal Symptom Expression

MSE was designed to assess how many patients become free or virtually free of MG symptoms and is achieved when the MG-ADL total score is 0 or 1 for patients receiving therapy.³⁵

Health-Related Quality of Life

MG-QoL15r Scale

MG-QoL15r is a 15-item, self-administered, patient-reported outcome scale designed to assess quality of life in patients with MG.³⁶ The 15 items relate to the physical, social, and psychological aspects of well-being and are associated with corresponding response scales, each scored from 0 to 2. The total score is the sum of the 15 individual item scores, and ranges from 0 to 30. Higher scores indicate a more severe impact of the disease on aspects of the patient’s life.^36,37 An MID has not been established for the MG-QoL15r in patients with MG.⁹

Safety

Safety outcomes included TEAEs of interest, clinical laboratory tests, vital signs, electrocardiograms, physical examinations, the Columbia-Suicide Severity Rating Scale, and immunogenicity. AEs were defined as any untoward medical occurrence in a patient who was administered a study drug, regardless of whether it was related to the study drug. AEs were classified according to the national Cancer Common Terminology Criteria for Adverse Events (CTCAE) and coded using the Medical Dictionary for Regulatory Activities, version 22.0. SAEs were defined as any AE that resulted in death, was life-threatening, required hospitalization or prolongation of existing hospitalization, resulted in persistent or significant disability or incapacity, resulted in a congenital anomaly or birth defect, or required intervention to prevent any of the aforementioned outcomes. Although AEs of interest were not prespecified in the protocol, an amendment of the Statistical Analysis Plan (February 1, 2021, amendment 1.0) included statistical analyses specific to infections (such as Neisseria infections, injection-site reactions, and hypersensitivity reactions [e.g., anaphylactic reactions, hepatic events, and malignancies]). The AE reporting period started on day 1, with the first administration of the study drug, and continued until the last study visit; patients were assessed 40 days after their last dose of the study drug for any ongoing AEs or any new SAEs that developed after the last study visit. A safety monitoring committee evaluated study safety data, as appropriate, for any unblinding of data for review.

Based on input received, changes from baseline in the MG-ADL, QMG, MGC, and MG-QoL15r scores were identified as important outcomes to patients and to clinicians making treatment decisions. The expert also identified the achievement of MSE, responses on the MG-ADL and the QMG, and the achievement of minimal manifestation status per MGFA-PIS as important outcome measures. Time to rescue therapy was not identified by clinicians as being very important for clinical decision-making. The need for rescue therapy was noted by the clinical expert to be a relatively uncommon event that may be indicative of treatment failure, but patients expressed a desire for new treatments that provide sustained benefit with reduced exacerbations. Findings for time to rescue therapy are summarized in Appendix 1. Among patients with refractory gMG, mean change in the MG-ADL score and the proportion of MG-ADL responders were key inputs in the sponsor’s pharmacoeconomic model. Harms of treatment were identified as important by patients, clinician groups, and the clinical expert, with infections deemed to be notable harms.

Table 8: Summary of Outcome Measures and Their Measurement Properties

CI = confidence interval; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA = Myasthenia Gravis Foundation of America; MGFA-PIS = Myasthenia Gravis Foundation of America Post-Intervention Status; MG-QoL15 = Myasthenia Gravis Quality of Life 15-item; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item – Revised; MID = minimal important difference; QMG = Quantitative Myasthenia Gravis; SD = standard deviation; VAS = visual analogue scale.

Statistical Analysis

Sample Size and Power Calculation

The RAISE study, with a primary efficacy end point of change from baseline to week 12 in MG-ADL score, initially had a planned enrolment of 130 patients, based on the assumption that there would be a difference in treatment group LS means of 2.3, with an SD of 3.4, and 65 patients per group, which would provide a power to detect a difference between groups of approximately 94%, based on a 2-sided alpha of 0.05, with assumed rates of rescue therapy and dropout of up to 10% and 5%, respectively. In protocol amendment version 2.0 (December 18, 2020), the total sample size was increased to 156 patients (78 patients per group) to account for higher variability in the primary end point than was originally assumed and to maintain the power of the study.

Statistical Testing

Multiplicity Control

The primary efficacy end point and ranked secondary end points were included in the statistical hierarchy, which used a fixed-sequential testing procedure to account for multiplicity, where statistical testing of an end point was investigated only if the null hypothesis for the previous end point had been rejected (i.e., if P ≤ 0.05). The statistical analysis of the secondary efficacy end points accounted for multiplicity and controlled the family-wise type I error rate at a 2-sided alpha level of 0.05 by using a parallel gatekeeping testing framework, with a different testing procedure for each of the type I error families. This was only to be done if the primary end point was significant at 2-sided type I error of 0.05.

Family 1: Family 1 included the key secondary end points. Testing was done using a fixed-sequential testing procedure in the following order:

1. change from baseline to week 12 in the QMG score

2. change from baseline to week 12 in the MGC score

3. change from baseline to week 12 in the MG-QoL15r score.

A secondary end point analysis in family 1 was considered statistically significant if the 2-sided P value was 0.05 or less for the end point analysis, providing that the primary efficacy end point analysis and all secondary end point analyses occurring earlier in the fixed-sequence were also statistically significant (i.e., had a 2-sided P value ≤ 0.05). This method was used to control the family-wise error rate at the 2-sided 0.05 level. At the first end point that was not significant, no further testing was to be done.

Family 2: If all secondary end points in family 1 were significant at a 2-sided type I error of 5%, family 2 was tested using a Holm procedure at 2-sided type I error. Family 2 included the following secondary end points:

1. time to first administration of rescue therapy over the 12-week treatment period

2. achievement of MSE, defined as an MG-ADL score of 0 or 1, at week 12 without rescue therapy

3. achievement of at least a 3-point reduction in MG-ADL score at week 12 without rescue therapy

4. achievement of at least a 5-point reduction in QMG score at week 12 without rescue therapy.

The Holm procedure is a multistep, step-down procedure. The procedure started with the smallest P value, which was compared to 0.0125. If the first test showed statistical significance, the treatment effect for the respective end point was considered significant. Then testing proceeded, comparing the next smallest P value to 0.017. If the second test showed statistical significance, testing proceeded, comparing the third smallest P value to 0.025. If the third test showed statistical significance, testing proceeded, comparing the largest P value to 0.05. The procedure stopped whenever a step yielded a nonsignificant result. Once an ordered P value was not significant, the remaining larger P values were not evaluated. In this case, the treatment effect could not be concluded for the remaining secondary end points.

Primary End Point

Change from baseline to week 12 in the MG-ADL score was the primary end point, comparing zilucoplan 0.3 mg/kg with placebo in the modified intention-to-treat (mITT) population, using LS mean with corresponding 2-sided 95% CIs and P values. The LS mean of each treatment group, and the LS mean differences between zilucoplan and placebo at week 1, week 2, week 4, and week 8 were also reported with the corresponding 2-sided 95% CIs. Treatment group differences were assessed using a mixed model for repeated measures (MMRM) analysis of covariance (ANCOVA), with treatment, baseline MG-ADL score, baseline QMG score, geographic region, treatment by visit (interaction term), baseline MG-ADL score by visit (interaction term), baseline QMG score by visit (interaction term), and geographic region by visit (interaction term) as fixed effects, and the patient as a random effect. Terms in the MMRM were treated as categorical for treatment, geographic region, and visit (ordinal), and as continuous for baseline scores in the MG-ADL and the QMG. An unstructured covariance structure was used for the repeated measures, wherein if the unstructured covariance structure model failed to converge, a heterogeneous compound symmetric covariance structure was used. Model assumptions (independence, normality, homoscedasticity) were checked using normal probability plots of studentized residuals by actual residuals and a random variation plot of studentized residuals by predicted values.

Subgroup Analyses

Subgroup analyses for change from baseline to week 12 in the MG-ADL score were performed in the mITT population using the following subgroups, in addition to the gMG refractory status (yes or no):

• age (< 65 years or ≥ 65 years)

• sex (female or male)

• ethnicity (Hispanic or Latino, not Hispanic or Latino)

• weight in kg (< 43, 43 to 56, 56 to < 77, 56 to < 150, or ≥ 150)

• body mass index in kg/m² (< 18.5, ≥ 18.5 to < 25, ≥ 25 to < 30, ≥ 30 to < 40, or ≥ 40)

• region (East Asia, Europe, and North America)

• MGFA disease class at baseline (II [IIa, IIb], III [IIIa, IIIb], or IV [IVa, IVb])

• duration of disease at baseline (< median or ≥ median)

• baseline MG-ADL score (≤ 9 or ≥ 10)

• baseline QMG score (≤ 17 or ≥ 18)

• diagnosis of thymoma

• prior occurrence of a crisis (yes or no), prior thymectomy (yes or no)

• prior steroid therapy (yes or no)

• steroid therapy taken at baseline (yes or no)

• prior NSIST (yes or no)

• NSIST at baseline (yes or no)

• prior history of IVIg or SCIg or PLEX (yes or no)

• timing of patient enrolment relative to COVID-19 pandemic periods (before, during, or after)

• timing of week 12 visit relative to COVID-19 pandemic periods (before, during, or after).

For patients with and without refractory disease, the change from baseline to week 12 in MG-ADL score was assessed in each stratum separately, using an MMRM with treatment and visit (categorical) as fixed effects, baseline MG-ADL score, baseline QMG score, region (East Asia, Europe, or North America) as covariates, and baseline-by-time interaction and treatment-by-time interaction. An unstructured covariance structure was used. In the event that the unstructured covariance structure model failed to converge, an autoregressive (1) covariance structure was used. The Kenward-Roger approximation was used to estimate the denominator degrees of freedom. The LS mean and SE of each treatment group, and the LS mean differences between zilucoplan and placebo were reported for the week 12 visit, along with the corresponding 2-sided 95% CIs for each subgroup stratum.

Subgroup analyses were considered exploratory and not powered to detect treatment differences between groups; no formal statistical testing of treatment-by-subgroup interactions, nor of treatment effects within subgroups, was conducted. Subgroup analyses were only performed for subgroups in which there were at least 5 patients in each subgroup level.

Missing Data and Censoring

The number of missing values by visit was provided by treatment group and overall using descriptive statistics for the mITT population. For the primary efficacy analysis, missing or censored data were imputed with the baseline or the last available MG-ADL score (including unscheduled visit), whichever was worse, from the time of the subsequent ICE, defined as treatment failure:

• ICE1: data occurring after a patient received rescue therapy

• ICE2: data occurring after a patient had any death or myasthenic crisis

• ICE3: any other monotone missing data were assumed to be MAR; that is, it was assumed that the patient had remained on their assigned study drug throughout the study and did not discontinue the study.

If a patient received rescue therapy, efficacy end points that occurred after rescue therapy were censored. Based on the results of the zilucoplan phase II study and the eculizumab phase III study, approximately 10% of patients were expected to receive rescue therapy and 5% were expected to be missing MG-ADL data during the 12-week treatment period.

Sensitivity Analyses

A sensitivity analysis was conducted to evaluate deviations from the MAR pattern using a jump to reference (J2R) approach and an MI method under the MNAR assumption to handle any missing scores. Intermittent missing data were imputed using the Markov chain Monte Carlo method. Monotone missing and/or censored data were imputed using the placebo distribution, irrespective of individual treatment assignment. Variables included in the imputation model were baseline MG-ADL score and MG-ADL scores at previous visits. If a patient started rescue therapy, the most recent MG-ADL score was used, which may have occurred at an unscheduled visit. If the MI model did not converge or produce estimates (e.g., due to overspecification), the set of imputation variables could be modified. Each of the imputed datasets were analyzed using the specified MMRM model. The treatment effects and SEs from the LS mean treatment difference were combined across the 100 imputed datasets, using Rubin’s rule to produce an overall P value. LS means and 95% CIs were plotted from week 1 to week 12. A second sensitivity analysis was conducted to evaluate the robustness of the MAR assumptions in the primary analysis, using tipping-point analyses in which results were statistically significant (P < 0.05); the analysis finds a (tipping) point in the spectrum of assumptions (from less conservative to more conservative), at which point conclusions change from being favourable to the experimental treatment to being unfavourable, and clinical judgment is applied to the plausibility of the assumptions underlying the tipping point. Patients who received rescue therapy (ICE1) or experienced a death or myasthenic crisis (ICE2) were imputed for baseline score or last available MG-ADL score, whichever was worse from time of the ICE. The tipping-point analysis for monotone missing data (ICE3) was assumed to be MAR in the primary analysis. Delta adjustments were made to assumed responses on the ICE3 in each treatment group, including scenarios in which patients with missing data who were randomized to zilucoplan were less likely to respond than patients with missing data who were randomized to placebo.

Supplemental Analyses

Supplemental analyses were conducted to handle missing data and data after rescue therapy using MI. Monotone linear regression imputation methods (100 imputed datasets) were employed to impute missing postbaseline scores at the scheduled study visits, using a sequential approach to produce a monotonic missing data pattern. Imputation was performed by treatment group, with the underlying imputation distribution based on the reason for the missing data. The MI method allowed for a pattern-mixture model approach, which assumed that the data are MNAR under the scenarios in which a patient’s discontinuation from the study was related to the study drug or the patient received rescue medications.

Key Secondary End Points

The key secondary end points of change from baseline to week 12 in the QMG score, the MGC score, and the MG-QoL15r score were analyzed using the same statistical approach as the primary end point.

Secondary End Points

Achievement of MSE, Defined as an MG-ADL Score of 0 or 1 Without Rescue Therapy

The MG-ADL value of 0 or 1 at week 12 without rescue therapy in the mITT population was calculated as follows: week 12 MG-ADL value = baseline MG-ADL score + week 12 change from baseline in MG-ADL score, where a value of 1.0 or lower was considered valid. The number and percentage of patients achieving MSE without rescue therapy were summarized by treatment group for each scheduled visit. ORs with corresponding 95% CIs and P values were estimated. The treatment effects and SEs were combined across 100 imputed datasets to produce an overall P value. Missing data were imputed using nonresponder imputation (i.e., study participants with missing data at the time point of interest were treated as responders). Any patients using rescue therapy (ICE1) and any patients who experienced death or myasthenic crisis (ICE2) were assumed to be nonresponders from the time of the ICE. Any other monotone missing data (ICE3) were assumed to be MAR before the end point was dichotomized.

MG-ADL Responder

An MG-ADL responder was defined as the achievement of at least a 3-point reduction in MG-ADL score at week 12 without rescue therapy. The MG-ADL responder rate was calculated with the same analysis that was used to estimate the achievement of MSE.

QMG Responder

A QMG responder was defined as the achievement of at least a 5-point reduction in QMG score at week 12 without rescue therapy.

Safety

Incidence rates were summarized descriptively for AEs, TEAEs, SAEs, TEAEs leading to study discontinuation, and deaths. The number and percentage of patients who experienced at least 1 TEAE were presented by treatment group, and overall, for each specific system organ class and preferred term, by maximum severity and by relationship to the study drug for each treatment group. Analyses of AEs were performed using the safety population.

Exploratory End Point

Minimal Manifestation Status per MGFA-PIS

The achievement of minimal manifestation status per MGFA-PIS was evaluated as a dichotomous end point, which was having or not having a response of minimal manifestations at week 12 without rescue therapy during the 12-week treatment period.

Table 9: Statistical Analysis of Efficacy End Points in the RAISE Study

ANCOVA = analysis of covariance; ICE = intercurrent event; J2R = jump to reference; MAR = missing at random; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA-PIS = Myasthenia Gravis Foundation of America Post-Intervention Status; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item – Revised; MMRM = mixed model for repeated measures; MNAR = missing not at random; MSE = minimal symptom expression; NR = not reported; NRI = nonresponder imputation; QMG = quantitative myasthenia gravis.

Note: No data were missing for the primary or key secondary end points at week 12 due to COVID-19.

^aIf a patient was administered a rescue therapy, any data strictly after the start date of the first rescue therapy administration were flagged as ICE1.

^bIf a patient experienced death, data after the death date was flagged as ICE2. If a patient experienced a myasthenic crisis, any data after the start date of the adverse event were flagged as ICE2.

^cAny other monotone missing data were flagged as ICE3 in each relevant dataset. ICE3 was defined as any missing value for which all values at subsequent visits were missing.

Source: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

Analysis Populations

Analysis sets included in the RAISE study are summarized in Table 10.

Table 10: Analysis Populations of the RAISE Study

ITT = intention to treat; MG-ADL = Myasthenia Gravis Activities of Daily Living; mITT = modified intention to treat.

Source: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

Results

Patient Disposition

At the data cut-off date of December 30, 2021, 174 patients with anti-AChR antibody–positive gMG were enrolled and randomized in the RAISE study and included in the mITT population (Table 11). Of the patients who were enrolled, 166 patients (95.4%) completed the study, including 82 of 86 patients (95.3%) in the zilucoplan group and 84 of 88 patients (95.5%) in the placebo group. The 8 patients (4.6%) who discontinued the study were balanced between treatment groups (i.e., 4 patients in each group). The primary reasons for discontinuation were as follows: 3 (1.7%) were due to withdrawal by patient (1 patient [1.2%] in the zilucoplan group and 2 patients [2.3%] in the placebo group), 2 (1.1%) were due to death (1 [1.2%] patient in the zilucoplan group and 1 [1.1%] patient in the placebo group), 2 (1.1%) were due to an AE (2 patients [2.3%] in the zilucoplan group and no patient in the placebo group), and 1 (0.6%) was due to physician decision (no patient in the zilucoplan group and 1 [1.1%] patient in the placebo treatment group). Among the discontinuations due to death, the event in the zilucoplan group was related to COVID-19 and the event in the placebo group was due to a cerebral hemorrhage. Among the discontinuations due to AEs in the zilucoplan group, 1 patient withdrew due to aphthous ulcers, and 1 patient withdrew due to hepatic enzyme increased. The discontinuation related to physician decision was due to a suspected reaction to the meningococcal vaccine.

In the prespecified refractory subgroup (88 patients), 85 patients (96.6%) completed the study, including 43 of 44 patients (97.7%) in the zilucoplan group and 42 of 44 patients (95.5%) in the placebo group (Table 11). Overall discontinuation rates were low, with 3 patients (3.4%) who discontinued the study (i.e., 1 patient and 2 patients in the zilucoplan group and placebo group, respectively). The primary reasons for discontinuation were as follows: 2 (2.3%) were due to withdrawal by patient (1 patient each per group), and 1 patient (1.1%) was due to physician decision (in the placebo group). Most patients (96.6%) entered the RAISE-XT extension study after completion of the RAISE study.

Table 11: Patient Disposition in the RAISE Study (Overall mITT Population and Refractory Subpopulation)

mITT = modified intention to treat; NA = not applicable.

^aSafety follow-up was not required for patients who enrolled in the RAISE-XT study.

Sources: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

Baseline Characteristics

Baseline characteristics of the overall trial population and the refractory subpopulation in the RAISE study summarized in Table 12 are limited to those that are most relevant to this review or were felt to affect the outcomes or interpretation of the study results.

More patients in the overall trial population (n = 174) were female (56.9%) than male (43.1%), and the mean age was 53.0 years (SD = 15.1 years). Most patients had a diagnosis of MGFA disease class III at screening (67.2%), generalized symptoms at disease onset (64.4%), disease onset at a mean of 43.8 years (SD = 18.0 years), and a disease duration of mean 9.2 years (SD = 9.9 years). Higher scores in the stratified randomization were more common for the MG-ADL (62.1% of patients had a score ≥ 10) and the QMG (56.3% of patients had a score ≥ 18) than for other scores. Disease characteristics and history were similar in the 2 treatment groups, except a higher proportion of patients had undergone thymectomy in the zilucoplan group (52.3%) than in the placebo group (42.0%). Patients had similar between-group mean baseline scores in the MG-ADL and the QMG.

In the refractory subpopulation (88 of 174 [50.6%]), more patients were female (64.8%) than male (35.2%). The mean age was 51.8 years (SD = 14.0 years), 69.3% of patients had been diagnosed with MGFA disease class III , 69.3% of patients had generalized symptoms at disease onset, mean age at disease onset was 40.0 years (SD = 16.7 years), and mean disease duration was 11.9 years (SD = 10.6 years). Disease characteristics and history were similar in the 2 treatment groups, except a higher proportion of patients had undergone thymectomy in the zilucoplan group than in the placebo group (75.0% versus 61.4%).

In both the overall population and the refractory subpopulation, demographic characteristics, symptoms at onset, time since most recent MG crisis, and baseline scores in the MG-ADL and QMG were similar. There were differences in the overall population and the refractory subpopulation in mean age at disease onset (43.8 years versus 40.0 years), mean disease duration (9.2 years versus 11.9 years), the proportion of patients with a diagnosis of thymoma (22.4% versus 29.5%), the proportion of patients who had undergone thymectomy (47.1% versus 68.2%), and the proportion of patients who had experienced an MG crisis (32.8% versus 50.0%).

Table 12: Baseline Characteristics in the RAISE Study (Overall mITT Population and Refractory Subpopulation)

BMI = body mass index; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGFA = Myasthenia Gravis Foundation of America; mITT = modified intention to treat; QMG = Quantitative Myasthenia Gravis; SD = standard deviation.

^aTime since most recent crisis (months) was calculated as: (date of study day 1 minus date of crisis) divided by (365.25 divided by 12).

Sources: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

Baseline medications were defined as medications that started before and continued after a patient received the assigned study drug. Baseline medications for gMG in the overall population and the refractory subpopulation are summarized in Table 13.

Among patients with gMG in the overall population, most patients (96.0%) were receiving gMG-specific baseline medications, including pyridostigmine (80.5%), prednisone (40.8%), prednisolone (20.7%), mycophenolate mofetil (19.5%), and azathioprine (17.8%). More patients in the zilucoplan group than in the placebo group were receiving prednisone (44.2% versus 37.5%) and prednisolone (23.3% versus 18.2%). Fewer patients in the zilucoplan group than in the placebo group were receiving azathioprine (15.1% versus 20.5%). No patient with gMG was receiving IVIg, SCIg, PLEX, or rituximab at baseline.

Among patients in the refractory subpopulation, the proportion of patients receiving gMG-specific baseline medications was, overall, similar to the proportion in the overall population, including pyridostigmine (79.5%), prednisone (37.5%), azathioprine (23.9%), and mycophenolate mofetil (22.7%). More patients in the zilucoplan group than in the placebo group were receiving prednisone (40.9% versus 34.1%) and prednisolone (36.4% versus 31.8%). Fewer patients in the zilucoplan group than in the placebo group were receiving azathioprine (22.7% versus 25.0%) and mycophenolate mofetil (20.5% versus 25.0%). No patient with refractory gMG was receiving IVIg, SCIg, PLEX, or rituximab at baseline.

In the RAISE trial, prior gMG medications were defined as any medication that started before the first administration of a study drug (refer to Table 32 in Appendix 1). The proportions of patients who had received at least 1, at least 2, and at least 3 prior lines of therapy were 100%, about 95%, and about 81%, respectively, in each study group in the RAISE trial. The majority of patients in both study groups had received prior acetylcholinesterase inhibitors (about 95% of patients), steroids (about 85% of patients), and NSISTs (around 70% of patients) (Table 32).

Table 13: gMG-Specific Baseline Medications in the RAISE Study (Overall Population and Refractory Subpopulation, Safety Set)

gMG = generalized myasthenia gravis; Ig = immunoglobulin; NOS = not otherwise specified; SC = subcutaneous.

Note: Within this table, where the concomitant and preferred terms are the same, only the concomitant term is displayed.

Sources: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

Exposure to Study Treatments

In the RAISE study, the mean duration of exposure to the study drug was 81.9 days (SD = 11.0 days) in the zilucoplan group and 81.1 days (SD = 13.0 days) in the placebo group (Table 14). Most patients received the study drug for 84 days or longer (73.3% and 64.8% of patients in the zilucoplan and placebo group, respectively). At week 12, total exposure was 19.8 patient-years in the zilucoplan group and 20.0 patient-years in the placebo group. Ten patients (11.6%) in the zilucoplan group and 15 patients (17.0%) in the placebo group missed at least 1 dose of the study drug. The mean number of doses missed was 5.4 (SD = 5.2; range, 1 to 15 doses) in the zilucoplan and 6.2 (SD = 10.9; range, 1 to 44 doses) in the placebo group.

In the refractory subpopulation, the mean duration of exposure to the study drug was 83.8 days (SD = 3.7 days) in the zilucoplan group and 80.9 days (SD = 14.4 days) in the placebo group. At week 12, total exposure was 10.4 patient-years in the zilucoplan group and 9.9 patient-years in the placebo group. Four patients (9.1%) in the zilucoplan and 11 patients (25.0%) in the placebo group missed at least 1 dose of the study drug. The mean number of doses missed was 7.3 (SD = 5.4; range, 3 to 15 doses) in the zilucoplan group and 3.4 (SD = 3.6; range, 1 to 12 doses) in the placebo group.

Table 14: Treatment Exposure in the RAISE Study (Overall Population and Refractory Subpopulation, Safety Set)

SD = standard deviation.

Sources: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

The mean duration of time that patients were receiving corticosteroids and NSISTs before baseline was 988.6 days (median, 218.5 days) and 1,187.4 days (median, 788.0 days), respectively, in the overall trial population. In the refractory population, the mean duration was 1,304.3 days (median, 412.5 days) and 1,457.0 days (median, 947.0 days), respectively.

Concomitant Medications

Concomitant gMG therapies and concomitant-only gMG therapies are summarized in Table 15.

Concomitant gMG therapies included medications that were taken at least once after the first administration of the study drug during the treatment period. Overall, of the 169 patients (97.1%) who received any concomitant gMG therapy, 84 patients (97.7%) were in the zilucoplan group, and 85 patients (96.6%) were in the placebo group. The most common (≥ 10% of patients overall) concomitant gMG therapies were pyridostigmine (82.8%), prednisone (41.4%), prednisolone (20.7%), mycophenolate mofetil (19.0%), and azathioprine (17.8%). Concomitant gMG therapies were generally balanced between treatment groups.

Concomitant-only gMG therapies were defined as any medication that started after the first administration of the study drug and continued during the treatment period. A total of 18 patients (10.3%) received any concomitant-only gMG therapy (12 patients [14.0%] and 6 patients [6.8%] in the zilucoplan and placebo groups, respectively). Overall, the rate of patients receiving concomitant-only gMG therapies was lower than patients receiving concomitant gMG therapies.

Among patients in the refractory subpopulation, 85 patients (96.6%) received any concomitant gMG medications. The most common (≥ 10% of patients overall) concomitant gMG therapies were pyridostigmine (81.8%), prednisone (38.6%), prednisolone (34.1%), azathioprine (23.9%), mycophenolate mofetil (22.7%), cyclosporin (11.4%), and tacrolimus (11.4%). Concomitant-only gMG therapies were not reported for the refractory subpopulation.

Table 15: Concomitant and Concomitant-Only gMG Therapies in the RAISE Study (Overall Population and Refractory Subpopulation, Safety Set)

gMG = generalized myasthenia gravis; NOS = not otherwise specified; NR = not reported.

^aConcomitant-only gMG therapies were defined as any medication that started after the first administration of the study drug and continued during the treatment period.

Sources: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

Efficacy

Findings for key efficacy outcomes in the RAISE study at the data cut-off date of December 30, 2021, are summarized in Table 16.

Primary End Point: Change From Baseline to Week 12 in MG-ADL Score

In the overall population, the LS mean change from baseline to week 12 in the MG-ADL score was –4.39 (SE = 0.45) in the zilucoplan group and –2.30 (SE = 0.44) in the placebo group. The LS mean difference between groups in the MG-ADL score was –2.09 (95% CI, –3.24 to –0.95; P < 0.001), favouring zilucoplan (Figure 2). In the refractory subpopulation, the LS mean change from baseline to week 12 in the MG-ADL score was –4.72 (SE = 0.58) in the zilucoplan group and –1.62 (SE = 0.58) in the placebo group (LS mean difference of –3.11; 95% CI, –4.69 to –1.52; nominal P < 0.001).

Subgroup Analyses

Findings across all subgroups for mean change from baseline to week 12 in the MG-ADL score were consistent with the primary analysis result, which favoured zilucoplan over placebo, and included results identified as potential treatment modifiers for the following subgroups: duration of disease at baseline, MGFA disease class at baseline, baseline MG-ADL score, baseline QMG score, refractory MG, any prior MG crisis, prior treatments (steroid, immunosuppressive [nonsteroidal], IVIg, SCIg, or PLEX therapy). The sample size was small for many subgroups.

Missing or Censored Data

In the overall population, the number of patients with no ICEs in the MG-ADL score was 80 of 86 patients (93.0%) and 77 of 88 patients (87.5%) in the zilucoplan and placebo groups, respectively. The number of patients with ICE1 (received rescue therapy) in the MG-ADL score was 4 patients (4.7%) and 10 patients (11.4%) in the zilucoplan and placebo groups, respectively. The number of patients with ICE2 (any death or myasthenic crisis) in the MG-ADL score was 1 patient (1.2%) and 1 patient (1.2%) in the zilucoplan and placebo groups, respectively.

Figure 2: LS Mean Change From Baseline to Week 12 in MG-ADL Score Using MMRM ANCOVA (Overall Population, mITT)

ANCOVA = analysis of covariance; CFB = change from baseline; CI = confidence interval; LS = least squares; LSM = least squares mean; MG-ADL = Myasthenia Gravis Activities of Daily Living; mITT = modified intention to treat; MMRM = mixed model for repeated measures.

Source: RAISE Clinical Study Report.¹⁴

Sensitivity Analyses

Two sensitivity analyses were conducted to assess the impact of missing scores on the primary efficacy end point of LS mean change from baseline to week 12 in the MG-ADL score. Findings were similar to the primary analysis when analyzed using MMRM ANCOVA, based on the J2R approach (LS mean difference of –2.03; 95% CI, –3.16 to –0.89; nominal P < 0.001) and based on the tipping-point approach (shift of 25.95 points [LS mean difference of –1.514; 95% CI, –3.057 to 0.029; P = 0.0544] for the zilucoplan group and –17.26 points [LS mean difference of –1.470; 95% CI, –2.981 to 0.042; P = 0.0567] for the placebo group).

Supplemental Analyses

Findings were similar to the primary analysis when analyzed using MMRM ANCOVA, based on the pattern-mixture approach (LS mean difference of –2.13; 95% CI, –3.27 to –0.98; P < 0.001).

Key Secondary End Points

Change From Baseline to Week 12 in the QMG Score

In the overall population, the LS mean change from baseline to week 12 in the QMG score was −6.19 (SE = 0.56) in the zilucoplan group and −3.25 (SE = 0.55) in the placebo group (LS mean difference of –2.94; 95% CI, –4.39 to –1.49; P < 0.001). In the refractory subpopulation, the LS mean change from baseline to week 12 in the QMG score was –6.08 (SE = 0.76) in the zilucoplan group and –2.76 (SE = 0.75) in the placebo group (LS mean difference of –3.32; 95% CI, –5.42 to –1.23; nominal P < 0.001).

Change From Baseline to Week 12 in the MGC Score

In the overall population, the LS mean change from baseline to week 12 in the MGC score was –8.62 (SE = 0.81) in the zilucoplan group and –5.42 (SE = 0.79) in the placebo group (LS mean difference of –3.20; 95% CI, –5.24 to –1.16; P = 0.0023). In the refractory subpopulation, the LS mean change from baseline to week 12 in the MGC score was –7.85 (SE = 1.09) in the zilucoplan group and –4.17 (SE = 1.07) in the placebo group (LS mean difference of –3.68; 95% CI, –6.65 to –0.72; nominal P = 0.0156).

Change From Baseline to Week 12 in the MG-QoL15r Score

In the overall population, the LS mean change from baseline to week 12 in the MG-QoL15r score was –5.65 (SE = 0.77) in the zilucoplan group and –3.16 (SE = 0.76) in the placebo group (LS mean difference of –2.49; 95% CI, –4.45 to –0.54; P = 0.0128). In the refractory subpopulation, the LS mean change from baseline to week 12 in the MG-QoL15r score was –5.63 (SE = 0.96) in the zilucoplan group and –2.36 (SE = 0.95) in the placebo group (LS mean difference of –3.28; 95% CI, –5.89 to –0.67; nominal P = 0.0145).

Additional Secondary End Points

Achievement of MSE at Week 12 Without Rescue Therapy

In the overall population, the percentage of patients who achieved MSE (an MG-ADL score of 0 or 1) at week 12 without rescue therapy was 14.0% in the zilucoplan group and 5.8% in the placebo group (OR = 2.608; 97.5% CI, 0.793 to 9.209; P = 0.0885). The difference in the percentage of patients who achieved MSE was 8.2% (97.5% CI, –1.9% to 18.3%). In the refractory subpopulation, the number of patients who achieved MSE at week 12 without rescue therapy was 2 of 44 patients (4.5%) and 0 of 43 patients in the zilucoplan and placebo groups, respectively.

Achievement of at Least a 3-Point Reduction in MG-ADL Score at Week 12 Without Rescue Therapy

In the overall population, the percentage of patients who were MG-ADL responders (achieved at least a 3-point decrease in the MG-ADL score) at week 12 without rescue therapy was 73.1% in the zilucoplan group and 46.1% in the placebo group (OR = 3.184; 98.75% CI, 1.391 to 7.293; P < 0.001). The difference in the percentage of patients who were MG-ADL responders was 27.0% (98.75% CI, 9.0% to 44.9%). In the refractory subpopulation, the number of patients who were MG-ADL responders at week 12 without rescue therapy was 33 of 44 patients (75.0%) and 17 of 42 patients (40.5%) in the zilucoplan and placebo groups, respectively.

Achievement of at Least a 5-Point Reduction in QMG Score at Week 12 Without Rescue Therapy

In the overall population, the percentage of patients who were QMG responders (achieved at least a 5-point decrease in the QMG score) at week 12 without rescue therapy was 58.0% in the zilucoplan group and 33.0% in the placebo group (OR = 2.865; 98.33% CI, 1.319 to 6.225; P = 0.0012). The difference in the percentage of patients who were QMG responders was 25.0% (98.33% CI, 7.3% to 42.7%). In the refractory subpopulation, the number of patients who were QMG responders at week 12 without rescue therapy was 24 of 43 patients (55.8%) and 11 of 41 patients (26.8%) in the zilucoplan and placebo groups, respectively.

Exploratory End Point

Minimal Manifestation Status per MGFA-PIS at Week 12 Without Rescue Therapy

In the overall population, the number of patients who achieved minimal manifestation status per MGFA-PIS at week 12 without rescue therapy was 22 of 78 (28.2%) in the zilucoplan group and 16 of 83 (19.3%) in the placebo group. The difference in the percentage of patients who achieved minimal manifestation status per MGFA-PIS was 10.9% (95% CI, –1.5% to 23.2%). The number of patients who achieved minimal manifestation status per MGFA-PIS at week 12 without rescue therapy was not reported in the refractory subpopulation.

Table 16: Key Efficacy Results in the RAISE Study (Overall mITT Population and Refractory Subpopulation)

ANCOVA = analysis of covariance; CDA-AMC = Canada’s Drug Agency; CI = confidence interval; ICE = intercurrent event; LS = least squares; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA-PIS = Myasthenia Gravis Foundation of America Post-Intervention Status; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item – Revised; mITT = modified intention to treat; MMRM = mixed model of repeated measures; MSE = minimal symptom expression; NR = not reported; OR = odds ratio; QMG = Quantitative Myasthenia Gravis; SD = standard deviation; SE = standard error.

Note: Baseline was defined as the last available predose value before the first injection of the study drug in the treatment period or, if missing, the screening value. MG-ADL scores after rescue therapy (ICE1) or any death or myasthenic crisis (ICE2) were censored and considered as treatment failures. Any missing data due to ICE1 or ICE2 were imputed based on baseline MG-ADL score or on the last available MG-ADL score, whichever was worse. Other missing scores were handled with the maximum likelihood estimation method under the missing at random assumption. Analysis was based on an MMRM ANCOVA model, using an unstructured correlation matrix, with treatment, baseline MG-ADL score, baseline QMG score, region (East Asia, Europe, and North America), and the interactions terms of treatment by visit and baseline MG-ADL score by visit as fixed effects; patients were added as random effects in the model. The MMRM ANCOVA includes week 1, week 2, week 4, week 8, and week 12.

^aPatients with no missing score at week 12.

^bLS mean difference was calculated as zilucoplan minus placebo.

^cP value was derived using an MMRM ANCOVA model (based on imputed data after treatment failure), using an unstructured correlation matrix, with treatment, baseline MG-ADL score, baseline QMG score, region, with the interaction terms of treatment by visit and baseline MG-ADL score by visit as fixed effects and patients as random effects in the model.

^dP value corresponds to the primary analysis of the secondary end point of change from baseline to week 12 in the QMG score; after multiplicity adjustment, a result was considered statistically significant if the primary analysis of change from baseline to week 12 in the MG-ADL score was also statistically significant at alpha of 0.05 using 2-sided statistical testing.

^eP value corresponds to the primary analysis of the secondary end point of change from baseline to week 12 in the MGC score; after multiplicity adjustment, a result was considered statistically significant if the primary analysis of change from baseline to week 12 in the MG-ADL score and the change from baseline in QMG score were also statistically significant at alpha of 0.05 using 2-sided statistical testing.

^fP value corresponds to the primary analysis of the secondary end point of change from baseline to week 12 in the MG-QoL15r score; after multiplicity adjustment, a result was considered statistically significant if the primary analysis of change from baseline to week 12 in the MG-ADL score and change in baseline in MGC and QMG scores were also statistically significant at alpha of 0.05 using 2-sided statistical testing.

^gAchievement of MSE is defined as having an MG-ADL score of 0 or 1.

^hRisk difference was not included in the sponsor’s planned analyses; the absolute risk difference with associated CI was requested by the CDA-AMC review team for interpretation purposes. Based on the testing procedure used for the secondary end points, the CIs were 98.75% for MG-ADL responder, 98.3% for QMG responder, and 97.5% for the achievement of MSE in the statistical hierarchy using the Holm’s procedure.

^jOdds ratio was estimated between treatment groups using a logistic regression model, with treatment as factor, and baseline MG-ADL score, baseline QMG score, and region as covariates at each imputed dataset. Treatment effects were combined across 100 imputed datasets to produce an overall treatment effect. An OR greater than 1 indicated a greater likelihood of response to zilucoplan than to placebo. Based on the testing procedure used for the secondary end points, the CI was 98.75% for MG-ADL responder, 98.3% for QMG responder, and 97.5% for the achievement of MSE in the statistical hierarchy using the Holm’s procedure. The associated CI was generated using a post hoc analysis.

^jP value was adjusted in the statistical hierarchical testing after the Holm’s procedure, using the threshold of 0.0125 for the MG-ADL score, 0.017 for the QMG score, and 0.025 for the achievement of MSE.

^kAn MG-ADL responder is defined as having at least a 3-point improvement (decrease) in the MG-ADL score.

^lA QMG responder is defined as having at least a 5-point improvement (decrease) in the QMG score.

^mThis end point was not adjusted for multiple testing, so there is a potential for type I error to be increased, and findings should be considered supportive.

ⁿOdds ratio (95% CI) was not included in the sponsor’s planned analyses; the relative effect was requested by the CDA-AMC review team for interpretation purposes.

Sources: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

Harms

Harms data for the RAISE study at the data cut-off date of December 30, 2021, are summarized in Table 17.

Adverse Events

In the overall population, the number of patients with at least 1 TEAE was 66 of 86 patients (76.7%) in the zilucoplan group and 62 of 88 patients (70.5%) in the placebo group. The most common TEAEs, occurring in greater than 5% of patients in either the zilucoplan group or the placebo group, respectively, were headache (15.1% versus 15.9%), injection-site bruising (16.3% versus 9.1%), MG (10.5% versus 9.1%), diarrhea (10.5% versus 2.3%), injection-site pain (9.3% versus 3.4%), urinary tract infection (8.1% versus 4.5%), contusion (8.1% versus 3.4%), increased lipase (8.1% versus 1.1%), nasopharyngitis (5.8% versus 3.4%), vomiting (3.5% versus 5.7%), rash (3.5% versus 5.7%), and increased amylase (5.8% versus 2.3%). In the refractory subpopulation, the number of patients with at least 1 TEAE was 39 (88.6%) in the zilucoplan group and 34 (77.3%) in the placebo group. The most common TEAEs, occurring in greater than 10% of patients in either the zilucoplan group or the placebo group, respectively, were headache (20.5% versus 15.9%), MG (13.6% versus 13.6%), injection-site bruising (15.9% versus 11.4%), diarrhea (15.9% versus 2.3%), and vomiting (4.5% versus 11.4%).

Serious Adverse Events

In the overall population, the number of patients with at least 1 SAE was 11 (12.8%) in the zilucoplan group and 13 (14.8%) in the placebo group. The most common SAEs reported in at least 2% of patients in either the zilucoplan group or the placebo group, respectively, were MG (2.3% versus 5.7%), COVID-19 (1.2% versus 2.3%), and COVID-19 pneumonia (1.2% versus 2.3%). In the refractory subpopulation, the number of patients with at least 1 SAE was 6 patients (13.6%) in the zilucoplan group and 8 patients (18.2%) in the placebo group. SAEs specified by system organ class were not reported for the refractory subpopulation.

Withdrawals Due to Adverse Events

In the overall population, the number of patients who stopped the study treatment due to AEs was 4 (4.7%) in the zilucoplan group and 2 (2.3%) in the placebo group. Withdrawals due to AEs in the zilucoplan group were due to (1 patient [1.2%] each) aphthous ulcer, mouth ulceration, COVID-19, and increased hepatic enzyme; the TEAE of aphthous ulcer and the TEAE of COVID-19 were considered serious, with the latter having had a fatal outcome. Withdrawals due to AEs in the placebo group were due to (1 patient [1.1%] each) cerebral hemorrhage and hyperemesis gravidarum; both TEAEs were considered serious and the TEAE of cerebral hemorrhage had a fatal outcome. In the refractory subpopulation, 1 patient (2.3%) stopped the study treatment in the zilucoplan group, and no patients stopped the study treatment in the placebo group. Reasons for withdrawal due to AEs were not reported for the refractory subpopulation.

Mortality

In the overall population of the RAISE study, 2 patients died: 1 patient (1.2%) in the zilucoplan group who experienced an SAE leading to death due to COVID-19 and COVID-19 pneumonia, and 1 patient (1.1%) in the placebo group who experienced an SAE leading to death due to cerebral hemorrhage. In the refractory subpopulation, no deaths were observed.

Adverse Events of Special Interest

Infections

In the overall population, the number of patients who experienced infections was 23 (26.7%) in the zilucoplan group compared with 16 (18.2%) in the placebo group (between-group difference, 8.6%; 95% CI, –3.8% to 20.9%). Of these, 4 patients (4.7%) in the zilucoplan group and 4 patients (4.5%) in the placebo group had serious infections. AEs of special interest were not reported for the refractory subpopulation.

Table 17: Harms Results in the RAISE Study (Overall Population and Refractory Subpopulation, Safety Set)

AE = adverse event; CDA-AMC = Canada’s Drug Agency; CI = confidence interval; NR = not reported; SAE = serious adverse event; TEAE = treatment-emergent adverse event.

^aRisk difference (95% CI) was not included in the sponsor’s planned analyses; the absolute risk difference was requested by the CDA-AMC review team for interpretation purposes.

Sources: RAISE Clinical Study Report,¹⁴ sponsor’s Summary of Clinical Evidence.²⁷

Critical Appraisal

Internal Validity

Randomization appeared to be adequate in the RAISE trial; the treatment groups were balanced, overall, for demographic and disease characteristics. A greater proportion of patients with gMG in the zilucoplan than in the placebo group had undergone thymectomy in the overall population (52.3% versus 42.0%) and in the refractory subpopulation (75.0% versus 61.4%), which had the potential to influence treatment effects. However, no systemic differences between groups were detected on other key prognostic factors or in the medical history of patients, which indicates that randomization was likely successful, and the risk of selection bias was low. Randomization was stratified by baseline MG-ADL score (≤ 9 versus ≥ 10), QMG score (≤ 17 versus ≥ 18), and geographic region (East Asia, Europe, and North America). The instruments used to evaluate the primary and secondary efficacy outcomes (MG-ADL, QMG, MGC, MG-QoL15r) were appropriate, and their psychometric properties have been investigated in patients with MG, although no MIDs have been estimated for the MG-QoL15r. Minimal manifestation status per MGFA-PIS without rescue therapy was based on clinician-assessed patient symptoms of MG after the initiation of MG-specific therapy; this measure was intended to capture patients who may not meet the definition of complete stable remission or pharmacologic remission but who have muscle weakness, based on careful examination. This was an exploratory end point, and no MID has been validated in the indicated population. Results for minimal manifestation status per MGFA-PIS at week 12 were not reported for the refractory subpopulation. There was a low risk of bias for allocation concealment because patients and study staff were blinded to treatment assignment, both treatments were identical, and the unblinding of treatment assignment was not permitted before the initiation of rescue therapy. Patients with refractory disease represented about one-half of the enrolled patients with gMG. The selection criteria for the refractory subgroup were specified a priori and in line with the criteria used to define patients with refractory gMG in other RCTs (e.g., the REGAIN trial for eculizumab). Although randomization was not stratified by refractory status, baseline characteristics were, overall, similar between treatment groups, so concerns regarding the prognostic balance of the refractory subpopulation were low. Subgroup analyses were not adjusted for multiplicity, nor were they specifically powered to detect differences among strata. Otherwise, results from the refractory subgroup can be considered consistent with the analyses of the overall population in the RAISE trial and share the limitations of those analyses. Results for the refractory subgroup showed consistency with the overall trial population across all outcomes.

There was a notable proportion of patients with important protocol deviations in both the zilucoplan group (33 patients [38.4%]) and the placebo group (32 patients [36.4%]). Prohibited concomitant medication use important protocol deviations occurred in 24.7% of patients, which included changes to gMG conventional medications (10 patients and 8 patients in the zilucoplan and placebo groups, respectively), the use of prohibited concomitant medications (9 patients and 8 patients, respectively), and changes to cholinesterase inhibitor dosing less than 10 hours before evaluation (6 patients and 6 patients, respectively). Additionally, deviations from the inclusion criteria occurred in 7.5% of patients; acetylcholinesterase inhibitor therapy was not withheld for at least 10 hours before the QMG assessment (6 patients and 6 patients in the zilucoplan and placebo groups, respectively), the corticosteroid regimen was changed in the 30 days before baseline (1 patient in the zilucoplan group), and inpatient treatment with IVIg was not reported as a rescue therapy (1 patient in the zilucoplan group). The proportion of patients with protocol deviations, such as the use of prohibited medications or changes to gMG conventional medications during the study, were balanced between groups.

MI methods were used to account for missing data in the primary and secondary end points, based on assumptions of MNAR, in which missingness may be related to the study drug or the receipt of rescue medication, and assumptions of MAR. Based on the specified approach, each of these was imputed with either the baseline value or the last observed value (whichever was worse). Because the rate of ICE1 (receipt of rescue therapy with IVIg, PLEX, or eculizumab) was higher in the placebo group (11%) than in the zilucoplan group (5%), this approach appeared to be overly pessimistic and possibly introduced bias in favour of zilucoplan; however, the imbalance was not large enough to raise serious concerns about biased treatment effects. In sensitivity analyses to account for the censoring of patients who experienced treatment failure in the zilucoplan versus placebo groups, respectively, the proportions of patients were aligned with prior clinical trial data (from the zilucoplan phase II and eculizumab phase III studies) and were similar across scores on the MG-ADL, QMG, MGC, and MG-QoL15r for the receipt of rescue therapy (4.7% versus 11.4%, respectively), any death or myasthenic crisis (1.2% versus 1.1%, respectively), and patients with missing data but assumed to have remained on treatment (1.2% versus 0.0%, respectively, for MG-ADL; 2.3% versus 1.1%, respectively, for QMG and MGC; and 3.5% versus 2.3%, respectively, for MG-QoL15r). Overall, the supplemental analyses and sensitivity analyses of the primary analysis did not adequately assess the potential bias related to missing data. However, concerns about loss to follow-up were low, because approximately 95% of patients with gMG in the overall population and the refractory subpopulation completed the RAISE study, with balanced proportions between treatment arms.

External Validity

Per sponsor request, the focus of this review was on the sponsor’s reimbursement request, which was narrower than the Health Canada indication. The reimbursement request aligned with the criteria for the refractory subgroup of the RAISE trial, which was zilucoplan as add-on therapy for the treatment of adult patients with anti-AChR antibody–positive refractory gMG, defined as no achievement of symptom control after treatment for at least 1 year with 2 or more of the following therapies: prednisone, azathioprine, mycophenolate, cyclosporine, cyclophosphamide, methotrexate, tacrolimus, other corticosteroids for gMG, other ISTs; or a history of treatment with at least 1 of these therapies for 1 year or more and the need for chronic PLEX, IVIg, or SCIg at least every 3 months for the 12 months before treatment with zilucoplan. The requested reimbursement criteria also aligned with the reimbursement criteria for the comparator therapy, eculizumab, which received a CADTH positive final recommendation in 2020.

Patients in the RAISE trial who did not have refractory disease (49%) were not included in the reimbursement request. Although the clinical expert consulted for this review agreed that there is a current unmet need in patients with nonrefractory disease who have responded inadequately to the existing standard of gMG therapy, this population was not the focus on this review.

Looking at the types and duration of prior conventional treatments received by patients in the RAISE trial, the clinical expert agreed that all were adequately managed on conventional therapy for gMG at time of enrolment and were reflective of patients who experience an unmet need in the clinical setting in Canada. The concomitant medications received by patients in the RAISE trial represented the range of options currently available in clinical practice in Canada, and the clinical expert felt that exposure to the corticosteroids and NSISTs being used by patients at study enrolment exceeded the typical duration for maximal response of 3 to 4 months for prednisone (or other corticosteroids) and 6 to 18 months for NSISTs.

The RAISE trial inclusion criteria required that corticosteroid and IST doses and types of treatment not be changed for at least 30 days before study baseline, and that therapy changes not be anticipated during the 12-week treatment period. All standard-of-care medications for gMG were to be kept at the same dose throughout the study, including corticosteroids and IST drugs. If an escalation of gMG therapy (i.e., rescue therapy) became necessary because of major deterioration in a patient’s clinical status or because of the risk of MG crisis, per the investigator’s judgment, the patient was allowed to receive IVIg or PLEX treatment as rescue therapy. After the first 12 weeks of the study, if the investigator determined that a reduction in standard-of-care therapy for gMG would be a reasonable course of action, a dose reduction could be initiated. The clinical expert consulted on this review was not concerned about dose changes in concomitant medications over the course of the 12-week treatment period. The expert felt that patients were adequately managed on conventional therapy for gMG at time of enrolment and that exposure to corticosteroids and NSISTs at study enrolment exceeded the typical duration for maximal response. Per the final product monograph, patients eligible for zilucoplan should continue to receive standard therapy throughout the pivotal trial.¹

At baseline, no patients in the RAISE trial had received eculizumab or rituximab (patients who received rituximab in the 12 months before baseline were not eligible for the study); as such, the trial population reflects patients in Canada who have limited or no access to these treatments because of a lack of funding by public drug plans.

According to the expert, stratifying patients during randomization by baseline scores on the MG-ADL and the QMG appeared to be an appropriate way to ensure equal distributions of patients by disease severity. The expert noted that no specific patient populations that might otherwise be considered eligible in clinical practice were missed, and that the patients enrolled in the RAISE study were representative of those observed in clinical practice. Although more patients had undergone thymectomy in the zilucoplan group than in the placebo group in the overall population (52% versus 42%, respectively) and in the refractory subpopulation (75% versus 61%, respectively), this did not contribute significantly to the treatment effects, according to the expert. In line with patients who were not eligible for the RAISE trial, patients with MuSK-positive MG should not be treated with a complement inhibitor such as zilucoplan based its mechanism of action, the expert emphasized; rather, patients with MuSK-positive MG might benefit from treatment with a neonatal FcR inhibitor (e.g., efgartigimod alfa).

The trial’s eligibility criteria, which included an MGFA disease class of II to IV and an MG-ADL total score of at least 6, would select patients with symptomatic gMG most in need of intervention, according to the clinical expert consulted for this review. Very few patients with class IV gMG (fewer than 10 patients in the overall trial population) were included in the trial. The clinical expert consulted for this review indicated that a subset of patients with MGFA disease class I (ocular MG) or class V and an MG-ADL score of less than 6 would be suitable for treatment. Specifically, the clinical expert indicated that patients with ocular MG or mild symptoms can still be refractory to other therapies, and that patients with MGFA disease class V (on a ventilator) who have no contraindications could potentially benefit from zilucoplan. However, the clinical expert noted that zilucoplan has not been studied in these patients, and the results of the trial cannot be directly generalized to these groups of patients.

The expert emphasized that given the relatively high proportion of patients with refractory disease in the RAISE trial (about 50% of patients in the overall population), the lower proportions of patients receiving prednisone (about 40%) and azathioprine or mycophenolate mofetil (up to 20% each) in the overall trial population than in clinical practice was a surprise. Although the number of patients who were on pyridostigmine was numerically higher in the zilucoplan group than in the placebo group, the expert noted that, per the study protocol of withholding this treatment for 10 hours, very few patients added pyridostigmine after the trial started and, combined with its short half-life, the potential for concomitant treatment with pyridostigmine to influence treatment effects is likely very low.

Moreover, the RAISE trial did not provide evidence for comparisons between zilucoplan and other currently available active treatments for gMG. The expert considered that the most relevant comparators for zilucoplan among patients with refractory gMG include other complement inhibitors (i.e., eculizumab, ravulizumab), neonatal FcR inhibitors (e.g., efgartigimod alfa), chronic IVIg, and chronic PLEX; rituximab as a comparator is less applicable because of its limited access in Canada for patients with gMG, its off-label use, the lack of rigorous clinical trial evidence in patients with anti-AChR antibody–positive MG, and some evidence of improved benefit among patients with MuSK-positive MG.

The expert agreed that the primary end point of change from baseline in MG-ADL score was an important outcome for the evaluation of treatment response and is aligned with clinical practice, including the threshold used in the RAISE trial. MIDs used for the QMG and MGC scores were also noted by the expert to align with the literature for thresholds validated in patients with MG.

According to the expert, most patients with gMG would be assessed at approximately 12 weeks to evaluate treatment response, which is aligned with the RAISE trial, with additional assessments conducted at the 3-month or 4-month time point to assess responsiveness or maintenance of response. The randomized controlled period of the RAISE trial was 12 weeks. The longer-term treatment effect of zilucoplan can only be assessed in the open-label extension period of the RAISE-XT study.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered to be most relevant to CDA-AMC expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group.^43,44

• High certainty: We are very confident that the true effect lies close to that of the estimate of the effect.

• Moderate certainty: We are moderately confident in the effect estimate. The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. We use the word likely for evidence of moderate certainty (e.g., X intervention likely results in Y outcome).

• Low certainty: Our confidence in the effect estimate is limited. The true effect may be substantially different from the estimate of the effect. We use the word may for evidence of low certainty (e.g., X intervention may result in Y outcome).

• Very low certainty: We have very little confidence in the effect estimate. The true effect is likely to be substantially different from the estimate of effect. We describe evidence of very low certainty as very uncertain.

In accordance with the GRADE approach, evidence from RCTs started as high-certainty evidence and could be rated down for concerns related to study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias.

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. The target of the certainty of evidence assessment was the presence or absence of a clinically important effect for change from baseline to week 12 in the MG-ADL, the QMG, and the MGC scores, based on thresholds identified in the literature. The certainty of evidence assessments for change from baseline to week 12 in the MG-QoL15r score, the number of patients achieving MSE at week 12 without rescue therapy, an MG-ADL responder rate at week 12 without rescue therapy, a QMG responder rate at week 12 without rescue therapy, the number of patients with minimal manifestation status per MGFA-PIS at week 12 without rescue therapy, and infections, were based on the presence or absence of any (nonnull) effect.

Results of GRADE Assessments

Table 2 presents the GRADE summary of findings for zilucoplan versus placebo in patients with anti-AChR antibody–positive gMG.

Long-Term Extension Studies

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the CDA-AMC review team.

Description of Studies

RAISE-XT (NCT04225871) is an ongoing, phase III, multicentre, open-label extension study of zilucoplan in patients with gMG. At the conclusion of the 12-week treatment period of the RAISE parent study, all eligible patients had the option to enrol in the RAISE-XT study. The objectives of the RAISE-XT study were to evaluate the long-term efficacy, safety, and tolerability of zilucoplan in patients with gMG who had previously participated in a qualifying zilucoplan study (either the phase II MG0009 study [NCT03315130] or the RAISE study). The design of the RAISE-XT study is depicted in Figure 3.

Results of 3 interim analyses in the RAISE-XT study are presented in this report, with clinical cut-off dates of September 8, 2022, May 11, 2023, and November 11, 2023. All data cut-offs include data for the overall population, whereas the November 2023 interim data cut-off also includes data for the refractory subpopulation. Because both the sponsor’s submission and the proposed Health Canada indication recommends a zilucoplan dose of 0.3 mg/kg, this Clinical Review report will present data for patients who were randomized to receive 0.3 mg/kg of zilucoplan or placebo in the double-blind parent studies (MG0009 or the RAISE study); data for patients who received 0.1 mg/kg of zilucoplan are not presented in this report. At the time of review, the RAISE-XT study is ongoing; it is expected to collect efficacy and safety data for up to 7 years. Results are presented for the end points at 12 weeks in the RAISE-XT study. Results at 84 weeks in the RAISE-XT study are presented in Table 33 in Appendix 1.

Figure 3: The RAISE-XT Study Design

E = extension; Wk = week.

Note: At day E1, all patients received 0.3 mg/kg of zilucoplan, regardless of the dose they received in the parent study.

Source: RAISE-XT Clinical Study Report.⁴⁵

Populations

Eligible patients were 18 years or older and had completed a qualifying zilucoplan study (the phase II MG0009 study or the RAISE study). The 2 groups entering the RAISE-XT study were defined as follows:

• The placebo and zilucoplan 0.3 mg/kg group includes patients from the RAISE study who were randomized to placebo in the treatment period and then received zilucoplan 0.3 mg/kg when they entered the RAISE-XT study. It also includes patients in the MG0009 study who were randomized to placebo in the treatment period and then received zilucoplan 0.3 mg/kg in the extension part of the MG0009 study and in the RAISE-XT study.

• The zilucoplan 0.3 mg/kg and zilucoplan 0.3 mg/kg group includes patients from the RAISE study who were randomized to zilucoplan 0.3 mg/kg in the treatment period and received zilucoplan 0.3 mg/kg in the RAISE-XT study. It also includes patients in the phase II MG0009 study who were randomized to zilucoplan 0.3 mg/kg in both the treatment period and the extension periods of the MG0009 study and then received zilucoplan in the RAISE-XT study.

Refractory status at baseline was recorded for patients in the RAISE study but not for patients in the MG0009 study. Therefore, all patients identified as refractory in the RAISE-XT study had entered the study from the RAISE study and had been classified as gMG refractory at baseline in the RAISE study. Patients were not reassessed for refractory status or redefined after entry into the RAISE-XT study.

As in the RAISE study, eligible patients must also have been vaccinated against Neisseria meningitidis. Patients from a concurrent clinical trial that involved an experimental therapeutic intervention, except for a prior zilucoplan trial, were excluded. Participation in observational studies and registry studies was permitted. Patients were excluded if they had commenced any disallowed medication, per the exclusion criteria from the parent zilucoplan study, or received an altered dose of any other concomitant medication, unless medically indicated. In addition, patients with any new or worsening medical conditions (since entry into the parent zilucoplan study) or hypersensitivity to zilucoplan, any of its excipients, or to the placebo were excluded. Patients who discontinued the study drug in the RAISE study for any reason before the week 12 visit were not eligible for the RAISE-XT study.

Interventions

The visit schedule and assessments during the first 12 weeks of the RAISE-XT study were identical to the 12-week, double-blind treatment period of the RAISE parent study. All patients in the RAISE-XT study received daily self-administered SC doses of zilucoplan 0.3 mg/kg, starting at the extension day 1 visit (E1), regardless of treatment received in the parent studies (a small number of patients received zilucoplan 0.1 mg/kg in the phase II MG0009 study). Zilucoplan was administered as described in the RAISE study. During the first 12 weeks of the RAISE-XT study, patients were to return to the clinic at week E1, week E2, week E4, week E8, and week E12 for study assessments.

Concomitant medications that were permitted in the RAISE-XT study were the same as those permitted in the RAISE study (i.e., doses of standard-of-care treatments for gMG that could not be increased during the study but could be decreased, if deemed to be a reasonable course of action by the investigator, and rescue therapy as needed).

Outcomes

The primary outcome of the RAISE-XT study was the incidence of TEAEs, including any TEAEs, serious TEAEs, withdrawal due to TEAEs, AEs leading to death, and TEAEs of interest. AEs were classified for severity according to CTCAE version 5.0 or later. For any AEs for which it was not possible to provide a CTCAE grading, the events were assessed using a standard intensity classification (grade 1 for mild, grade 2 for moderate, and grades 3, 4, and 5 for severe). A TEAE was defined as an AE that started on or after the time of first administration of the study drug and up to and including 40 days after the final dose (or last contact, depending on which occurred first). Secondary efficacy outcomes included change from baseline in the RAISE-XT study to week E12 in the MG-ADL score, the QMG score, the MGC score, and the MG-QoL15r score. Exploratory outcomes in the RAISE-XT study included minimal manifestation status per MGFA-PIS at week E12 without rescue therapy, responder rates for the MG-ADL, QMG, and MGC scores at week E12 without rescue therapy, and the achievement of MSE (an MG-ADL score of 0 or 1) at week E12 without rescue therapy.

Statistical Analysis

The plan was to enrol approximately 200 patients in the RAISE-XT study from the MG0009 and RAISE studies. Assuming patients remained in the study for an average of 2 years, this would provide approximately 400 patient-years for the zilucoplan safety database. The safety evaluation was conducted on the safety set (all patients who received at least 1 dose of the study drug in the RAISE-XT study). AEs were summarized using descriptive statistics (number, percentage of patients, frequency of AEs) by system organ class and preferred term. AEs that started before the date of first administration of the study treatment were not considered TEAEs. For all safety assessments, baseline was the open-label baseline, defined as the last available assessment before first administration of the study drug in the open-label extension periods of the MG0009 study and the RAISE study. For patients coming from the MG0009 study, this was the last available assessment before the first study drug administration in the open-label extension portion of the MG0009 study. For patients coming from the RAISE study, this was the last available assessment before the first study drug administration in the RAISE-XT study. When dates were missing or partially missing, AEs were assumed to be treatment-emergent unless evidence existed indicating otherwise.

TEAEs were tabulated and presented as the number of patients, percentage of patients, and frequency of events. The number and percentage of patients who experienced each TEAE of interest were summarized by treatment group. Change from baseline to week E12 in scores of the MG-ADL, QMG, MGC, and MG-QoL15r were estimated using an MMRM ANCOVA, with continuous items (baseline MG-ADL score, baseline QMG score) and categorical items (geographic region [Europe, Japan, North America], parent study factor, and baseline MG-ADL score by visit [interaction term]) as fixed effects, and the patient as a random effect. Separate models in the MMRM ANCOVA were fitted for week 1 to week 12 (double-blind treatment period) and week E1 to week E12 (open-label treatment period). The MMRM used an unstructured covariance structure for the repeated measures, such that if the model failed to converge, an autoregressive (1) covariance structure was used. Kenward-Roger approximation was used to estimate the degrees of freedom of the denominator. The LS mean for each treatment group were reported for all visits, with corresponding 2-sided 95% CIs, and plots from week 1 to week E12 were overlaid on the same graph. Total score and change from baseline scores in the MG-ADL were summarized by treatment group and overall, and by scheduled visit, using descriptive statistics. No imputation methods were used for missing total scores on the MG-ADL, QMG, MGC, or MG-QoL15r. Data after the receipt of rescue medication were not imputed.

In the RAISE-XT study, the mITT population included all enrolled patients who received at least 1 dose of the study drug and who had at least 1 postdosing MG-ADL score. The safety set included all patients who received at least 1 dose of the study drug.

The interim analyses reported efficacy results through week E12, with assessments up to week E48 (September 2022 data cut-off), week E84 (May 2023 data cut-off), and week E108 (November 2023 data cut-off). A Clinical Study Report was provided for the September 2022 data-cut-off only. Sponsor-submitted information for the overall gMG population for the September 2022 data cut-off was available for baseline patient characteristics, baseline gMG medications, concomitant gMG therapies, and patient exposure; information on patient disposition and treatment exposure in the overall population was also available and was presented using the longest follow-up available (November 2023 data cut-off). Results of efficacy and harms were presented for the longest follow-up available (November 2023 data cut-off) for both the overall population and the refractory subpopulation.

Results

Patient Disposition

Of the 200 patients enrolled in the RAISE-XT study, 93 patients had received zilucoplan 0.3 mg/kg and 90 patients had received placebo in their parent study. At the November 11, 2023, data cut-off, most patients in the intention-to-treat population were ongoing in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group (79%) and in the placebo and zilucoplan 0.3 mg/kg group (64%) (Table 18). The number of patients who had discontinued the study was 20 (22%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 32 (36%) in the placebo and zilucoplan 0.3 mg/kg group. The primary reasons for discontinuation in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo and zilucoplan 0.3 mg/kg group were voluntary withdrawal by the patient (1% and 13%), AEs (2% and 8%), physician decision (2% and 7%), death (4% and 2%), safety reasons (1% and 4%), other reasons (1% and 0%), and missing reason (0% and 1%).

Table 18: Patient Disposition in the RAISE-XT Study (Overall ITT Population, November 2023 Data Cut-Off)

E = extension; ITT = intention to treat; mITT = modified intention to treat.

Sources: RAISE-XT Clinical Study Report,⁴⁵ sponsor’s Summary of Clinical Evidence.²⁷

Baseline Characteristics

Baseline characteristics of the patients in the RAISE-XT study were available for the September 8, 2022, data cut-off (Table 19). In the overall gMG population, patients in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo and zilucoplan 0.3 mg/kg group were similar in mean age (53 years and 54 years), proportion of female participants (56% and 53%) and male participants (44% and 47%), and proportion of patients enrolled from North America (57% and 54%).

Table 19: Baseline Characteristics of Patients in the RAISE-XT Study (Overall ITT Population, September 2022 Data Cut-Off)

BMI = body mass index; ITT = intention to treat; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGFA = Myasthenia Gravis Foundation of America; QMG = Quantitative Myasthenia Gravis; SD = standard deviation.

Sources: RAISE-XT Clinical Study Report,⁴⁵ sponsor’s Summary of Clinical Evidence.²⁷

Baseline Medications

Baseline medications were defined as any medications that started before the RAISE-XT study and continued thereafter (classified as prior and concomitant medications). A total of 89 patients (97%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 85 patients (95%) in the placebo and zilucoplan 0.3 mg/kg group were receiving any gMG-specific medication at baseline (Table 20). The most common (≥ 10% of patients overall) gMG medications in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo and zilucoplan 0.3 mg/kg group were pyridostigmine (81% and 78%), prednisone (43% and 39%), mycophenolate mofetil (19% and 21%), prednisolone (20% and 18%), and azathioprine (16% and 17%). No patient with gMG received IVIg, SCIg, PLEX, or rituximab at baseline.

Table 20: Baseline gMG-Specific Medications in the RAISE-XT Study (Overall ITT Population, September 2022 Data Cut-Off)

gMG = generalized myasthenia gravis; Ig = immunoglobulin; ITT = intention to treat; SC = subcutaneous.

Note: Baseline medications include any medications that started before the study drug in the MG0011 study and continued after (classified as prior and concomitant medications).

Source: RAISE-XT Clinical Study Report.⁴⁵

Concomitant Medications and Cointerventions

As of the September 2022 data cut-off date, 90 patients (97%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 86 patients (96%) in the placebo and zilucoplan 0.3 mg/kg group received any concomitant gMG therapy (Table 21). The most common (≥ 10% of patients overall) concomitant gMG therapies in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo and zilucoplan 0.3 mg/kg group were pyridostigmine (83% and 82%), prednisone (46% and 40%), prednisolone (22% and 20%), mycophenolate mofetil (19% and 21%), and azathioprine (16% and 17%). The use of concomitant gMG therapies was generally balanced between the treatment groups. Overall, the rate of patients receiving concomitant-only gMG therapies (i.e., medication that had started after the first administration of study drug and continued during the treatment period) was lower than that of patients receiving concomitant gMG therapies (i.e., medication that had been taken at least once after the first administration of the study drug during the treatment period) (Table 21). The most common (≥ 10% of patients overall) concomitant-only gMG therapies in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo and zilucoplan 0.3 mg/kg group were prednisone (23% and 22%), pyridostigmine (16% and 13%), and prednisolone (12% and 10%). The use of concomitant-only gMG therapies was generally balanced between the treatment groups.

Table 21: Concomitant gMG Therapies in the RAISE-XT Study (Overall ITT Population, September 2022 Data Cut-Off)

gMG = generalized myasthenia gravis; ITT = intention to treat.

^aConcomitant-only gMG therapies were defined as any medication that started after the first administration of the study drug and continued during the treatment period.

Sources: RAISE-XT Clinical Study Report,⁴⁵ sponsor’s Summary of Clinical Evidence.²⁷

Exposure to Study Treatments

At the clinical data cut-off date of November 11, 2023, the mean duration of study drug was 892 days (SD = 428 days) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 743 days (SD = 428 days) in the placebo and zilucoplan 0.3 mg/kg group (Table 22). The mean duration of exposure to treatment was 2.5 years (SD = 1.2 years) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 2.1 years (SD = 1.1 years) in the placebo and zilucoplan 0.3 mg/kg group. The number of patients who missed at least 1 dose of the study drug was 34 patients in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 37 patients in the placebo and zilucoplan 0.3 mg/kg group. The mean number of missed doses was 12.2 (SD = 18.6) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 10.0 (SD = 13.2) in the placebo and zilucoplan 0.3 mg/kg group.

Table 22: Patient Exposure in the RAISE-XT Study (Overall Population, Safety Set, November 2023 Data Cut-Off)

E = extension; SD = standard deviation.

Sources: RAISE-XT Clinical Study Report,⁴⁵ sponsor’s Summary of Clinical Evidence.²⁷

Efficacy

Secondary End Points

Secondary efficacy end points in the RAISE-XT study at the data cut-off date of November 11, 2023, included change from parent-study baseline to week E12 in the MG-ADL, QMG, MGC, and MG-QoL15r scores (Table 23); results at 84 weeks in the RAISE-XT study are presented in Table 33 in Appendix 1.

Change From Parent-Study Baseline to Week E12 in the MG-ADL Score

In the overall population, the LS mean change from parent-study baseline to week E12 in the MG-ADL score was –5.90 (SE = 0.47) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –6.17 (SE = 0.59) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the mean change from parent-study baseline to week E12 in the MG-ADL score was –6.61 (SE = 0.63) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –6.24 (SE = 0.71) in the placebo and zilucoplan 0.3 mg/kg group.

Change From Parent-Study Baseline to Week E12 in the QMG Score

In the overall population, the LS mean change from parent-study baseline to week E12 in the QMG score was –8.78 (SE = 0.66) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –8.53 (SE = 0.79) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the mean change from parent-study baseline to week E12 in the QMG score was –8.18 (SE = 0.71) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –8.34 (SE = 1.09) in the placebo and zilucoplan 0.3 mg/kg group.

Change From Parent-Study Baseline to Week E12 in the MGC Score

In the overall population, the LS mean change from parent-study baseline to week E12 in the MGC score was –11.77 (SE = 0.86) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –12.30 (SE = 1.12) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the mean change from parent-study baseline to week E12 in the MGC score was –11.83 (SE = 1.17) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –13.34 (SE = 1.35) in the placebo and zilucoplan 0.3 mg/kg group.

Change From Parent-Study Baseline to Week E12 in the MG-QoL15r Score

In the overall population, the LS mean change from parent-study baseline to week E12 in the MG-QoL15r score was –9.92 (SE = 0.95) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –8.07 (SE = 1.08) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the mean change from parent-study baseline to week E12 in the MG-QoL15r score was –9.46 (SE = 1.15) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and –9.34 (SE = 1.32) in the placebo and zilucoplan 0.3 mg/kg group.

Exploratory End Points

Exploratory end points in the RAISE-XT study at the data cut-off date of November 11, 2023, included the achievement of MSE at week E12 without rescue therapy, the MG-ADL responder rate at week E12 without rescue therapy, the QMG responder rate at week E12 without rescue therapy, and the minimal manifestation status per MGFA-PIS at week E12 without rescue therapy (Table 23).

Achievement of MSE at Week E12 Without Rescue Therapy

In the overall population, the number of patients who achieved MSE (an MG-ADL score of 0 or 1) at week E12 without rescue therapy was 18 (19.4%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 7 (7.8%) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the number of patients who achieved MSE at week E12 without rescue therapy was 8 of 43 (18.6%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 1 of 42 (2.4%) in the placebo and zilucoplan 0.3 mg/kg group.

MG-ADL Responder Rate at Week E12 Without Rescue Therapy

In the overall population, the number of patients who were MG-ADL responders at week E12 without rescue therapy was 71 of 84 (84.5%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 68 of 83 (81.9%) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the number of patients who were MG-ADL responders at week E12 without rescue therapy was 34 of 39 (87.2%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 27 of 36 (75.0%) in the placebo and zilucoplan 0.3 mg/kg group.

QMG Responder Rate at Week E12 Without Rescue Therapy

In the overall population, the number of patients who were QMG responders at week E12 without rescue therapy was 66 of 82 (80.5%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 59 of 82 (72.0%) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the number of patients who were QMG responders at week E12 without rescue therapy was 28 of 38 (73.7%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 25 of 36 (69.4%) in the placebo and zilucoplan 0.3 mg/kg group.

Minimal Manifestation Status per MGFA-PIS at Week E12 Without Rescue Therapy

In the overall population, the number of patients who achieved minimal manifestation status per MGFA-PIS at week E12 without rescue therapy was 29 of 77 (37.7%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 29 of 78 (37.2%) in the placebo and zilucoplan 0.3 mg/kg group. In the refractory subpopulation, the number of patients who achieved minimal manifestation status per MGFA-PIS at week E12 without rescue therapy was not reported.

Table 23: Key Efficacy Results in the RAISE-XT Study (Overall mITT Population and Refractory Subpopulation, November 2023 Data Cut-Off)

CI = confidence interval; E = extension; LS = least squares; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA-PIS = Myasthenia Gravis Foundation of America Post-Intervention Status; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item – Revised; mITT = modified intention to treat; MSE = minimal symptom expression; NR = not reported; QMG = Quantitative Myasthenia Gravis; SE = standard error.

^aA MG-ADL responder is defined as having at least a 3-point improvement (decrease) in the MG-ADL score.

^bA QMG responder is defined as having at least a 5-point improvement (decrease) in the QMG score. This end point was not adjusted for multiple testing, so there is a potential for the type I error to be increased, and the findings should be considered supportive.

Sources: RAISE-XT Clinical Study Report,⁴⁵ sponsor’s Summary of Clinical Evidence.²⁷

Harms

The primary end point of the RAISE-XT study was the incidence of TEAEs. The duration of follow-up for harms was from baseline to week 12 of the RAISE-XT study, with a safety follow-up visit 40 days after the last dose of the study drug. Harms data in the RAISE-XT study for the data cut-off of November 11, 2023, is summarized in Table 24.

Adverse Events

The number of patients in the overall population who experienced at least 1 TEAE was 89 of 93 patients (95.7%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 88 of 90 (97.8%) in the placebo and zilucoplan 0.3 mg/kg group. The most common TEAEs occurring in 10% of patients or greater in either the zilucoplan 0.3 mg/kg and 0.3 mg/kg group or the placebo and zilucoplan 0.3 mg/kg group were MG (29% and 29%), COVID-19 (39% and 31%), headache (19% and 22%), nasopharyngitis (23% and 17%), arthralgia (20% and 13%), diarrhea (19% and 13%), fatigue (18% and 13%), nausea (15% and 17%), upper respiratory tract infection (14% and 18%), urinary tract infection (16% and 14%), pain in extremity (16% and 10%), cough (12% and 12%), fall (11% and 10%), back pain (11% and 10%), rash (10% and 10%), vomiting (11% and 6%), injection-site bruising (4% and 11%), and oropharyngeal pain (1% and 10%).

The number of patients in the refractory subpopulation who experienced at least 1 TEAE was 82 of 85 patients (96.5%) in the zilucoplan 0.3 mg/kg group (the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and the placebo and zilucoplan 0.3 mg/kg group combined). The most common TEAEs occurring in 10% of patients or greater in the refractory subpopulation in either the zilucoplan 0.3 mg/kg and 0.3 mg/kg group or the placebo and zilucoplan 0.3 mg/kg group were MG (42% and 38%), COVID-19 (44% and 24%), headache (21% and 21%), arthralgia (26% and 10%), nasopharyngitis (16% and 19%), urinary tract infection (19% and 14%), upper respiratory tract infection (19% and 12%), diarrhea (16% and 12%), nausea (12% and 17%), fall (14% and 12%), back pain (14% and 12%), and pain in extremity (16% and 10%).

Serious Adverse Events

The number of patients in the overall population who experienced at least 1 serious TEAE was 41 (44.1%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 33 (36.7%) in the placebo and zilucoplan 0.3 mg/kg group. Serious TEAEs occurring in 2% of patients or greater in either the zilucoplan 0.3 mg/kg and 0.3 mg/kg group or the placebo and zilucoplan 0.3 mg/kg group were MG (9.7% and 11.1%), COVID-19 pneumonia (4.3% and 1.1%), myocardial infarction (4.3% and 0.0%), pneumonia (3.2% and 1.1%), cholecystitis (2.2% and 1.1%), Staphylococcus bacteremia (2.2% and 0.0%), atrial fibrillation (2.2% and 0.0%), cardiac arrest (2.2% and 0.0%), cellulitis (2.2% and 2.2%), and large intestine polyp (0.0% and 2.2%). The number of patients in the refractory subpopulation who experienced at least 1 serious TEAE was 20 of 43 (46.5%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 15 of 42 (35.7%) in the placebo and zilucoplan 0.3 mg/kg group.

Withdrawal Due to Adverse Events

TEAEs resulting in permanent withdrawal from the study drug in the overall population were reported in 21 patients (11.4%), 9 (9.7%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 12 (13.3%) in the placebo and zilucoplan 0.3 mg/kg group. The most common TEAE resulting in permanent withdrawal from the study drug was MG (6 patients [3.3%]). The number of patients in the refractory subpopulation who experienced at least 1 serious TEAE was 11 of 85 (12.9%), 2 of 43 (4.7%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 9 of 42 (21.4%) in the placebo and zilucoplan 0.3 mg/kg group.

Deaths

A total of 6 patients died in the RAISE-XT study. The number of patients who experienced a TEAE leading to death was 3 (3.2%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 1 (1.1%) in the placebo and zilucoplan 0.3 mg/kg group. In the zilucoplan 0.3 mg/kg and 0.3 mg/kg group, the TEAEs leading to death were cardiac arrest (2 patients [2.2%]) and head injury (1 patient [1.1%]). In the placebo and zilucoplan 0.3 mg/kg group, the TEAE leading to death was death (1 patient [1.1%]). All 4 TEAEs leading to death resulted in permanent withdrawal from the study drug. Two patients in the refractory subpopulation died during the study; 1 patient in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group had a TEAE of death. No details were reported for the deaths in the refractory subpopulation.

AEs of Special Interest

The number of patients who experienced any infections was 67 (72.0%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 65 (72.2%) in the placebo and zilucoplan 0.3 mg/kg group. Of these, the number of patients who experienced serious infections was 16 (17.2%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 11 (12.2%) in the placebo and zilucoplan 0.3 mg/kg group. The number of patients in the refractory subpopulation who experienced any infections was 34 of 43 (79.1%) in the zilucoplan 0.3 mg/kg and 0.3 mg/kg group and 33 of 42 (78.6%) in the placebo and zilucoplan 0.3 mg/kg group.

Table 24: Summary of Harms Results in the RAISE-XT Study (Overall Population and Refractory Subpopulation, Safety Set, November 2023 Data Cut-Off)