Reimbursement Review

Ravulizumab (Ultomiris)

Sponsor: Alexion Pharma GmbH

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 on Application Submitted for Review

AChEI = acetylcholinesterase inhibitor; AChR = acetylcholine receptor; gMG = generalized myasthenia gravis; NOC = Notice of Compliance; NSIST = nonsteroidal immunosuppressive therapy; q.8.w. = every 8 weeks.

Introduction

Myasthenia gravis (MG) is a chronic autoimmune disease in which antibodies against acetylcholine receptors (AChRs) or functionally associated molecules in the neuromuscular junction disrupt nerve impulse conduction, resulting in localized or generalized skeletal muscle weakness.^1-4 In 10% to 15% of all patients with MG, weakness remains restricted exclusively to the eyes (ocular MG).^1,3,5,6 The remainder of patients either are diagnosed with or progress within a few years to generalized myasthenia gravis (gMG), and their symptoms include eyelid drooping and double vision, altered facial expression, difficulty chewing and swallowing food, difficulty speaking, and in patients with more severe disease, problems with limb movement and breathing.^3,6 Collectively, symptoms of MG negatively impact patients’ health-related quality of life (HRQoL).³ Globally, the prevalence rates of MG range from 150 cases to 200 cases per 1 million population, and the incidence rates range from 4 cases to 30 cases per million person-years.^3,7 In Canada, the prevalence of MG is approximately 263 cases to 320 cases per 1 million population and its incidence is approximately 23 cases per 1 million population annually.⁸ Autoantibodies against AChR can be detected in approximately 80% of patients with gMG.³ According to the clinical expert consulted by Canada’s Drug Agency (CDA-AMC) for this review, 15% to 20% of patients with gMG have refractory disease. The prognosis of MG varies by symptom severity and response to treatment. Patients with severe or refractory symptoms are at higher risk for complications compared with those with mild or nonbulbar symptoms. MG may be associated with increased mortality.⁹

According to the clinician expert, the goal of treatment in most patients with gMG is to reduce disease symptoms (ideally aiming for minimal disease manifestation) as well as adverse effects of MG therapy and to allow the patient to function and work normally with good HRQoL. Other goals of treatment include avoiding MG exacerbations and myasthenic crisis, minimizing hospitalizations and intensive care unit admissions, and reducing the numbers and doses of therapies required for symptom control.

The initial symptomatic treatment for most patients with gMG is an acetylcholinesterase (AChE) inhibitor such as pyridostigmine. According to the expert, peripherally acting AChE inhibitors are used as symptomatic treatments for temporarily alleviating muscle weakness in these patients. A minority of patients with MG have a spontaneous remission or respond to AChE inhibitors; however, most patients need treatment with corticosteroids and/or steroid-sparing drugs when they do not reach their treatment goals with AChE inhibitors.

Corticosteroids are the first-line immunosuppressive therapy (IST) for patients with MG who remain symptomatic while on AChE inhibitors or those who desire better symptom control. Early use of oral steroids in patients with pure ocular symptoms may delay or reduce the risk of generalization and worsening of the underlying symptoms.^10,11 Patients who do not respond to corticosteroids or who have comorbidities that are not suitable for treatment with corticosteroids may initiate treatment with nonsteroidal immunosuppressive therapies (NSISTs).¹² The use of conventional therapy (AChE inhibitors, corticosteroids, and/or NSISTs) is associated with the slow onset of benefit, incomplete efficacy in eliminating MG symptoms, and significant toxicities.^13-15 Other treatment options, such as immunomodulating therapies like plasma exchange or plasmapheresis (PLEX) and IV immunoglobulin (IVIg), are usually reserved for use on a short-term basis for life-threatening situations or crises, for use in some patients before surgery or corticosteroid treatment, to obtain a rapid treatment response, or for when treatments have been ineffective.¹⁰ Novel biologic treatments include efgartigimod alfa, eculizumab, and rituximab.^12,16-21 Surgery (thymectomy) is a treatment option in select circumstances.²⁰ Thymectomy is recommended in nonthymomatous, anti–AChR antibody–positive adult patients with gMG aged 50 years or younger if they are stable (as an early treatment option), as well as in patients with or without AChR antibodies if ISTs (which include both corticosteroids and NSISTs) elicit an insufficient response or are intolerable due to side effects.¹²

Ravulizumab is a terminal complement inhibitor that specifically binds to the complement protein C5 with high affinity, thereby inhibiting its cleavage to C5a and C5b and preventing the generation of membrane attack complex.²² In January 2023, ravulizumab was approved by Health Canada for the treatment of adult patients with anti–AChR antibody–positive gMG.²² It is supplied as a 10 mg/mL or 100 mg/mL concentrate for IV infusion. The recommended ravulizumab maintenance dosing in adult patients with gMG (aged 18 years or older) with a body weight greater than or equal to 40 kg is based on the patient’s body weight with maintenance doses administered every 8 weeks, starting 2 weeks after the loading dose. Patients must be vaccinated against meningococcal infections before, or at the time of, initiating ravulizumab unless the risks of delaying ravulizumab therapy outweigh the risks of developing a meningococcal infection.²² Ravulizumab was initially reviewed by CDA-AMC for the treatment of adults with anti–AChR antibody–positive gMG and received a negative funding recommendation from the Canadian Drug Expert Committee (CDEC) on August 24, 2023.²³ The drug was resubmitted for review by the sponsor on the basis of the availability of new evidence from the pivotal CHAMPION-MG trial: the efficacy and harms of ravulizumab in 2 post hoc subgroups, the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort; additional open-label extension (OLE) follow-up data (with the final analysis representing up to 3.5 years of follow-up duration); and a new indirect treatment comparison (ITC). In this resubmission, the sponsor’s reimbursement request is as add-on therapy for adult patients with anti–AChR antibody–positive gMG whose symptoms persist despite adequate treatment with AChE inhibitors, corticosteroids, and/or NSISTs.

Perspectives of Patients, Clinicians, and Drug Programs

The information in this section is a summary of input provided by the patient and clinician groups that responded to the call for input by CDA-AMC and from clinical experts consulted by CDA-AMC for the purpose of this review. The full patient and clinician group submissions received by CDA-AMC are available in the consolidated patient and clinician group input document for this review on the project website.

Patient Input

CDA-AMC received 1 patient group submission from Muscular Dystrophy Canada (MDC). MDC is a health charity that supports people affected by muscular dystrophies and related muscle diseases in Canada. MDC collected information from 215 patients impacted by MG through a health care experience survey and semistructured phone or virtual interviews. These patients consisted of 83 males and 132 females aged between 22 years and 78 years from all provinces in Canada. MDC also conducted an MG Canadian journey mapping project among patients living with MG via the completion of virtual interviews, round table sessions, surveys, and HRQoL measures. Respondents indicated that MG has a significant impact on productivity; fatigue, energy levels, and quality of sleep; respiratory health; mobility and strength; independence; relationships and social participation; eyes, vision, speech, and swallowing; mental health; quality of life; and the well-being of respondents’ families. MDC added that according to the respondents, while supportive treatments have had positive health outcomes, there are concerns about the long-term and sustained benefits of these treatments.

MDC noted that patients with gMG seek better control over their condition to minimize the impact of symptoms, side effects, and disease exacerbations on their lives, allowing them to maintain their independence and avoid serious hospital admissions. MDC added that patients stated that they would be willing to deal with the side effects of medications if these aspects of MG were better controlled. According to MDC, respondents indicated that currently available therapies may decrease MG exacerbations but do not improve their overall HRQoL.

Based on the patient group input, the factors that are important in evaluating different treatments include treatment administration, potential side effects, the duration and frequency of treatments, convenience (e.g., travel time and parking for clinic visits), and financial impact (costs).

MDC explained that patients value new treatments that offer improved disease control and symptom relief, and extend dosing intervals.

In terms of diagnostic testing, MDC stated that 85% of respondents reported significant difficulty in getting diagnosed. The vast majority of patients found the test to be cost-effective but noted the overall lengthy process with many missed opportunities (e.g., delays, misdiagnoses, costs incurred). A diagnosis received as part of hospitalization was reported as a seamless experience.

MDC believed that there is a pressing need for improved treatment options to address the ongoing challenges faced by patients with MG and ravulizumab provides a new treatment option for patients with MG that has demonstrated efficacy, safety, and improved dosing convenience compared to other treatment options.

Clinician Input

Input From Clinical Experts Consulted by the Review Team for This Resubmission

The clinical expert consulted for this review identified unmet needs associated with currently available treatments for patients with gMG whose symptoms persist even if they have been treated with conventional medications for this disease (such as AChE inhibitors, corticosteroids, and/or NSISTs). The unmet needs are as follows: although multiple treatment options are available to patients with gMG, some patients do not have adequate response to the existing treatments and 15% to 20% of patients have refractory disease and require alternative therapeutic options; patients may become intolerant of ISTs; and some exiting treatments are only suitable for select patients. Because of its unique mechanism of action, ravulizumab — a complement C5 inhibitor — could be another treatment option for patients living in Canada with gMG whose disease had an inadequate response to or did not respond to or who cannot tolerate conventional ISTs.

The clinical expert indicated that patients in Canada with significant symptoms of gMG whose disease has not responded to or who cannot tolerate conventional immunosuppressants would be eligible to receive treatment with ravulizumab. The initiation of ravulizumab therapy could be considered when patients are experiencing significant symptoms of gMG; are resistant to conventional therapies, requiring multiple concomitant ISTs (which include corticosteroids and NSISTs); or are patients who cannot tolerate the significant side effects of conventional immunosuppressants.

The expert noted that in clinical practice, regular follow-up visits with a neuromuscular specialist or a neurologist are required to monitor the patient’s response to treatment, using certain MG-specific scales (e.g., the Myasthenia Gravis Activities of Daily Living [MG-ADL]) tool as well as the treating physician’s clinical examination.

The expert also noted that treatment with ravulizumab should be discontinued when a patient does not respond well to the treatment or experiences significant side effects of the treatment such as meningococcal infections, or when a patient prefers to discontinue treatment. In addition, if the patient shows long-term stability of neurologic status and is perceived to have achieved remission, the clinician may suggest holding the treatment under observation.

The expert indicated that the initiation of treatment with ravulizumab and the follow-up assessments could be provided by a neurologist with expertise in MG management.

Clinician Group Input

One input was received from the Neuromuscular Disease Network for Canada (NMD4C), a new pan-Canadian network that brings together clinical, scientific, technical, and patient expertise to improve care, research, and collaboration in neuromuscular disease. The mission of NMD4C is to improve the care, research, and treatment of neuromuscular diseases for all people living in Canada. The information presented in this submission was gathered from 8 clinicians with experience in treating gMG.

NMD4C noted that the goals of therapy are to minimize morbidity and mortality from MG, keep patients out of hospital, improve quality of life, prevent repeated attacks, and prevent prolonged, untreated, or partially treated MG. According to the clinician input, the current mainstay of therapies for MG includes supportive therapies, symptomatic treatments, and disease-modifying strategies. NMD4C provided opinions consistent with the clinical expert consulted for this review regarding the unmet needs with currently available treatments: limited available options for active and refractory disease, limited response to the traditional immunotherapies, the slow onset of treatment action, and serious side effects.

NMD4C believes ravulizumab inhibits immune-mediated damage to the neuromuscular junction rather than being a symptomatic treatment. The place in therapy for ravulizumab would likely be in patients with inadequate MG response: after treatment with pyridostigmine, and after treatment with either steroids and/or other ISTs, including steroid-sparing immunosuppressive drugs. Ravulizumab is likely to affect the treatment paradigm of patients with refractory and nonrefractory MG who are not responsive to first-line and second-line therapies or require chronic IVIg infusions or PLEX. It may also be considered in patients who are intolerant of other immunomodulatory treatments.

NMD4C stated that patients with MG should try other treatments before initiating ravulizumab, which requires periodic IV infusions over an extended period, is unlikely to induce long-term disease remission, is likely to be expensive, will likely not be available in smaller cities or nonspecialized centres, and requires extensive expertise.

Based on the clinician group input, adult patients who are seropositive for anti–AChR antibodies and have gMG would be best suited for treatment with ravulizumab; there are no data on the efficacy of ravulizumab for a minority population of patients with MG — including those aged younger than 18 years, those who had thymectomy within the past year, those with thymic carcinoma or who are pregnant or breastfeeding, those with anti–muscle-specific kinase (anti–MuSK) or anti–LRP4 antibodies, or seronegative patients. Patients with MG who are anti–AChR antibody–positive who have not responded to pyridostigmine and to steroids and/or oral ISTs or are additionally dependent on periodic PLEX or chronic IVIg therapy are most in need.

NMD4C noted that the diagnosis must have been confirmed clinically and supported by confirmatory laboratory tests before treatment with ravulizumab.

NMD4C explained that the outcomes that are used to determine whether a patient is responding to treatment in clinical practice include increased survival, the avoidance of emergency department visits or hospital admissions (including intensive care unit admissions), the avoidance of the need for rescue therapy as well as maintenance therapy with IVIg and plasmapheresis, a reduction in the dose and/or duration of concomitant steroids, a reduction in the level of fatigable weakness, and an improvement in activities of daily living and quality of life.

Based on the NMD4C input, the factors that should be considered when deciding to discontinue treatment include the amount of clinical improvement or response, the duration of time spent in a clinically stable state, adverse events (AEs) associated with the treatments, and inconvenience associated with the therapy.

NMD4C added that an appropriate setting for treatment includes a clinical team with general knowledge of MG, as well as appropriate nursing experience in venous access issues, managing IV medications, and managing potential adverse effects. It is recommended that ravulizumab be prescribed by neurology specialists with expertise in MG diagnosis, assessment, monitoring, and the management of patients with MG.

Drug Program Input

Input was obtained from the drug programs that participate in the CDA-AMC reimbursement review process. Please refer to Table 5 for further information. The following were identified as key factors that could potentially impact the implementation of the CDA-AMC recommendation for ravulizumab:

• 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

• generalizability

• care provision issues

• system and economic issues.

Clinical Evidence

Systematic Review

Description of Studies

The CHAMPION-MG trial (N = 175)²⁴ was a phase III, double-blind, multicentre, placebo-controlled randomized controlled trial (RCT) with an OLE period of up to 4 years. The primary objective of the CHAMPION-MG trial was to evaluate the safety and efficacy of ravulizumab compared with placebo in complement inhibitor–naive adult patients with gMG. The randomized controlled period (RCP) portion of the trial is complete (the data cut-off date was May 11, 2021; the database was locked on June 30, 2021) and the OLE concluded on May 25, 2023 (the last patient’s last visit).

In this resubmission, the sponsor provided new clinical evidence to support its revised reimbursement request: ravulizumab as add-on therapy for adult patients with anti–AChR antibody–positive gMG whose symptoms persist despite adequate treatment with AChE inhibitors, corticosteroids, and/or NSISTs. Post hoc analyses were performed in 2 cohorts, the concomitant IST optimized cohort (N = ███, █████ and the refractory concomitant IST optimized cohort (N = █████████); these cohorts were the focus of this Clinical Review Report. The purpose of the post hoc analyses was to evaluate the efficacy and safety of 26-week treatment with ravulizumab or placebo in patients with anti–AChR antibody–positive gMG. Therefore, the inclusion criteria of the previous Study ADAPT (efgartigimod alfa versus placebo) and Study REGAIN (eculizumab versus placebo) were adopted to select patients who would be considered to have either IST optimization, or IST optimization with refractory disease (patients who do not achieve symptom control after 12 months of treatment with 2 or more ISTs, or 1 or more IST and chronic IVIg or PLEX treatment based on the criteria for eculizumab in the pivotal Study REGAIN) at the time of study enrolment from the CHAMPION-MG trial, respectively. The primary efficacy end point (change from baseline in the MG-ADL score at week 26) was the same as that in the primary analysis in the full population. Other outcomes in these analyses included change from baseline in the Quantitative Myasthenia Gravis (QMG) total score, improvement of at least 5 points in the QMG total score from baseline, change from baseline in the Myasthenia Gravis Quality of Life 15-item Scale - Revised (MG-QoL15r) score, change from baseline in the Quality of Life in Neurological Disorders (Neuro-QoL) fatigue score, improvement of at least 3 points in the MG-ADL total score from baseline, the incidence of clinical deterioration and/or an MG crisis, and safety. In the subgroup population of patients who received optimized IST or patients who were refractory despite having received optimized IST, all outcomes analyzed were exploratory. Baseline demographic and disease characteristics of the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort were consistent with the full CHAMPION-MG trial population in terms of the distribution of the Myasthenia Gravis Foundation of America (MGFA) classification, the baseline MG-ADL score (approximately 9 points across the 3 groups), the baseline QMG score (ranging from 14 points to 15 points across the 3 groups), and age at diagnosis.

Efficacy Results

The evidence examined in this review was informed by 2 post hoc cohorts: the concomitant IST optimized cohort (patients receiving concomitant optimized IST, with or without refractory disease), and the refractory concomitant IST optimized cohort (patients receiving concomitant optimized IST, with refractory disease).

MG-ADL Total Score

The MG-ADL total score ranges from 0 to 24; higher scores indicate a greater severity of symptoms and a more significant impact on a patient’s daily activities. Although no minimal important difference (MID) has been estimated, an improvement of approximately 2 points in the MG-ADL total score represents the minimal clinically important difference (MCID) and is a recommended response threshold in patients with MG.²⁵ This MID was estimated to interpret the change from baseline in an individual patient; an MID for between-group effects has not been estimated. In the concomitant IST optimized cohort, during the RCP, the least squares mean (LSM) (95% confidence interval [CI]) change from baseline to week 26 in the MG-ADL total score was █████████████████ in the ravulizumab group compared to █████████████████ in the placebo group. The LSM treatment difference was ███████████████████████). In the refractory concomitant IST optimized cohort, the LSM (95% CI) change from baseline to week 26 in the MG-ADL total score was ████████████████) in the ravulizumab group compared to ████████████████) in the placebo group. The LSM treatment difference was ███████████████████████). ███████████████████████████████████████████████████████████

QMG Total Score

The QMG total score ranges from 0 to 39, where higher scores indicate greater disease severity. A definite MID for the QMG scale has not been established, although a threshold ranging between 2 points and 3 points depending on disease severity has been suggested for patients with MG,^26,27 which represents the MCID for the QMG scale. This MID was estimated to interpret the change from baseline in an individual patient; an MID for between-group effects has not been estimated. In the concomitant IST optimized cohort, the LSM (standard error of the mean [SEM]) change from baseline to week 26 in the QMG total score was ████ (████) in the ravulizumab group and ████████) in the placebo group. The LSM treatment difference was ████ (████████████████████). In the refractory concomitant IST optimized cohort, the LSM (SEM) change from baseline to week 26 in the QMG total score was ████ (████) in the ravulizumab group and ████ (████) in the placebo group during the RCP. The LSM treatment difference was ███████████████████████). ███████████████████████████████████████████████████████████████████

QMG 5-Point or Greater Response

A clinical responder was defined as a patient who achieved a 5-point or greater improvement in the QMG total score. A 5-point or greater improvement in the QMG total score represents a higher threshold and exceeds the established MCID of 2 points or 3 points for the QMG total score scale. In the concomitant IST optimized cohort, █████ of patients in the ravulizumab group compared to ████ in the placebo group achieved at least a 5-point improvement. The between-group difference was ███████████████████ and the OR was █████ (██████████████████████). In the refractory concomitant IST optimized cohort, the proportion of patients meeting the clinical responder definition for QMG improvement was █████ in the ravulizumab group versus ████ in the placebo group, with a between-group difference of █████ (█████████████) and an OR of █████ (95% █████████████; | | ██████). ████████████████████████████████████████████████████████.

MG-ADL 3-Point or Greater Response

Patients who achieved at least a 3-point improvement in the MG-ADL total score were considered responders. A 3-point or greater improvement in MG-ADL total score represents a higher threshold and exceeds the established MCID of 2 points for the MG-ADL total score scale. In the full analysis set (FAS) of the CHAMPION-MG trial, this outcome was tested after a prior nonsignificant result of the hierarchical testing procedure and therefore is at an increased risk of type I error (falsely rejecting the null hypothesis). In the concomitant IST optimized cohort, the proportion of patients who met the clinical responder definition for MG-ADL improvement of at least 3 points was ████% (███████████) in the ravulizumab group compared to ████% █████████████) in the placebo group, with a between-group difference of ███████████████████ and an OR of █████ (██████████). In the refractory concomitant IST optimized cohort, the proportion of patients meeting the clinical responder definition for MG-ADL improvement was █████ (███████████) in the ravulizumab group compared to ████% (███████████) in the placebo group, with a between-group difference of ███████████████████ and an OR of █████ (██████████). ██████████████████████████████████████████████████

MG-QoL15r Total Score and Neuro-QoL Fatigue Score

HRQoL and fatigue were assessed based on LSM change from baseline in the MG-QoL15r total score and Neuro-QoL fatigue score. An MID for MG-QoL15r or Neuro-QoL fatigue in patients with MG has not been estimated. Since the MG-QoL15r outcome did not reach statistical significance in the overall trial population, the P values for the subsequent secondary end points included in the prespecified hierarchical testing order including the Neuro-QoL fatigue score were considered nominal. The between-group differences in these outcomes for both the primary analysis and the 2 cohorts were consistent. In the concomitant IST optimized cohort, the LSM change from baseline to week 26 in the MG-QoL15r total score was ████ (███████████) in the ravulizumab group and ████ (███████████) in the placebo group during the RCP. The LSM treatment difference was ████ (███████████████████). In the refractory concomitant IST optimized cohort, the LSM change from baseline to week 26 in the MG-QoL15r total score was ████████████████ in the ravulizumab group and ████ (███████████) in the placebo group during the RCP. The LSM treatment difference was ████ (███████████████████). In the concomitant IST optimized cohort, the LSM change from baseline to week 26 in the Neuro-QoL fatigue score was ████ (█████████████) in the ravulizumab group and ████ (███████████) in the placebo group during the RCP. The LSM treatment difference was ████ (███████████████████). In the refractory concomitant IST optimized cohort, the LSM change from baseline to week 26 in the Neuro-QoL fatigue score was -███ (████████████) in the ravulizumab group and ████ (████████████) in the placebo group during the RCP. The LSM treatment difference was ████ (███████████████████). █████████████████████████████████████████████████████████

Incidence of Clinical Deterioration and MG Crisis

During the RCP, in the concomitant IST optimized cohort, █████████████ in the ravulizumab group and ███████████████ in the placebo group reported clinical deterioration. In the refractory concomitant IST optimized cohort, ████████ (██████ in the ravulizumab group and ██████████████ in the placebo group reported clinical deterioration. In each cohort, an MG crisis was reported by || patient from the placebo group. No patients from the ravulizumab group reported an MG crisis.

Harms Results

███████████████████████████████████████████████████████████████████████████

The percentage of patients with any AEs was █████ (██ patients) of placebo-treated patients and █████ (██ patients) of ravulizumab-treated patients in the concomitant IST optimized cohort and █████ (██ patients) of placebo-treated patients and █████ (██ patients) of ravulizumab-treated patients in the refractory concomitant IST optimized cohort. The most commonly reported AEs in these 2 cohorts included diarrhea (concomitant IST optimized cohort = █████ in the ravulizumab group versus █████ in the placebo group; refractory concomitant IST optimized cohort = █████ in the ravulizumab group versus █████ in the placebo group) and headache (concomitant IST optimized cohort = █████ in the ravulizumab group versus █████ in the placebo group; refractory concomitant IST optimized cohort = █████ in the ravulizumab group versus █████ in the placebo group). Most AEs were grade 1 and grade 2 in severity.

In the concomitant IST optimized cohort, the percentage of patients with serious adverse events (SAEs) was higher in the ravulizumab group (█████) compared to the placebo group (█████). In the refractory concomitant IST optimized cohort, █████ of ravulizumab-treated patients and █████ of patients in the placebo group experienced at least 1 SAE. Commonly reported SAEs were not reported for these 2 cohorts.

In both cohorts, there was ████████ treated with ravulizumab who discontinued the study drug due to an AE. In both cohorts, there were █████████ treated with ravulizumab who died, and ██ death was reported for the placebo group. Meningococcal infection was considered a notable harm for treatment with ravulizumab. No events of meningococcal infection were reported during the RCP in the entire study population.

Table 2: Summary of Key Results From the Randomized Controlled Period of CHAMPION-MG Trial, Cohort Analyses

AE = adverse event; CI = confidence interval; FAS = full analysis set; IST = immunosuppressive therapy; LSM = least squares mean; MG-ADL = Myasthenia Gravis Activities of Daily Living; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item Scale - Revised; MMRM = mixed model of repeated measures; Neuro-QoL = Quality of Life in Neurological Disorders; OR = odds ratio; QMG = Quantitative Myasthenia Gravis; SAE = serious adverse event; SD = standard deviation; SEM = standard error of the mean; vs. = versus; WDAE = withdrawal due to adverse event.

^aAll efficacy outcomes in Table 2 analyzed in the full population were multiplicity-adjusted using a hierarchical testing strategy. However, none of the P values in the cohort analyses were multiplicity-adjusted.

Sources: Clinical Study Report for the CHAMPION-MG trial,²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Critical Appraisal

This is a resubmission of the initial ravulizumab review. In the previous review on ravulizumab, CDEC issued a negative reimbursement recommendation in 2023 based on the evidence submitted to CDA-AMC. The current review focuses on 2 post hoc cohorts that were identified from the FAS in the CHAMPION-MG trial: a concomitant IST optimized cohort and a refractory concomitant IST optimized cohort. The criteria and definitions used for patient selection for these 2 cohorts were considered reasonable and acceptable in clinical practice, according to the clinical expert consulted for this review. Based on the patient characteristics at baseline (which were similar to those reported in the full population) and the patients’ previous IST treatment history, the selected cohorts of patients represented a heavily pretreated patient population that, despite a lengthy duration of IST treatment, continued to experience significant gMG symptoms. Theoretically, post hoc analyses that are not prespecified in a trial’s analysis plan can be at risk of bias due to selective reporting. However, the sponsor provided clear justification for the selected subpopulations using criteria informed by the other gMG trials to provide directly relevant information related to a previous negative reimbursement recommendation. Additionally, the same prespecified analysis methods of the CHAMPION-MG trial were used to analyze the subgroup data, and results for all relevant end points were presented. As a result, any concern for selective reporting is minimized. In these 2 subgroups, patients’ baseline demographic and disease characteristics were generally well balanced between treatment groups and were similar to the FAS population. Also similar to the FAS, minor imbalances were observed for ████████████████████ (██████████████████████████████████████████████████████████████████████████████████████████████████) and ████████████████████████. The imbalances in the FAS could be due to the small sample size, which would have been exacerbated in the smaller subgroups. The clinical expert consulted by the review team noted that these may not have a significant impact on result interpretation. Tests for subgroup differences between the subgroup populations and the rest of the FAS were performed for the MG-ADL score and QMG score. However, these analyses were post hoc and the CHAMPION-MG trial was not powered to find a difference between the groups. Otherwise, results from these post hoc subgroups can be interpreted as having consistency with the main analyses of the CHAMPION-MG study and sharing the same limitations of those analyses. In addition, it should be considered that the small sample size in these 2 subgroups (the concomitant IST optimized cohort accounted for ███ of the full population and the refractory concomitant IST optimized cohort was ████ of the full population) could have resulted in insufficient power to detect true between-group differences, and multiplicity was not controlled using hierarchical testing. Therefore, there was an increasing risk of type I error (i.e., falsely rejecting the null hypothesis) rate for the investigated outcomes that achieved statistical significance at a conventional alpha of 0.05.

The reimbursement request also includes patients who have had an adequate trial of AChE inhibitors and no ISTs; these patients are not included in the 2 post hoc cohorts but were studied in the FAS in the CHAMPION-MG trial. A similar proportion of patients who received AChE inhibitors but no ISTs were enrolled in the CHAMPION-MG and ADAPT trials (i.e., approximately 10% of the overall trial populations). The NMD4C clinician group and the clinical expert consulted for this review agreed that while the place in therapy for ravulizumab would also include patients who had received AChE inhibitors but not ISTs, it would be reasonable for patients with gMG to try other treatments before initiating ravulizumab, which requires periodic IV infusions over an extended period, is likely to be expensive, may not be available in smaller cities or nonspecialized centres, and requires extensive expertise.

Patient selection for these 2 subgroups was based on the inclusion criteria from other RCTs: Study ADAPT and Study REGAIN. According to the clinical expert consulted for this review, the definitions used to identify patients are reasonable and adequately reflect the patients who experience unmet needs in the treatment of gMG in clinical settings in Canada (i.e., patients with refractory gMG as well as those whose disease showed response to a prior gMG treatment but responded inadequately to existing standards of gMG therapy). Furthermore, based on the duration of prior ISTs and concomitant ISTs used in the 2 subgroups before enrolment in the CHAMPION-MG trial, the clinical expert consulted for this review considered patients in these subgroups to have entered the CHAMPION-MG trial on optimized ISTs.

Because the concomitant conventional therapy was required to remain stable during the RCP, except in the case of rescue therapy, ravulizumab was not compared to any individual or combination conventional therapy as it would typically be used in clinical practice (i.e., altering doses or adding additional medications to suit patients’ current symptoms or other needs). Similar protocol requirements for concomitant ISTs are common across trials in patients with gMG to ensure consistency and to prevent the confounding of trial results. Notably in Study ADAPT (efgartigimod alfa versus placebo) and Study REGAIN (eculizumab versus placebo), a change in the type or dose of concomitant conventional care was not allowed unless deemed medically necessary.

The RCP of the CHAMPION-MG trial was 26 weeks. The longer-term treatment effect of ravulizumab can only be assessed in the OLE period of this study. Moreover, the CHAMPION-MG trial did not provide evidence for the comparisons between ravulizumab and other currently available active treatments for gMG.

Long-Term Extension Studies

Description of Studies

One long-term extension study is summarized here to provide evidence regarding the long-term efficacy and safety of ravulizumab in patients with gMG. In the current resubmission, OLE data from the pivotal study of up to 3.5 years is available. At the end of the 26-week RCP, all patients were eligible to enter the OLE and receive open-label ravulizumab. Following the 26-week RCP, patients in the placebo group received a blinded loading dose of ravulizumab and patients in the ravulizumab group received a blinded ravulizumab dose of 900 mg. Starting week 28, all patients began open-label ravulizumab maintenance doses every 8 weeks.

Efficacy Results

In the OLE set in the ravulizumab-ravulizumab group, the LSM change from RCP baseline in the MG-ADL total score was −4.0 (95% CI, −5.3 to −2.8) at week 164 during the OLE period. In the placebo-ravulizumab group, the LSM change from RCP baseline was −3.6 (95% CI, −4.8 to −2.3) at week 164. In the ravulizumab-ravulizumab concomitant IST optimized cohort, the LSM change from OLE baseline in the MG-ADL total score was ███████████████ at week 28 and ████████████████ at week 164. In the placebo-ravulizumab concomitant IST optimized cohort, the LSM change from OLE baseline was █████████████████ at week 28 and █████████████████ at week 148. In the ravulizumab-ravulizumab refractory concomitant IST optimized cohort, the LSM change from OLE baseline in the MG-ADL total score was ███████████████ at week 28 and ████████████████ at week 148. In the placebo-ravulizumab refractory concomitant IST optimized cohort group, the LSM change in the MG-ADL total score from OLE baseline was █████████████████ at week 28 and █████████████████ at week 148. In the OLE set, in the ravulizumab-ravulizumab and placebo-ravulizumab groups, the LSM change from RCP baseline in the QMG total score was █████████████████ and █████████████████ at week 164, respectively. In the ravulizumab-ravulizumab and placebo-ravulizumab groups of the concomitant IST optimized cohort, the LSM change from OLE baseline in the QMG total score was ████████████████ at week 164 and █████████████████ at week 148, respectively.

In the ravulizumab-ravulizumab and placebo-ravulizumab groups of the refractory concomitant IST optimized cohort, the LSM change from OLE baseline in the QMG total score was ███████████████ and █████████████████ at week 148, respectively. In the OLE set, based on a 5-point or greater improvement in the QMG total score from the RCP baseline, the proportion of clinical responders in the ravulizumab-ravulizumab group and placebo-ravulizumab group was ██████████████████ and █████████████████ at week 164, respectively.

In the ravulizumab-ravulizumab and placebo-ravulizumab groups of the concomitant IST optimized cohort, the proportion of patients with a 5-point or greater improvement in the QMG total score from the RCP baseline was ██████████████████ and █████████████████) at week 164, respectively.

In the ravulizumab-ravulizumab and placebo-ravulizumab groups of the refractory concomitant IST optimized cohort, the proportion of patients with a 5-point or greater improvement in their QMG total score from the RCP baseline was ██████████████████ and ██████████████████ at week 164, respectively.

In the OLE set, in the ravulizumab-ravulizumab and placebo-ravulizumab groups, the LSM change from RCP baseline in the MG-QoL15r total score was ████████████████ and █████████████████ at week 164, respectively.

In the OLE set, in the ravulizumab-ravulizumab and placebo-ravulizumab groups, the LSM change from RCP baseline in the Neuro-QoL fatigue score was ███████████████████ and ███████████████████ at week 164, respectively.

In the OLE set, based on a 3-point or greater improvement in the MG-ADL total score from RCP baseline, the proportion of clinical responders in the ravulizumab-ravulizumab and placebo-ravulizumab groups was ██████████████████ and ██████████████████ at week 164, respectively.

Harms Results

During ravulizumab treatment, 96.4% of patients experienced at least 1 AE. The most commonly reported AEs (≥ 10% of total patients) in the ravulizumab-treated set (RTS) were COVID-19 (36.1%), headache (23.1%), diarrhea (17.2%), arthralgia (13.6%), back pain (13%), nausea (13%), urinary tract infection (12.4%), nasopharyngitis (11.8%), fatigue (10.7%), and dizziness (10.1%). The total number of patients with ravulizumab infusion interruption due to AEs was 8 (4.7%) patients and 74 (43.8%) patients who reported potential infusion reactions during ravulizumab treatment. SAEs reported by 1 patient or more included COVID-19 (6 patients), MG (5 patients), COVID-19 pneumonia (4 patients), cellulitis and pneumonia (3 patients each), and erysipelas, urinary tract infection, spinal compression fracture, intervertebral disc protrusion, transient ischemic attack, congestive cardiac failure, pyrexia, dyspnea, dysphagia, dehydration, and nephrolithiasis (2 patients each).

In the RTS, there were 8 deaths throughout the entirety of the study period. Two deaths occurred during the RCP (due to COVID-19 pneumonia and cerebral hemorrhage) and 6 deaths occurred during the OLE period due to the following reasons: COVID-19 (3 patients), toxicity due to various drugs (1 patient), dehydration (1 patient), and an unknown reason (1 patient).

████ patients had AEs that were unrelated to the study drug and led to discontinuation. No meningococcal infections were reported during the study.

Critical Appraisal

The lack of control group precludes causal statements about benefit and harm compared with any comparator. The open-label nature of the study may increase the risk of bias in determining the magnitude of the safety outcomes and efficacy end points that include more subjective assessments because the lack of blinding may affect patients’ expectations of treatment. The direction and magnitude of these potential biases remain unclear. Patients of the OLE were patients who did not drop out of the placebo-controlled study (92%), which puts the results at some risk of selection bias that is likely to be favouring ravulizumab. Of the patients who started the placebo-controlled phase, 123 patients completed the OLE. There is therefore a risk of bias due to missing outcome data in both the placebo-ravulizumab and ravulizumab-ravulizumab arms, with the impact on the direction of treatment effect over time not clear. The limitations of the post hoc analyses of the subgroups mirror those discussed in the systematic review section; however, these concerns are minimized as the findings for these groups were generally consistent with the overall population.

Indirect Comparisons

Description of Studies

One sponsor-submitted network meta-analysis (NMA) report compared the efficacy and harms of ravulizumab relative to currently available active treatments (efgartigimod alfa, eculizumab, IVIg, and rituximab) for the treatment of adult patients with anti–AChR antibody–positive gMG in the concomitant IST optimized cohort as well as the refractory concomitant IST optimized cohort.²⁹ In total, 7 RCTs were included in the ITC. The sponsor also submitted a matching-adjusted indirect comparison (MAIC) report comparing ravulizumab relative to efgartigimod alfa at various time points, although this was for the overall trial populations rather than for specific subgroups.³⁰

Efficacy Results

Comparative evidence of ravulizumab to other active treatments for gMG in the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort was available through sponsor-submitted NMA. Based on the results of the NMA, the evidence is insufficient to conclude whether ravulizumab differs from efgartigimod alfa, rituximab, or IVIg in terms of change from baseline in the MG-ADL total score or QMG total score in the concomitant IST optimized cohort, or whether ravulizumab differs from eculizumab in terms of change in the MG-ADL total score or QMG total score in the refractory concomitant IST optimized cohort.

Evidence from the MAIC report is insufficient to conclude whether ravulizumab differs from efgartigimod alfa in improvement in the MG-ADL total score in patients with gMG who received a previous stable dose of IST.

Harms Results

The evidence from the NMA was not sufficient to conclude whether ravulizumab differs from the other active treatments in the risk of discontinuation of the study drug due to AEs in the overall patient population with gMG.

Critical Appraisal

A priori protocols for the sponsor-submitted NMA or MAIC were not available; therefore, it cannot be known whether the analyses presented were selected from multiple analyses of the data.

One of the major concerns for NMA is that the included trials could be highly heterogeneous in terms of study design and patient characteristics at baseline. Seven RCTs were included in the NMA: 6 for the analyses in the concomitant IST optimized cohort and 2 for the analyses in the refractory concomitant IST optimized cohort. Heterogeneities were identified in the analysis populations, which included study design (phase of study, study time points, and outcome measures in different ways) and patient characteristics at baseline (age, gender, and baseline MG-ADL scores).

These differences would undermine the validity of the NMA, which relies on the transitivity assumption being upheld (i.e., that the trials are similar on all important effect modifiers). The limited number of included studies did not allow for metaregression or other techniques to adjust for differences in effect modifiers across studies within the NMA. The rarity of the population of interest limits the size and number of clinical studies completed with potential comparators and adds to the practical challenges when indirectly comparing treatment options.

Usually, including post hoc subgroups in the analyses may raise concerns since these analyses are not prespecified and can be at risk of bias due to selective reporting (e.g., there is a risk that the presented results are selected from multiple analyses of the data based on their direction, magnitude, or statistical significance). However, the sponsor provided clear justification for the selected subpopulations using criteria informed by other MG trials. Additionally, various sensitivity analyses were conducted to examine the robustness of the results from base-case analysis, and results of the sensitivity analyses were consistent with the base-case analysis. As a result, any concern for selective reporting was minimized.

In the NMA, given the lack of closed loops in the networks, consistency in the ITC analyses could not be tested. All comparisons were therefore informed only by indirect evidence, which increases the level of uncertainty. Efficacy data were sparse in the NMA for the comparison of ravulizumab versus other active treatments. Overall, the 95% credible intervals (CrIs) for the point estimates were wide for the efficacy and harms outcomes and spanned the null when compared with other regimens; therefore, confidence in the effect estimates for efficacy of the study drugs was limited due to imprecision indicated by the wide CrIs for these outcomes and precluded any conclusions as to which treatment may be favoured.

In this NMA, 2 efficacy outcomes were analyzed (the MG-ADL score and the QMG score). Therefore, the relative treatment effect of ravulizumab versus relevant comparators on other important clinical outcomes such as patients’ survival or HRQoL remains unknown. Harms were only assessed in a full population instead of the IST optimized cohorts, which were the focus of this current review.

In the MAIC analysis, various patient characteristics at baseline were considered for involvement in the adjusted analyses, including age, sex, MGFA class, disease duration, MG-ADL score, steroid use at study entry, and NSIST use at study entry. It was not clear whether other potential effect modifiers were missing. Therefore, there remains a risk that the results are biased due to residual confounding. In this analysis, the effect sample size in the CHAMPION-MG trial after matching was substantially reduced by ███ for patients treated with ravulizumab, and ███ for those treated with placebo, suggesting that results were heavily influenced by a subset of the sample in the trial that may not be representative of the full sample, nor generalizable to the original population represented by the CHAMPION-MG trial. In general, the 95% CIs for the point estimates often crossed the null (except for at 4 weeks) and precluded definitive conclusions as to which treatment may be favoured. In this MAIC analysis, change in the MG-ADL score was the only assessed outcome; therefore, other relevant outcomes were not assessed, such as HRQoL, symptom relief, or safety.

Studies Addressing Gaps in the Evidence From the Systematic Review

No studies addressing gaps were submitted by the sponsor.

Conclusions

Evidence from 1 phase III, multicentre, double-blind, randomized, placebo-controlled trial (the CHAMPION-MG study) suggested that administration of ravulizumab in adult patients with anti–AChR antibody–positive gMG resulted in statistically significant improvements compared with placebo in activities of daily living (change from baseline in the MG-ADL total score) and gMG disease severity (change from baseline in the QMG total score and the proportion of patients with a QMG 5-point response) after 26 weeks of treatment. Two post hoc cohorts of the CHAMPION-MG trial aligned with the reimbursement request under review: the concomitant IST optimized cohort (████████) and the refractory concomitant IST optimized cohort (█████████), which were the focus of this resubmission. The clinical benefit observed in the 2 cohorts was consistent with the results shown in the overall trial population. Evidence from the 2 post hoc cohorts suggested that in adult patients with anti–AChR antibody–positive gMG, ravulizumab likely results in clinically meaningful improvements in activities of daily living (the MG-ADL total score) compared with placebo. Results for other outcomes related to relief in MG disease severity (change from baseline in the QMG total score, at least a 5-point improvement in the QMG total score, and at least a 3-point improvement in the MG-ADL total score) were supportive of the analyses of changes in the MG-ADL total score. Note that confidence in the between-group differences for efficacy in the 2 cohorts were limited due to imprecision, which was indicated by the associated CIs that included small effects close to the null or that crossed the null (the CI crossed the null for the QMG total score in the refractory concomitant IST optimized cohort). The treatment effect of ravulizumab on HRQoL (the MG-QoL15r score and the Neuro-QoL fatigue score) remains uncertain. Results of post hoc cohort analyses were subject to limitations such as the potential for randomization to not be upheld, a reduced sample size, and a lack of formal statistical approaches to control for inflated type I error rate in multiple comparisons. However, in all cases, the findings of the subgroup analyses were consistent with the analyses of the full trial population. Results of an OLE study (with up to 3.5 years of follow-up) suggested that improvement in daily living activities and disease severity was sustained during the long-term follow-up and that patients who switched from placebo to ravulizumab continuously reported improvements in their MG-ADL total score. The safety profile of ravulizumab in the 2 subgroups was consistent with that reported in the full population with no unexpected safety signals and was considered manageable by the clinical expert consulted for this review.

Evidence from sponsor-submitted ITCs (an NMA and an MAIC) comparing ravulizumab to other active treatments was insufficient to make conclusions on the relative efficacy of ravulizumab to these active treatments (i.e., the NMA examined ravulizumab versus efgartigimod alfa, IVIg, and rituximab in the concomitant IST optimized cohort, and ravulizumab versus eculizumab in the refractory concomitant IST cohort; the MAIC examined ravulizumab versus efgartigimod alfa in the overall population of the CHAMPION-MG trial). Limitations identified in these ITCs included a limited number of included studies, heterogeneity in trial characteristics and patient characteristics across these trials, and CrIs or CIs that crossed the null.

Introduction

The objective of this report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of ravulizumab (10 mg/mL and 100 mg/mL for IV infusion) in the treatment of adult patients with gMG. This is a resubmission and the focus of this report has been placed on comparing ravulizumab to relevant comparators (in particular, subsets of patients with gMG) and identifying gaps in the current evidence.

Disease Background

Content in this section has been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the review team.

MG is a chronic autoimmune disease in which antibodies against AChRs or functionally associated molecules in the neuromuscular junction disrupt nerve impulse conduction, resulting in localized or generalized skeletal muscle weakness.^1-4 In approximately two-thirds of patients with MG, the disease initially affects the extraocular muscles. These patients present with isolated ptosis, diplopia, or both, and no signs or symptoms of muscle weakness elsewhere.⁵ In 10% to 15% of all patients with MG, weakness remains restricted exclusively to the eyes (ocular MG).^1,3,5,6 The remainder of patients either are diagnosed with or progress within a few years to gMG, where symptoms affect muscle groups in the head, neck, trunk, and/or limbs and are not limited to only eye muscles.^31-35 Symptoms of gMG include eyelid drooping and double vision, altered facial expression, difficulty chewing and swallowing food, difficulty speaking, and in patients with more severe disease, problems with limb movement and breathing.^3,6 Collectively, symptoms of MG negatively impact patients’ HRQoL.³ The disease has a fluctuating natural history: MG exacerbation (an increase in symptoms in patients who were previously asymptomatic or minimally symptomatic) and myasthenic crisis (muscle weakness causing life-threatening difficulties with breathing and swallowing and requiring ventilator support) can occur gradually or without warning.³⁶ A widely used classification system from a task force of the MGFA stratifies patients by the extent and severity of muscle weakness, where class I equals isolated ocular muscle weakness, class II equals mild generalized weakness involving nonocular muscles, class III equals moderate generalized weakness involving nonocular muscles, class IV equals severe generalized weakness involving nonocular muscles, and class V equals intubation due to respiratory muscle weakness.³⁷ For MG symptoms that fluctuate, the patient’s most severe weakness was used to assess the MGFA clinical class.²

Globally, the prevalence rates of MG range from 150 cases to 200 cases per 1 million population, and they have steadily increased over the past 50 years, at least partly due to improvements in recognition, diagnosis, treatment, and an overall increase in life expectancy. More recent studies addressing incidence rates show a wide range from 4 cases to 30 cases per million person-years.^3,7 In Canada, the prevalence of MG is approximately 263 cases to 320 cases per 1 million population and its incidence is approximately 23 cases per 1 million population annually, based on the data in Ontario from 1996 to 2013.⁸ Autoantibodies against AChR, muscle-specific kinase (MuSK), and LRP4 can be detected in approximately 80%, 1% to 10%, and 1% to 3% of patients with gMG, respectively.³ Ten percent of patients with MG have a thymoma, and the prevalence of thymoma increases with increasing age and is associated with more severe disease.^3,38 According to the clinical expert consulted by this review, 15% to 20% of patients with gMG have refractory disease, which may be defined by not achieving symptom control even though sufficient or multiple conventional therapies have been given, although a commonly accepted definition of refractory disease is unavailable. A web-based observational disease and outcomes registry also estimated that approximately 15% of patients with MG is considered medically refractory to conventional treatment strategies.³¹ Diagnosis of MG and gMG is made based on the combination of relevant symptoms and signs, a positive test for specific autoantibodies (e.g., AChR, MuSK, and LRP4 autoantibodies), and neurophysiological testing. Thymic status should be determined by means of mediastinal imaging.³ In patients with mild symptoms, the disease may be underdiagnosed.

The prognosis of MG varies by symptom severity and response to treatment. Patients with severe or refractory symptoms, including those hospitalized for myasthenic crisis, are at higher risk for complications compared with those with mild or nonbulbar symptoms. MG may be associated with increased mortality. In a population-based study from Denmark, overall mortality was significantly increased for patients with anti–AChR antibody–positive MG compared with matched controls from the general population (mortality rate ratio = 1.41 [95% CI, 1.24 to 1.60]).⁹ However, other contemporary studies have not found an increased mortality rate, perhaps due to better disease management with the increasing availability of effective immunotherapies.²

Standards of Therapy

Content in this section has been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the review team.

According to the clinician expert consulted for this review, the goal of treatment in most patients with gMG is to reduce disease symptoms (ideally aiming for minimal disease manifestation) as well as adverse effects of MG therapy and to allow the patient to function and work normally with good HRQoL. Other goals of treatment include avoiding MG exacerbations and myasthenic crisis, minimizing hospitalizations (including intensive care unit admissions), and reducing the numbers and doses of therapies required for symptom control.

The initial symptomatic treatment for most patients with gMG is an AChE inhibitor such as pyridostigmine. According to the clinical expert consulted for this review, peripherally acting AChE inhibitors are used as symptomatic treatments for temporarily alleviating muscle weakness in these patients. A minority of patients with MG have a spontaneous remission or respond to AChE inhibitors; however, most patients need treatment with corticosteroids and/or steroid-sparing drugs when they do not reach their treatment goals with AChE inhibitors.

Corticosteroids are the first-line immunosuppressive therapy for patients with MG who remain symptomatic while on AChE inhibitors or who desire better symptom control. Early use of oral steroids in patients with pure ocular symptoms may delay or reduce the risk of generalization and the worsening of underlying symptoms.^10,11 Patients who do not respond to corticosteroids or who have comorbidities that are not suitable for treatment with corticosteroids may initiate treatment with NSISTs.¹² Among the NSISTs, azathioprine and mycophenolate mofetil are commonly used in gMG. The use of conventional therapy (AChE inhibitors, corticosteroids, and/or NSISTs) is associated with the slow onset of benefit, and incomplete efficacy in eliminating MG symptoms and significant toxicities. Corticosteroids, particularly if used at high doses or over prolonged periods, are associated with cataracts, cushingoid appearance, osteoporosis and fractures, glucose intolerance and diabetes, hypertension, infections, mood disturbances, and weight gain.^13-15

Other treatment options, such as immunomodulating therapies like PLEX and IVIg, are usually reserved for use on a short-term basis for life-threatening situations or crises, for use in some patients before surgery or corticosteroid treatment, to obtain a rapid treatment response, or for when treatments have been ineffective.¹⁰ The choice of whether to use IVIg or PLEX depends on patient comorbidities such as sepsis (PLEX-contraindicated) or renal failure (IVIg-contraindicated) and on treatment availability due to blood supply shortages.¹⁰ Importantly, PLEX or IVIg have no long-term impact on the disease course in gMG.²⁰ Significant limitations of IVIg or PLEX³⁹ have been noted by clinicians in Canada: venous access issues, potentially serious cardiovascular and systemic adverse effects or contraindications to these therapies, waning efficacy in spite of an initial response to IVIg or PLEX, considerable infusion or transfusion requirements that make these efforts unsustainable, a lack of supply of product (IVIg or subcutaneous immunoglobulin) that threatens to interrupt therapy abruptly to the detriment of patients, and finally, that PLEX is available in very few large centres in each province and IVIg can only be infused in hospital infusion rooms across the country, which contributes to considerable delay in instituting this therapy.

Novel biologic treatments such as efgartigimod alfa and eculizumab have emerged. Efgartigimod alfa is an FcRn inhibitor, and was approved by Health Canada in 2023 for the treatment of adult patients with gMG who are anti–AChR antibody–positive.¹⁶ This therapy also received a positive reimbursement recommendation with conditions from CDA-AMC as an add-on therapy for adult patients with anti–AChR antibody–positive gMG whose symptoms persist despite adequate treatment with AChE inhibitors, corticosteroids, and/or NSISTs.¹⁷ Efgartigimod alfa is currently undergoing negotiations for product listing agreements. Eculizumab, a monoclonal antibody that specifically binds to the complement protein C5 with high affinity, was approved for use in adult patients with gMG by Health Canada in 2018.¹⁸ Although eculizumab is indicated for use in all adult patients with gMG, CDEC recommended that it be reimbursed in patients with refractory gMG who are anti–AChR antibody–positive and have inadequate symptom control after undergoing at least 2 ISTs in the previous 12 months, or at least 1 IST and chronic IVIg or PLEX in the previous 12 months, along with other conditions.^19,20 A successful product listing agreement for eculizumab was not reached between the drug plans and the sponsor, and therefore is not listed in any jurisdictions in Canada. The international consensus guidance from 2021 states that rituximab (an immunomodulatory drug) is generally recommended only for patients with anti–MuSK antibody–positive MG considering the great uncertainty of its efficacy in anti–AChR antibody–positive MG.¹² The 2021 CDA-AMC technology review of rituximab concluded that rituximab may be associated with improvements in clinical status, the use of concurrent immunomodulatory therapies, quality of life, and various laboratory parameters in patients with MG compared to their previous treatments; however, the use of these findings in informing clinical and policy decisions was limited due to the substantial methodological limitations of the included literature in this report.²¹ Challenges with rituximab include the slow onset of action (a minimum of 3 months), a lack of efficacy in the anti–AChR antibody–positive gMG population, and significant side effects owing to its broad immunosuppressive mechanism — including life-threatening infections, hepatitis B reactivation, severe mucocutaneous reactions, and progressive multifocal leukoencephalopathy.^40-43 In many jurisdictions in Canada, rituximab is listed with relatively accessible criteria, although it remains challenging to access in certain provinces.

Lastly, surgery (thymectomy) is a treatment option in select circumstances.²⁰ Thymectomy is recommended in nonthymomatous adult patients with anti–AChR antibody–positive gMG aged 50 years or younger if they are stable (as an early treatment option), as well as in patients with or without AChR antibodies if ISTs elicit an insufficient response or are intolerable due to side effects.¹²

According to the clinician expert input received for the original submission for ravulizumab, as MG symptoms improve, doses of AChE inhibitors, corticosteroids, and then other ISTs are reduced and the frequency of IVIg or PLEX is reduced until the minimal maintenance therapy required for remission is identified. Patients whose symptoms persist despite treatment with adequate doses of corticosteroids, ISTs, and/or chronic IVIg or PLEX, or patients for whom the doses or frequencies of these therapies cannot be reduced, are considered to have refractory gMG, which accounts for approximately 10% to 15% of patients with gMG.

Drug Under Review

Ravulizumab is a terminal complement inhibitor that specifically binds to the complement protein C5 with high affinity, thereby inhibiting its cleavage to C5a and C5b and preventing the generation of membrane attack complex.²² The precise mechanism by which ravulizumab exerts its therapeutic effect in patients with gMG is unknown but is presumed to involve the reduction of terminal complement complex C5b-9 deposition at the neuromuscular junction.

In January 2023, ravulizumab was approved by Health Canada for the treatment of adult patients with anti-AChR antibody–positive gMG. It is supplied as a 10 mg/mL or 100 mg/mL concentrate for IV infusion. The recommended ravulizumab maintenance dosing in adult patients (aged 18 years or older) with gMG with a body weight greater than or equal to 40 kg is based on the patient’s body weight, with maintenance doses administered every 8 weeks, starting 2 weeks after the loading dose (Table 3). Patients must be vaccinated against meningococcal infections before, or at the time of, initiating ravulizumab, unless the risks of delaying ravulizumab therapy outweigh the risks of developing a meningococcal infection.²²

Table 3: Ravulizumab Weight-Based Dosing Regimen

gMG = generalized myasthenia gravis.

Source: Product monograph for ravulizumab.²²

In this resubmission, the sponsor’s current reimbursement request is as add-on therapy for adult patients with anti–AChR antibody–positive gMG whose symptoms persist despite adequate treatment with AChE inhibitors, corticosteroids, and/or NSISTs.

Key characteristics of ravulizumab are summarized in Table 4 with other treatments available for gMG.

Table 4: Key Characteristics of Ravulizumab and Other Drugs Used for the Treatment of gMG

AChEI = acetylcholinesterase inhibitor; AChR = acetylcholine receptor; gMG = generalized myasthenia gravis; IgG = immunoglobulin G; IgG1 = immunoglobulin G1; IST = immunosuppressive therapy; IVIg = IV immunoglobulin; NA = not applicable; PLEX = plasma exchange or plasmapheresis; PML = progressive multifocal leukoencephalopathy; PO = orally; TSL = tumour lysis syndrome.

^aRelevant Health Canada–approved indications.

^bSupplemental ravulizumab doses of 1,200 mg to 1,800 mg are given following PLEX and supplemental doses of 600 mg are given following IVIg.

^cSupplemental eculizumab doses of 300 mg to 600 mg are given following PLEX.

Sources: Product monographs for ravulizumab,²² efgartigimod alfa,¹⁶ eculizumab,¹⁹ pyridostigmine,⁴⁴ and rituximab.⁴⁵

Submission History

Ravulizumab was initially reviewed by CDA-AMC for the treatment of adults with anti-AChR antibody–positive gMG and received a negative funding recommendation from CDEC on August 24, 2023.²³ The recommendation was based on evidence assessed in the CDA-AMC Clinical Review Report for the initial submission of ravulizumab.⁴⁶ Key reasons for the negative recommendation included the following.

• There was a lack of IST dosing information to conclude whether IST was optimized for patients at the time of study enrolment. Without IST dosing information at study entry, the committee was uncertain if the dose and durations of corticosteroid and NSIST therapies at study enrolment were sufficient to achieve maximal responses. During the RCP, dosage changes in concomitant ISTs were discouraged; however, rescue therapy was allowed upon patients’ clinical deterioration and dosage changes were permitted if deemed medically necessary by the investigator. The committee noted that changes in concomitant MG therapies were not generalizable to clinical practice since changes to these medications were discouraged by the study protocol. Therefore, there was insufficient evidence on the treatment effect of ravulizumab compared to conventional therapies such as those used in clinical practice.

• There was insufficient evidence on the treatment effect of ravulizumab in patients who remained symptomatic and were not necessarily considered refractory. While subgroup analyses conducted among patients who had been exposed to at least 2 ISTs were consistent with the main analyses, these results were limited by smaller numbers of patients and the lack of formal statistical testing with adjustment for multiple testing (an increased risk of type I error). Given the lack of IST dosing information, it was unclear whether IST dosing was optimized at study entry in this subgroup.

• There was insufficient evidence on the treatment effect of ravulizumab in patients with refractory gMG. Patients enrolled in the CHAMPION-MG trial were not required to be refractory and therefore it was unknown how many patients were refractory and whether the results observed in the overall patient population would be the same in these patients.

• There was a lack of evidence on the relative efficacy and harms of ravulizumab compared to relevant comparators (e.g., rituximab, IVIg, PLEX). Evidence from sponsor-submitted ITCs and an observational study comparing ravulizumab to eculizumab suggested uncertainty in the relative efficacy of these drugs. In addition, the ITCs were limited by differences in the study populations, a small effective sample size, selection criteria, and demographic differences between included studies and the absence of adjustment for potentially important clinical covariates.

In response to the draft CDEC recommendation, the sponsor submitted a request for reconsideration in which the sponsor requested that CDEC reconsider the following.

• The study design of the CHAMPION-MG trial allowed for the robust comparison of ravulizumab with relevant standard of care treatments in gMG. The sponsor noted that the CHAMPION-MG trial enrolled patients on stable and optimized doses of ISTs with mean treatment durations well above the minimum treatment length requirement in both treatment arms. According to the sponsor, the CHAMPION-MG trial compared ravulizumab to current standard of care therapies at optimized doses.

• The sponsor noted that rituximab and IVIg or PLEX are not considered relevant comparators for ravulizumab in the treatment of anti–AChR antibody–positive gMG largely due to their limited accessibility and limited evidence to support their efficacy.

• There are significant unmet needs in patients with gMG who are symptomatic despite treatment due to the limitations of current standard of care. The sponsor noted that ravulizumab selectively targets the underlying disease pathogenesis and has a rapid onset of action, sustained clinical benefit, and a well-tolerated safety profile that fulfills the current gaps in the management of patients with gMG.

• The sponsor proposed that the reimbursement criteria for ravulizumab align with the unmet needs of patients, the studied population most represented by the pivotal CHAMPION-MG trial — specifically, patients with gMG who remain symptomatic (the MG-ADL score of ≥ 6 and the MGFA classification of class II to class IV) despite 2 or more prior ISTs. According to the sponsor, the results of the prespecified subgroup analyses in patients with more than 2 prior ISTs are consistent with the results of the primary analysis and demonstrated statistically significant MG-ADL improvements from baseline to week 26 in the ravulizumab group compared to placebo.

However, CDEC maintained its initial position that there was insufficient information on IST dosing to conclude whether patients in the CHAMPION-MG trial were on optimized IST dosages at study entry, that rituximab and IVIg or PLEX were relevant comparators, that it remained unclear from the evidence in the CHAMPION-MG trial whether ravulizumab could be more effective and safer than currently available treatments in patients with symptomatic but nonrefractory gMG, and that the proportion of patients with refractory gMG who received an adequate trial of IST could not be determined in the CHAMPION-MG trial.

The drug was resubmitted for review by the sponsor on June 4, 2024, on the basis of the availability of new evidence and a revised reimbursement request for ravulizumab as an add-on therapy for adult patients with anti–AChR antibody–positive gMG whose symptoms persist despite adequate treatment with AChE inhibitors, corticosteroids, and/or NSISTs. The new evidence included the following.

• The efficacy and harms of ravulizumab in 2 post hoc subgroups in the CHAMPION-MG trial: These subgroups were the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort.²⁸ The sponsor noted that there is no universally accepted definition of IST “optimization” in clinical practice due to variability in the disease symptoms, adverse effects of the current therapies, and individual patient characteristics. Therefore, the inclusion criteria of Study ADAPT (efgartigimod alfa versus placebo) and Study REGAIN (eculizumab versus placebo) were adopted to select patients who would be considered to have IST optimization, or were IST optimized but still with refractory disease at the time of study enrolment in the CHAMPION-MG trial, respectively. These criteria were based on total treatment duration and stable dose duration.

• NMA:³⁰ The relative efficacy of ravulizumab was assessed against efgartigimod alfa, rituximab, IVIg, and eculizumab in the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort. In addition, the sponsor submitted an MAIC report comparing the efficacy of ravulizumab relative to efgartigimod alfa in the overall trial populations.³⁰

• OLE data⁴⁷ in the CHAMPION-MG trial: In the end-of-study analysis (the last patient’s last visit on May 25, 2023), a total of 161 patients had entered the OLE period. The efficacy and safety of ravulizumab were assessed in the 2 cohorts during the OLE phase.

The sponsor noted that the new evidence addressed concerns raised by CDEC during the initial submission; these concerns were related to IST optimization during the CHAMPION-MG trial, the treatment effect of ravulizumab in patients with refractory disease, comparative evidence of ravulizumab versus relevant comparators, and the long-term efficacy and harms of ravulizumab.

Perspectives of Patients, Clinicians, and Drug Programs

The full patient and clinician group submissions received are available in the consolidated patient and clinician group input document for this review on the project website.

Patient Group Input

This section was prepared by the review team based on the input provided by patient groups.

CDA-AMC received 1 patient group submission from MDC. MDC is a health charity that supports people affected by muscular dystrophies and related muscle diseases 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 in semistructured interviews. Surveys were shared with members via e-blasts, personalized invites, and online patient groups. MDC also conducted an MG Canadian journey mapping project among adults living with MG via virtual interviews, round table sessions, surveys, and HRQoL measures (EQ visual analogue scale, EQ-5D, MG-ADL, and MG-QoL). MDC collected information from 215 patients impacted by MG through a health care experience survey and semistructured phone or virtual interviews. These patients consisted of 83 males and 132 females aged between 22 years and 78 years from all provinces in Canada.

In addition, MDC sought patients’ opinions on the value of having ravulizumab approved for use in Canada from those affected by gMG. In total, 14 patients (11 females and 3 males) aged 32 years to 69 years (mean age of 53 years) provided input on their experience with MG and their knowledge of ravulizumab. These patients were recruited from all regions across Canada (the Atlantic, Central, and Prairies regions) except for the North (Nunavut, the Northwest Territories, and Yukon).

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

With respect to the currently available treatments, MDC noted that patients reported negative experiences with steroids and prednisone, the slow onset of medication effects, and a feeling of trial and error with medications, but a positive experience with IVIg. MDC added that, while according to the respondents, supportive treatments have had positive health outcomes, there are concerns about the long-term and sustained benefits of these treatments.

MDC noted that patients with gMG seek better control over their condition to minimize the impact of symptoms, side effects, and disease exacerbations on their lives, allowing them to maintain their independence and avoid serious hospital admissions. MDC added that patients stated that they would be willing to deal with the side effects of medications if these aspects of MG were better controlled. According to MDC, respondents indicated that currently available therapies may decrease MG exacerbations but do not improve their overall HRQoL.

Based on the patient group input, the important factors in evaluating different treatments include treatment administration, potential side effects, the duration and frequency of treatments, convenience (e.g., travel time and parking for clinic visits), and financial impact (costs).

MDC explained that according to the recent round table of patients with gMG in Canada, patients value new treatments that offer improved disease control and disease symptoms and extend dosing intervals.

In terms of diagnostic testing, MDC stated that 85% of respondents reported significant difficulty in getting diagnosed. The vast majority of patients found the test to be cost-effective but noted the overall lengthy process with many missed opportunities for improvement (e.g., delays, misdiagnoses, costs incurred). A diagnosis received as part of hospitalization was reported as a seamless experience.

MDC believed that there is a pressing need for improved treatment options to address the ongoing challenges faced by patients with MG and that ravulizumab provides a new treatment option for patients with MG that has demonstrated efficacy, safety, and improved dosing convenience compared to other treatment options.

Clinician Input

Input From Clinical Experts Consulted by the Review Team

All CDA-AMC review teams include at least 1 clinical specialist with expertise regarding 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, 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 gMG.

Unmet Needs

The clinical expert consulted for this review identified unmet needs associated with currently available treatments for patients with gMG whose symptoms persist even if they have been treated with conventional medications for this disease (such as AChE inhibitors, corticosteroids, and/or NSISTs). The unmet needs are as follows: although multiple treatment options are available to patients with gMG, some patients do not have an adequate response to the existing treatments, and 15% to 20% of patients have refractory disease and require alternative therapeutic options; patients may become intolerant of ISTs (e.g., from treatment with corticosteroids); and some exiting treatments are only suitable for select patients. Treatments with a more rapid onset of action, longer-lasting benefits, improved efficacy in patients with refractory gMG, and fewer side effects are needed.

Place in Therapy

The clinical expert consulted for this review noted that treatment with traditional immunosuppressants sometimes is not satisfactory (e.g., prednisone is associated with a wide spectrum of side effects, especially in the chronic use; azathioprine may not be effective and may require 9 months to 12 months to reach its maximum effect; mycophenolate mofetil may not be effective either and poses a risk of teratogenicity). In addition, there were no good-quality RCTs supporting the use of these drugs, even though they are widely accepted and routinely used in clinical practice. The expert stated that because of its unique mechanism of action (a complement C5 inhibitor), ravulizumab could be another treatment option for patients living in Canada with gMG whose disease has had an inadequate response to or has not responded to conventional ISTs or those who cannot tolerate traditional ISTs.

Patient Population

According to the clinical expert consulted for this review, patients with significant symptoms of gMG and resistance to conventional therapies, patients who require multiple concomitant ISTs, or patients who may experience significant side effects or intolerance of standard immunosuppressants may benefit from treatment with ravulizumab.

The expert also noted that candidates for ravulizumab therapy (primarily patients with refractory gMG and those with severe but nonrefractory gMG) would be identified through the judgment of an expert neurologist based on clinical evaluation following serologic testing for AChR antibodies and, potentially, following a chest CT scan to rule out thymoma and thymic carcinoma. The expert also noted that although patients with thymoma were excluded from the clinical trials of complement inhibitors, there is no reason to expect that these patients would not benefit from these drugs. These new treatments are likely to reduce the chronic use of corticosteroids, diminish side effects of immunosuppressants, and decrease the number of patients with refractory disease.

Assessing Response to Treatment

The clinical expert stated that a patient’s response to treatment with ravulizumab would be assessed by certain MG scales (e.g., a decrease of approximately 2 points in the MG-ADL total score and/or a decrease of approximately 3 points in the QMG score, every 1 month to 3 months), by a physician’s clinical examination of MG symptoms and signs, and via the reduction of other MG therapies, such as corticosteroid use, chronic IVIg or PLEX, and the number of emergency department visits and hospitalizations.

Patients need regular follow-up visits with a neuromuscular specialist or neurologist to monitor the degree of response to treatment. Depending on the severity of the patients’ symptoms, the follow-up visits can range from every 1 month to 3 months when ravulizumab is initiated, and every 3 months thereafter if the patients have a stable neurologic status.

Discontinuing Treatment

The clinical expert consulted for this review stated that ravulizumab would be discontinued in patients who do not respond favourably to treatment with ravulizumab (e.g., less than 2-point improvement of MG-ADL total score), continue to require therapy with multiple concomitant ISTs, or exhibit recurrent MG crises or exacerbations of symptoms requiring rescue therapies with IVIg or PLEX. Ravulizumab would also be discontinued in patients who experience serious side effects or secondary infection such as meningococcal meningitis, or in patients who prefer to discontinue. At present, there is no available data or anecdotal experience to inform decision-makers and clinicians on whether it may be plausible to hold the treatment once the patient achieves sustained stable neurologic status or whether in some patients, lifelong treatment with complement inhibitors — including ravulizumab — would be required to maintain the achieved clinical benefit.

Prescribing Conditions

The clinical expert consulted for this review stated that the initiation of ravulizumab would be supervised by a neurologist with expertise in managing patients with MG (a neuromuscular expert). Given the small group of neuromuscular experts in Canada and the high demand on their time, it may not be realistic to restrict the use of ravulizumab to academic expert settings only. Therefore, a trained neurologist may be starting the therapy and providing the follow-up assessments.

Clinician Group Input

This section was prepared by the review team based on the input provided by clinician groups.

One input was received from NMD4C, a new pan-Canadian network that brings together clinical, scientific, technical, and patient expertise to improve care, research, and collaboration in neuromuscular disease. The mission of NMD4C is to improve the care, research, and treatment of neuromuscular diseases for all people living in Canada. The information presented in this submission was gathered from 8 clinicians with experience in treating gMG.

According to the clinician input, the current mainstay of therapies for MG includes supportive therapies, symptomatic treatments, and disease-modifying strategies. NMD4C added that outside the setting of crisis or imminent crisis, the effects of disease-modifying therapy are usually transient, lasting a few weeks, and additional ongoing immunotherapy is often needed.

NMD4C noted that the goals of therapy are to minimize morbidity and mortality from MG, keep patients out of hospital, improve quality of life, prevent repeated attacks, and prevent prolonged, untreated, or partially treated MG.

NMD4C stated that some of the treatment gaps include limited therapy options for active and refractory disease, inadequate response to the traditional immunotherapies, the slow onset of treatment action, side effects, a lack of response, and treatments for pediatric patients.

In terms of place in therapy, NMD4C believes ravulizumab addresses the underlying disease process by inhibiting the immune-mediated damage to the neuromuscular junction rather than being a symptomatic treatment. Ravulizumab will complement other treatments that act upstream in the immune-mediated damage. Conversely, the concomitant use of PLEX or IVIg may reduce the efficacy of ravulizumab by reducing its concentrations. The place in therapy for ravulizumab would likely be after treatment with pyridostigmine, and after treatment with either steroids and/or other ISTs, including steroid-sparing immunosuppressive drugs. Ravulizumab is likely to affect the treatment paradigm of patients with refractory and nonrefractory MG who are not responsive to first-line and second-line therapies or require chronic IVIg infusions or PLEX. It may also be considered in patients who are intolerant of other immunomodulatory treatments.

NMD4C stated that it would be appropriate to recommend that patients with gMG try other treatments before initiating ravulizumab, which requires periodic IV infusions over an extended period, is unlikely to induce long-term disease remission, is likely to be expensive, may not be available in smaller cities or nonspecialized centres, and requires extensive expertise. Ravulizumab will be contraindicated in patients who have had previous infection with Neisseria meningitidis; vaccination for this pathogen is required before consideration of treatment with ravulizumab.

Based on the clinician group input, adult patients who are seropositive for AChR antibodies and have gMG would be best suited for treatment with ravulizumab; there are no data on the efficacy of ravulizumab for a minority population of patients with MG — including those aged younger than 18 years, those who had thymectomy within the past year, those with thymic carcinoma, those who are pregnant or breastfeeding, patients with anti–MuSK or anti–LRP4 antibodies, or seronegative patients. Patients with MG who are seropositive for AChR antibodies who have not responded to pyridostigmine and to steroids and/or oral ISTs or are additionally dependent on periodic PLEX or chronic IVIg therapy are most in need.

NMD4C noted that the diagnosis must have been confirmed clinically and supported by confirmatory laboratory tests before treatment with ravulizumab.

NMD4C explained that the outcomes that are used to determine whether a patient is responding to treatment in clinical practice include increased survival, the avoidance of emergency department visits or hospital admissions (including admissions to the intensive care unit), the avoidance of the need for rescue as well as maintenance therapy with IVIg and plasmapheresis, a reduction in the dose and/or duration of concomitant steroids, a reduction in the level of fatigable weakness, and an improvement in activities of daily living and quality of life.

Based on the NMD4C input, the factors that should be considered when deciding to discontinue treatment include the amount of clinical improvement or response, the duration of time spent in a clinically stable state, the AEs associated with the treatments, and the inconvenience associated with the therapy.

NMD4C added that an appropriate setting for treatment includes a clinical team with general knowledge of MG, appropriate nursing experience in managing IV medications, and familiarity with venous access issues and managing potential adverse effects. It is recommended that ravulizumab be prescribed by neurology specialists with expertise in MG diagnosis, assessment, monitoring, and management of patients with MG.

Drug Program Input

The drug programs provide input on each drug being reviewed through the 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 for this review are summarized in Table 5.

Table 5: Summary of Drug Plan Input and Clinical Expert Response

AChEI = acetylcholinesterase inhibitor; AChR = acetylcholine receptor; CDA-AMC = Canada’s Drug Agency; CDEC = Canadian Drug Expert Committee; FWG = Formulary Working Group; gMG = generalized myasthenia gravis; IST = immunosuppressive therapy; ITC = indirect treatment comparison; IVIg = IV immunoglobulin; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; NSIST = nonsteroidal immunosuppressive therapy; PLA = product listing agreement; PLEX = plasma exchange or plasmapheresis; RCT = randomized controlled trial; vs. = versus.

Clinical Evidence

The objective of this Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of ravulizumab (10 mg/mL and 100 mg/mL for IV infusion) in the treatment of gMG. The focus will be placed on comparing ravulizumab to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of ravulizumab is presented in 3 sections with the CDA-AMC critical appraisal of the evidence included at the end of each section. The first section, the systematic review, includes pivotal studies and RCTs that were selected according to the sponsor’s systematic review protocol. The second section includes sponsor-submitted long-term extension studies. The third section includes indirect evidence from the sponsor. There can also be a fourth section in such reviews that includes additional studies considered by the sponsor to address important gaps in the systematic review evidence. However, no studies addressing gaps were submitted by the sponsor in the review of ravulizumab.

Included Studies

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

• 1 pivotal study (the CHAMPION-MG trial)²⁴ identified in the systematic review, including data from 2 post hoc cohorts²⁸ submitted for this resubmission

• 1 long-term extension study⁴⁷

• 1 NMA²⁹ and 1 MAIC analysis.³⁰

Systematic Review

Content in this section has been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

Description of Studies

Characteristics of the included study are summarized in Table 6.

Table 6: Details of CHAMPION-MG Study

AChR = acetylcholine receptor; gMG = generalized myasthenia gravis; IST = immunosuppressive therapy; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item Scale - Revised; MGC = Myasthenia Gravis Composite; MGFA = Myasthenia Gravis Foundation of America; MGFA-PIS = Myasthenia Gravis Foundation of America Post-intervention Status; Neuro-QoL = Quality of Life in Neurological Disorders; OLE = open-label extension; q.8.w. = every 8 weeks; QMG = Quantitative Myasthenia Gravis; RCP = randomized controlled period; RCT = randomized controlled trial.

Sources: Clinical Study Report for the CHAMPION-MG trial,²⁴ and addendum of the CHAMPION-MG trial.^28,47 Details included in the table are from the sponsor’s Summary of Clinical Evidence.

The CHAMPION-MG trial (N = 175)²⁴ was a phase III, double-blind, multicentre, placebo-controlled RCT with an OLE period of up to 4 years. The primary objective of the CHAMPION-MG trial was to evaluate the safety and efficacy of ravulizumab compared with placebo in complement inhibitor–naive adult patients with gMG.

The study design consisted of 3 periods: a screening period of up to 4 weeks, a 26-week double-blind RCP, and an OLE period of up to 4 years (or until the product is registered or approved and available by prescription [in accordance with country-specific regulations], or until the study drug can be provided via an Alexion post-trial access program, whichever occurs first). After screening, if all of the inclusion criteria and none of the exclusion criteria were met, patients were vaccinated against N. meningitidis, if not already vaccinated within the 3 years before their enrolment in the study.⁴⁸ Following screening, adult patients with anti–AChR antibody–positive gMG were enrolled at 85 centres in 13 countries (primarily North America and Europe; 5 sites in Canada) and randomized 1:1 to receive either a weight-based dose of ravulizumab or a matching placebo for 26 weeks. Randomization was stratified by region (North America, Europe, Asia-Pacific, and Japan). The matching placebo, consisting of a sterile clear solution in an identically packaged 30-mL vial, was administered on the same schedule. All investigative site personnel, sponsor staff and designees, staff directly associated with the conduct of the study, and patients were blinded to treatment assignment.

The RCP portion of the trial is complete (the data cut-off date was May 11, 2021; the database was locked on June 30, 2021) and the OLE concluded on May 25, 2023 (the date of the last patient’s last visit). Patients receiving an IST at screening could continue taking their baseline ISTs throughout the RCP and OLE portions.

Populations

Inclusion and Exclusion Criteria

Patients enrolled in the CHAMPION-MG study were male and female adults diagnosed with anti-AChR antibody–positive gMG at least 6 months before the screening visit. Patients had MGFA clinical classification of class II to class IV at screening and their MG-ADL profile score must have been 6 or greater at both screening and randomization. Patients who were currently receiving treatment with AChE inhibitors and/or ISTs (i.e., oral corticosteroids, azathioprine, mycophenolate mofetil, methotrexate, cyclosporine, tacrolimus, or cyclophosphamide) must have been receiving treatment and must have been on a stable dose for protocol-specified time periods. Patients were excluded from the study if they had any active or untreated thymoma, or a history of thymic carcinoma or thymic malignancy. Patients with a history of thymectomy, thymomectomy, or any thymic surgery within the 12 months before screening were also excluded. Additionally, if patients presented with clinical features that were consistent with an MG crisis or exacerbation or with clinical deterioration at the time of the screening visit, or before randomization, they were ineligible for study participation. Finally, patients did not advance past screening if they had received previous treatment with complement inhibitors. In total, 175 patients (86 patients in the ravulizumab group and 89 patients in the placebo group) from 13 countries were enrolled; of those, 6 patients were enrolled from sites in Canada.

In this resubmission, the sponsor provided new clinical evidence to support its revised reimbursement request: ravulizumab as add-on therapy for adult patients with anti-AChR antibody–positive gMG whose symptoms persist despite adequate treatment with AChE inhibitors, corticosteroids, and/or NSISTs. Post hoc analyses were performed in a concomitant IST optimized cohort and a refractory concomitant IST optimized cohort.

The sponsor noted that there is no universally accepted definition of IST “optimization” in clinical practice. Therefore, the inclusion criteria of the previous Study ADAPT (efgartigimod alfa versus placebo) and Study REGAIN (eculizumab versus placebo) were adopted to select patients who would be considered to have IST optimization and IST optimization with refractory disease at the time of study enrolment from the CHAMPION-MG trial, respectively. These criteria were based on “total treatment duration” and “stable dose duration” before screening. Therefore, the definitions of the 2 cohorts were as follows.

• Concomitant IST optimized cohort: This was a subset of the full population in the CHAMPION-MG trial (n = ███ of 175 patients in the full population, █████ included patients who were on at least 1 concomitant IST at screening and met the following:

  • if on a concomitant corticosteroid, the patient had been treated with that steroid for 3 or more months before screening and had been on a stable dose for 1 or more month

  • if on a concomitant NSIST (e.g., azathioprine), the patient had been treated with that NSIST for 6 or more months before screening and had been on a stable dose of that NSIST for 3 or more months.

• Refractory concomitant IST optimized cohort: This was a subset of the full population in the CHAMPION-MG trial ██ ██ of 175 patients in the full population █████████████ patients in the concomitant IST optimized cohort, █████). Patients in this refractory cohort met both the concomitant IST optimized cohort criteria as well as the definition of refractory disease per the inclusion criteria of Study REGAIN. Specifically, patients in this cohort of the CHAMPION-MG trial were defined as not achieving symptom control after 12 months of treatment with either of the following:

  • 2 or more ISTs

  • 1 or more IST and chronic IVIg or PLEX.

Interventions

During the randomization period, patients received a weight-based loading dose of ravulizumab or placebo on day 1, followed by maintenance doses of ravulizumab or placebo on day 15 and every 8 weeks thereafter until week 26. Both ravulizumab and placebo were administered by IV infusion. Following week 26, all patients could transition into the OLE phase, which continued for up to 4 years.

Patients received a body weight–based loading dose of ravulizumab (10 mg/mL concentrate) on day 1 as follows:

• a body weight of 40 kg or greater but less than 60 kg meant a loading dose of 2,400 mg

• a body weight of 60 kg or greater but less than 100 kg meant a loading dose of 2,700 mg

• a body weight of 100 kg or greater meant a loading dose of 3,000 mg.

Subsequently, patients received body weight–based maintenance doses of ravulizumab on day 15 and every 8 weeks (± 2 weeks) thereafter during the RCP (week 10 and week 18) as follows:

• a body weight of 40 kg or greater but less than 60 kg meant a maintenance dose of 3,000 mg

• a body weight of 60 kg or greater but less than 100 kg meant a maintenance dose of 3,300 mg

• a body weight of 100 kg or greater meant a maintenance dose of 3,600 mg.

Supplemental doses of ravulizumab were administered if IVIg or PLEX rescue therapy was provided on nondosing days (no supplemental drug was given if rescue therapy was provided before study drug administration on a dosing day). Supplemental ravulizumab was administered within 4 hours of each PLEX session or within 4 hours of the completion of each IVIg cycle. This is consistent with the recommendation in the product monograph.²² A fixed dose of 600 mg of ravulizumab was administered following IVIg (with a minimum infusion time of 0.4 hours to 0.5 hours). Patients received a body weight–based supplemental dose of ravulizumab following PLEX as follows:

• a body weight of 40 kg or greater but less than 60 kg meant if the most recent dose was 2,400 mg, the supplemental dose was 1,200 mg and the minimum infusion time was 1.0 hour; if the most recent dose was 3,000 mg, the supplemental dose was 1,500 mg and the minimum infusion time was 1.2 hours

• a body weight of 60 kg or greater but less than 100 kg meant if the most recent dose was 2,700 mg, the supplemental dose was 1,500 mg and the minimum infusion time was 1.0 hour; if the most recent dose was 3,300 mg, the supplemental dose was 1,800 mg and the minimum infusion time was 1.1 hours

• a body weight of 100 kg or greater meant if the most recent dose was 3,000 mg, the supplemental dose was 1,500 mg and the minimum infusion time was 1.0 hour; if the most recent dose was 3,600 mg, the supplemental dose was 1,800 mg and the minimum infusion time was 1.1 hours.

Allowed concomitant medications included AChE inhibitors, immunosuppressive drugs (corticosteroids, azathioprine, mycophenolate, methotrexate, tacrolimus, cyclosporine, and cyclophosphamide), and rescue therapy with IVIg or PLEX. Patients who entered the study receiving AChE inhibitors were to maintain the dose and schedule unless there was a compelling medical need and the sponsor approval had to be obtained before the change in order for the patient to remain on the study. AChE inhibitors were withheld for at least 10 hours before QMG and Myasthenia Gravis Composite assessments. Patients who entered the study receiving corticosteroids or IST were to maintain their dose and schedule; dose changes needed to be authorized by the sponsor and corticosteroid dose increases could not exceed the dose at baseline. Disallowed concurrent medications included rituximab and eculizumab (or other complement inhibitors).

Rescue therapy with IVIg or PLEX could be administered for clinical deterioration, defined per protocol as 1 of the following:

• there was an MG crisis (respiratory muscle weakness severe enough to necessitate intubation or to delay extubation following surgery, often accompanied by severe bulbar muscle weakness)

• there was a significant symptomatic worsening to a score of 3 points or a 2-point worsening from baseline on any 1 of the individual MG-ADL items other than double vision or eyelid droop

• patient health would be in jeopardy if rescue therapy was not given, in the opinion of the investigator or investigator-designated physician.

Outcomes

A list of efficacy and harms outcomes assessed in this Clinical Review Report is provided in Table 7, followed by descriptions of the outcome measures. Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence⁷² as well as any outcomes identified as important to this review according to the clinical expert consulted by the review team and input from patient and clinician groups and public drug plans. Using the same considerations, the review team selected end points that were considered to be most relevant to inform expert committee deliberations. Note that these outcomes were selected for the analyses of the full population in the CHAMPION-MG trial. However, all the new analyses in the 2 subgroups are post hoc and unadjusted for multiplicity.

Table 7: Main Outcomes Summarized From CHAMPION-MG Study

AE = adverse event; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item Scale - Revised; Neuro-QoL = Quality of Life in Neurological Disorders; QMG = Quantitative Myasthenia Gravis; SAE = serious adverse event.

^aStatistical testing for these end points was adjusted for multiple comparisons in the primary analyses (e.g., hierarchical testing) in the full population in the CHAMPION-MG trial, using a closed testing procedure. If an end point did not achieve statistical significance (P value > 0.05), then subsequent end points in the prespecified order were considered to be not statistically significant and reported P values were considered nominal. Numbers included in parentheses denote testing order. No end points were adjusted for multiple comparisons in the post hoc analyses in the 2 cohorts.

Sources: Clinical Study Report for the CHAMPION-MG trial.²⁴ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Primary Outcome

The primary efficacy outcome was the change from baseline in the MG-ADL total score at week 26. The MG-ADL is an 8-item survey to measure the severity of patient-reported MG symptoms and their impact on patients’ daily activities, with each response rated from 0 (normal) to 3 (most severe); the total MG-ADL score is the sum of the scores of the 8 items and ranges from 0 to 24.^49,50 A reduction in score indicates improvement and the MCID on the MG-ADL scale is estimated to be 2 points based on the change within an individual patient.²⁵ The MG-ADL assessment was performed by a properly trained clinical evaluator, delegated by the investigator. The MG-ADL is a commonly used assessment tool in clinical practice as no objective tests (e.g., blood or other laboratory test) are available for monitoring efficacy in patients with gMG.

Secondary Outcomes

The secondary efficacy outcomes (all measured at week 26) were as follows:

1. change from baseline in QMG total score

2. improvement from baseline of 5 or more points in the QMG total score

3. change from baseline in MG-QoL15r score

4. change from baseline in Neuro-QoL fatigue score

5. improvement from baseline of 3 or more points in the MG-ADL total score.

The QMG is a 13-item clinician-determined assessment of MG disease status, with each item graded for severity from 0 (normal function) to 3 (loss of ability to perform that function); the total score of QMG is the sum of the scores of the 13 items and ranges from 0 to 39.⁵¹ A higher score is indicative of more severe disease. A definite MID for the QMG scale has not been established, although a threshold ranging between 2 points and 3 points depending on disease severity has been suggested in previous studies enrolling patients with MG.^26,27 The MCID for QMG is estimated to be 3 points for patients with higher baseline QMG scores (> 16) within an individual patient.²⁷ A 5-point or greater improvement in the QMG total score from baseline at week 26 was selected as a robust clinical significance threshold for the second hierarchical secondary end point. The QMG assessment was performed by a properly trained clinical evaluator, delegated by the investigator.

The MG-QoL15r questionnaire (a revised version of the Myasthenia Gravis Quality of Life 15-item Scale) is a 15-item patient-reported HRQoL instrument for MG, with each item having 3 response options: 0 (not at all), 1 (somewhat), and 2 (very much); total scores range from 0 to 30, with higher scores indicating a greater extent of dissatisfaction with MG-related dysfunction.⁵² MCIDs have yet to be established for the MG-QoL15r questionnaire.

The Neuro-QoL fatigue subscale is an instrument evaluating the effect of fatigue on the quality of life of patients with neurologic disorders.⁵³ It comprises 19 patient-reported items assessing the frequency of different aspects of fatigue over the previous 7 days, answered using a 5-point scale ranging from 1 (never) to 5 (always); total scores range from 19 to 95. Higher scores indicate greater fatigue and a greater impact of MG on activities. MCIDs have yet to be established for the Neuro-QoL fatigue subscale.

The proportion of patients who experienced a 3-point or greater improvement in MG-ADL total score at week 26 was another secondary end point. The 3-point reduction threshold was selected to exceed the estimated MCID of 2 points for the MG-ADL scale.²⁵

Exploratory Outcomes

The incidence of clinical deterioration was 1 of the exploratory outcomes in the CHAMPION-MG trial. Clinical deterioration was defined as 1 of the following: an MG crisis (respiratory muscle weakness severe enough to necessitate intubation or delay extubation following surgery, often accompanied by severe bulbar muscle weakness); significant symptomatic worsening to a score of 3 points or a 2-point worsening from baseline on any 1 of the individual MG-ADL items other than double vision or eyelid droop; or the fact that patient health would be in jeopardy if rescue therapy were not given, in the opinion of the investigator or investigator-designated physician.

The incidence of an MG crisis was another exploratory outcome.

Harms Outcomes

Safety and tolerability of ravulizumab was evaluated by the incidence of patients with at least 1 AE, SAE, and adverse events of special interest (AESIs) over time. Meningococcal infections were collected as the AESI for this study. All AEs were treatment-emergent (unless otherwise specified) and were summarized according to the number and percentage of patients with events.

AEs were defined as:

• any abnormal laboratory result or other safety assessment

• the exacerbation of a chronic or intermittent pre-existing condition

• new conditions detected or diagnosed after study drug administration

• signs, symptoms, or the clinical sequelae of a suspected drug-drug interaction.

An SAE was defined as an AE at any dose that:

• resulted in death

• was life-threatening

• required hospitalization

• resulted in persistent disability or incapacity.

Table 8: Summary of Outcome Measures and Their Measurement Properties

CI = confidence interval; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MG-QoL15 = Myasthenia Gravis Quality of Life 15-item Scale; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item Scale - Revised; MGC = Myasthenia Gravis Composite; MGFA = Myasthenia Gravis Foundation of America; MID = minimal important difference; MMT = manual muscle testing; Neuro-QoL = Quality of Life in Neurological Disorders; QMG = Quantitative Myasthenia Gravis; r = correlation coefficient; SD = standard deviation; VAS = visual analogue scale.

Statistical Analysis

Statistical analyses of efficacy and harms outcomes in the full population in the CHAMPION-MG trial have been described in detail in the Clinical Review Report of the previous ravulizumab submission.

In the primary analysis of the full population, a sample size of approximately 160 patients was planned to ensure at least 90% nominal power to reject the null hypotheses of no treatment difference for the primary and secondary end points at a 2-sided alpha level of 0.05 and based on a t statistic for 2 independent samples. Sample size calculation in the CHAMPION-MG trial was not based on the concomitant IST optimized cohort or the refractory concomitant IST optimized cohort, which are the focus of this report.

In the full population, type I error was controlled using a closed-loop hierarchical testing strategy to limit the overall 2-sided type I error rate to alpha equalling 0.05. Between the primary and secondary efficacy outcomes, and within the secondary outcomes, a sequential hypothesis testing procedure was used for multiplicity adjustment. The primary outcome (change in the MG-ADL total score from baseline to week 26) was tested first, followed by 5 secondary outcomes (refer to Table 6 for the order of testing). All null hypotheses were tested at alpha equalling 0.05. If statistical significance was not achieved for an end point within the hierarchy, subsequent end points in the prespecified order were not to be considered statistically significant and all P values were considered nominal. Exploratory efficacy outcomes were not included in the hierarchical testing strategy and statistical tests of these outcomes were not adjusted for multiplicity. There was no hierarchical testing procedure used to control the overall type I error to control for multiple comparisons in the 2 post hoc cohorts.

Sensitivity analyses of the primary outcome included a placebo-based analysis (i.e., considering a potential missing not at random mechanism for the missing data) and a tipping point analysis. Sensitivity analyses of both the primary and secondary outcomes included identical mixed model of repeated measures (MMRM) analyses conducted in the per-protocol set and the modified FAS as well as an MMRM excluding randomization stratification region and an MMRM including rescue therapy received in the model.

In the primary analysis, subgroup analyses were performed for all secondary outcomes as per the primary analysis using an MMRM except with an additional subgroup covariate — by treatment interaction. The following prespecified subgroup analyses were performed based on geographic region, sex, race, age at first study drug infusion, IST use at baseline, years from diagnosis to informed consent, baseline MGFA classification, and baseline body weight. The study was not specifically powered to evaluate differences in outcomes among the individual strata.

Post Hoc Analyses in the Concomitant IST Optimized Cohort and Refractory Concomitant IST Optimized Cohort

The statistical analyses used were consistent with the statistical methods used for the primary analysis and followed the methods in the CHAMPION-MG trial’s statistical analysis plan (Table 9). All demographic, baseline, and disease characteristics were summarized by treatment group. The number of patients who received a stable dose of NSIST, corticosteroid, or both per inclusion triggers was summarized by treatment group. Descriptive statistics on the duration of use were provided, concomitant ISTs were used during the RCP, and changes in concomitant MG medication therapies used were summarized.

For primary and secondary continuous end points, changes from baseline at week 26 were analyzed using an MMRM model. The model included the change from baseline score as the response variable, fixed categorical effects of treatment, study visit and treatment-by-study-visit interaction, the randomization stratification variable geographic region, and a fixed covariate of baseline score. A difference in treatment effect between the ravulizumab and placebo treatment groups along with a 2-sided 95% CI and P value were calculated. Missing data were not imputed. In addition to these analyses, differences in treatment effect for the concomitant IST optimized cohort versus the remainder of the FAS for both change from baseline to week 26 in MG-ADL (and QMG) were assessed using an MMRM as specified for the primary end point with the inclusion of cohort status, treatment by cohort status interaction, and treatment by cohort-status-by-visit interaction. For the primary efficacy end point (change in the MG-ADL total score) and key secondary efficacy end points, sensitivity analyses were conducted in the concomitant IST optimized cohort by excluding patients with any increases in IST dosing during the RCP.

The secondary end points of at least a 5-point improvement in QMG total score from baseline to week 26 (and at least a 3-point improvement in MG-ADL) were analyzed using a generalized linear mixed model including the same explanatory covariates as for the continuous end points. Odds ratios (ORs) of the treatment effect (ravulizumab group compared to placebo) along with 2-sided 95% CIs and P values were calculated. An estimate of an OR greater than 1 indicated a beneficial treatment effect.

For the exploratory outcomes of incidence of clinical deterioration and MG crisis, the number of patients who experienced these events was summarized by treatment group.

Table 9: Statistical Analysis of Efficacy End Points in the Cohorts and Full Population in CHAMPION-MG Study

FAS = full analysis set; GLMM = generalized linear mixed model; IST = immunosuppressive therapy; MAR = missing at random; mFAS = modified full analysis set; MG-ADL = Myasthenia Gravis Activities of Daily Living; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item Scale - Revised; MMRM = mixed model of repeated measures; MNAR = missing not at random; OR = odds ratio; PPS = per-protocol set; QMG = Quantitative Myasthenia Gravis; RCP = randomized controlled period.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Analysis Populations

In the CHAMPION-MG study, the FAS included all patients who had received at least 1 dose of the study drug; patients were analyzed according to the treatment they were randomized to receive, regardless of the treatment received. The per-protocol set included all patients in the FAS without any major protocol deviations during the RCP. The overall safety analysis set in the CHAMPION-MG study included all patients who had received at least 1 dose of the study drug; patients were analyzed according to the treatment they actually received, and must have received that treatment for the entire duration of the RCP. The safety set in the 2 post hoc subgroups was a subset of the overall safety analysis set.

The populations used in the cohort analyses in this study are described in Table 10.

Table 10: Populations in Cohort Analyses in the CHAMPION-MG Study During Randomization Controlled Period

CS = corticosteroid; FAS = full analysis set; IST = immunosuppressive therapy; IVIg = IV immunoglobulin; NSIST = nonsteroidal immunosuppressive therapy; PLEX = plasma exchange or plasmapheresis.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Results

Patient Disposition

Patient disposition in the full population and 2 post hoc subgroups of the CHAMPION-MG trial is summarized in Table 11. In the full population, among 242 patients screened, 175 (72.3%) patients were randomized and 67 (27.7%) patients did not advance past screening. The most common reasons for not advancing past screening were not having a positive serologic test for anti-AChR antibodies (25 [10.3%] patients) and an MG-ADL score of 6 or less (16 [6.6%] patients). All 175 of the randomized patients were treated with the study drug.

Thirteen of the 175 (7.4%) randomized patients withdrew from the study before completing the RCP (6 patients in the placebo group and 7 patients in the ravulizumab group). The most common reasons for discontinuation were patient decision (3 patients), physician decision (3 patients), death (2 patients), and AEs (2 patients). Overall, 161 (92%) patients entered the OLE period.

In the concomitant IST optimized cohort █| patients (████) in the ravulizumab group and || (████) in the placebo group withdrew from the study before completing the 26-week treatment. In the refractory concomitant IST optimized cohort, || patients (█████) in the ravulizumab group and || (████) in the placebo group withdrew before completing the 26-week treatment. The reasons for early discontinuation in these 2 cohorts were similar to those in the full population.

Table 11: Patient Disposition in CHAMPION-MG Study

AE = adverse event; FAS = full analysis set; IST = immunosuppressive therapy; mFAS = modified full analysis set; NA = not applicable; NR = not reported; OL = open-label; PPS = per-protocol set.

Note: Patient disposition information in Table 11 was collected through the end of the week 26 randomized controlled period for all patients. Results were presented for patients who received 1 or more doses of ravulizumab in the OLE period and who completed (or would have completed but discontinued) a total of 52 weeks of treatment as of the data cut-off date of May 11, 2021.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Baseline Characteristics

Of the full CHAMPION-MG trial population (n = 175), ██████████████ (███/175 patients = █████) were receiving concomitant IST at baseline and met the dose optimization criteria used in Study ADAPT. In Study ADAPT, patients were required to be on a stable dose of at least 1 treatment for gMG (e.g., AChE inhibitors, corticosteroids, NSISTs) before screening and throughout the trial. These patients formed the concomitant IST optimized cohort; a ███████ number of patients were included in the ravulizumab arm (n ███) and the placebo arm (n = ██). The refractory concomitant IST optimized cohort included ██ patients (█████ of the FAS population) with refractory disease despite concomitant optimized ISTs.

Baseline demographic and disease characteristics of the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort were ██████████ with the full CHAMPION-MG study population in terms of distribution of ███████████████████████████████████████████████ ████████ ████████████████████████████████████████ ██████████████████████████████████████, as shown in Table 12.

The baseline characteristics outlined 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.

Table 12: Baseline Characteristics in CHAMPION-MG Study — Concomitant IST Optimized Cohort Versus Refractory Concomitant IST Optimized Cohort Versus FAS Population

FAS = full analysis population; IST = immunosuppressive therapy; IVIg = IV immunoglobulin; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGFA = Myasthenia Gravis Foundation of America; NSIST = nonsteroidal immunosuppressive therapy; PLEX = plasma exchange or plasmapheresis; QMG = Quantitative Myasthenia Gravis; SD = standard deviation.

Notes: The IST optimized cohort included patients on 1 or more concomitant IST at screening and who met the following criteria: if on a concomitant steroid, the patient had been treated with that steroid for the 3 or more months before screening and had been on a stable dose for 1 or more month; if on a concomitant NSIST, the patient had been treated with that NSIST for the 6 or more months before screening and had been on a stable dose of that NSIST for 3 or more months.

Patients in the refractory concomitant IST optimized cohort met both the concomitant IST optimized cohort criteria as well as the definition of refractory disease, which was defined as not achieving symptom control after 12 months of treatment with either 2 or more ISTs, or with 1 or more IST and chronic IVIg or PLEX.

Exposure to Study Treatments

During the RCP, treatment compliance was 94.4% in the placebo group and 96.5% in the ravulizumab group in the full population. Reasons for missed infusions were documented as either COVID-19–related or “other” reasons (due to AEs and unknown). Per the protocol, 12 (13.5%) patients in the placebo group and 7 (8.1%) patients in the ravulizumab group were administered supplemental study drug doses when PLEX or IVIg rescue therapy was administered.

In the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort, the results of study duration and treatment duration were similar to those observed in the full population (Table 13). The mean study duration in the ravulizumab group was 202.5 days, █████ days, and █████ days in the FAS, the concomitant IST optimized cohort, and the refractory concomitant IST optimized cohort, respectively. The mean study duration in the placebo group was 198.9 days, █████ days, and █████ days in the FAS, the concomitant IST optimized cohort, and the refractory concomitant IST optimized cohort, respectively. The mean treatment duration in the ravulizumab group was 179.6 days, █████ days, and █████ days in the FAS, the concomitant IST optimized cohort, and the refractory concomitant IST optimized cohort, respectively. The mean treatment duration in the placebo group was 176.5 days, █████ days, and █████ days in the FAS, the concomitant IST optimized cohort, and the refractory concomitant IST optimized cohort, respectively.

Table 13: Patient Exposure During Randomized Controlled Period of the CHAMPION-MG Study (FAS Population and Post Hoc Cohorts)

FAS = full analysis set; IST = immunosuppressive therapy; SD = standard deviation.

Sources: Clinical Study Report for the CHAMPION-MG trial,²⁴ cohort analyses of the CHAMPION-MG trial,²⁸ and the sponsor’s Summary of Clinical Evidence.

Table 14 provides details on the median doses of the concomitant IST for the concomitant IST optimized cohort (n = ███) and the refractory concomitant IST optimized cohort (n = ██) at study enrolment. Doses are broken down into 3 categories based on the IST(s) received at the time of study enrolment: NSIST only, corticosteroid only, or corticosteroid in combination with NSIST. Note that the following drugs are considered usual care in the CHAMPION-MG trial: pyridostigmine, prednisone, azathioprine, cyclosporine, mycophenolate mofetil, and tacrolimus. Table 38 in Appendix 1 summarizes the types and frequencies of patients’ use of ISTs at baseline in the CHAMPION-MG study’s FAS population.

Table 14: Median Stable Doses of IST Treatments in Concomitant IST Optimized Cohort and Refractory Concomitant IST Optimized Cohort in the CHAMPION-MG Study at Time of Study Enrolment

AZA = azathioprine; CS = corticosteroid; CyS = cyclosporine; IST = immunosuppressive therapy; MMF = mycophenolate mofetil; MTX = methotrexate; NA = not applicable; NSIST = nonsteroidal immunosuppressive therapy; TAC = tacrolimus.

Note: The mg/kg doses were reported for select ISTs, which were dosed based on body weight.

Sources: Cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

The median durations of IST treatment at baseline in the cohort populations were longer than the minimum inclusion requirements based on Study ADAPT (Table 15).⁶⁰ In the concomitant IST optimized cohort, patients receiving corticosteroids had a median treatment duration of █████████████████ in the placebo group and ███████████████████ in the ravulizumab group. Similar results were observed in the refractory concomitant IST optimized cohort (the median duration was ███████████████ in the placebo and ravulizumab groups, respectively), reflecting a far longer duration than the minimum requirement of 3 months in the cohort inclusion criteria. Similarly, the median duration of azathioprine treatment surpassed the minimum criteria of 6 months by a wide margin; the median duration of azathioprine was over █████████████) and over ██████████████) in the placebo group and ravulizumab group, respectively. This was also observed in the refractory concomitant IST optimized cohort, with a median treatment duration of █████ days and ████ days in the placebo and ravulizumab groups, respectively. Mycophenolate mofetil was the most commonly used NSIST apart from azathioprine in both cohorts. The median duration of mycophenolate mofetil was over █████████████) in the placebo group and over ███████████████) in the ravulizumab group in the concomitant IST optimized cohort, whereas in the refractory concomitant IST optimized cohort, the median duration was ███ days and █████ days in the placebo and ravulizumab groups, respectively. In addition to exceeding the minimum criteria for inclusion in the concomitant IST optimized cohort, the median durations of corticosteroid and NSIST use also exceeded the typical IST treatment duration for maximal responses of 2 months to 6 months for corticosteroids, and 9 months to 18 months for NSISTs as outlined in the clinician input received by CDA-AMC.^17,61

Table 40 and Table 41 in Appendix 1 summarize mean, median, range, and at least 6 months’ and at least 12 months’ duration of treatment for ISTs before the first dose of the study drug in the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort.

Table 15: Median Duration of IST Treatment in Concomitant IST Optimized Cohort and Refractory Concomitant IST Optimized Cohort at Time of Study Enrolment

AZA = azathioprine; CS = corticosteroid; CyS = cyclosporine; FAS = full analysis set; IST = immunosuppressive therapy; MMF = mycophenolate mofetil; MTX = methotrexate; NA = not applicable; TAC = tacrolimus.

Sources: Cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 39 in Appendix 1 summarizes the changes (replacement, addition or removal, or dose and/or frequency adjustment) in concomitant ISTs in the safety analysis set in the CHAMPION-MG trial.

Efficacy

MG-ADL Total Score

The MG-ADL total score ranges from 0 to 24, and higher scores indicate a greater severity of symptoms and a more significant impact on a patient’s daily activities. Although no MID has been estimated, an improvement of approximately 2 points in the total MG-ADL score is a recommended response threshold that indicates clinical improvement at the level of individual patients with MG.²⁵

In the concomitant IST optimized cohort, during the RCP, the LSM (95% CI) change from baseline to week 26 in the MG-ADL total score was █████████████████ in the ravulizumab group compared to █████████████████ in the placebo group, as summarized in Table 16. The LSM treatment difference was -████████████████ |███████. Differences in treatment effect for the concomitant IST optimized cohort versus the remainder of the FAS for change from baseline to week 26 in the MG-ADL total score were assessed using an MMRM. The results showed that there was █████████████████████ between the cohort and the remainder of the FAS (the LSM treatment difference was █████████████████, supporting the ██████████████████████████████████ determined by visual inspection.

In the refractory concomitant IST optimized cohort, the LSM (95% CI) change from baseline to week 26 in the MG-ADL total score was █████████████████ in the ravulizumab group compared to █████████████████ in the placebo group. The LSM treatment difference was █████████████████████████. Differences in treatment effect for the refractory concomitant IST optimized cohort versus the remainder of the FAS for change from baseline to week 26 in the MG-ADL total score were assessed using an MMRM. The results showed that there was █████████████████████ between the cohort and the remainder of the FAS (the LSM treatment difference was █████████████████, supporting the ██████████████████████████████████ determined by visual inspection.

The treatment effect of ravulizumab over placebo observed in the 2 cohorts is consistent with results from the primary analysis population in the CHAMPION-MG trial. Sensitivity analyses for the MG-ADL total score from baseline to week 26 were performed in the FAS, and the results were consistent with those from the primary analysis. Sensitivity analyses were performed in the concomitant IST optimized cohort by removing patients with increases in IST dosing during the RCP. Results of the sensitivity analysis were aligned with those in the primary analysis. Please refer to Table 39 (Appendix 1) for changes in IST dosing during the RCP in the safety analysis set in the CHAMPION-MG trial.

Table 16: Change From Baseline to Week 26 in MG-ADL Total Score in 2 Cohorts and FAS

CI = confidence interval; FAS = full analysis set; IST = immunosuppressive therapy; LSM = least squares mean; MG-ADL = Myasthenia Gravis Activities of Daily Living; MMRM = mixed model of repeated measures; SD = standard deviation; SEM = standard error of the mean.

^aEstimates were based on an MMRM that included treatment group, stratification factor region, and MG-ADL total score at baseline, study visit and study-visit-by-treatment-group interaction. The P value was from a restricted maximum likelihood–based MMRM that tested whether the LSMs for the 2 treatments were equal. An unstructured covariance structure was used. P values were unadjusted for multiplicity; as a result, there was an increased risk of type I error (i.e., falsely rejecting the null hypothesis).

^bThe P value was from an MMRM that included change from baseline at postdosing visits as the response variable; fixed categorical effects of treatment, study visit and treatment-by-study-visit interaction; the randomization stratification variable geographic region; and the fixed covariate of baseline score.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

QMG Total Score

The total QMG score ranges from 0 to 39, where higher scores indicate greater disease severity. A definite MID for the QMG scale has not been established, although a threshold ranging between 2 points and 3 points depending on disease severity was suggested for patients with MG.^26,27

In the concomitant IST optimized cohort, the LSM (SEM) change from baseline to week 26 in the QMG total score was █████████ in the ravulizumab group and █████████ in the placebo group (Table 17). The LSM treatment difference was █████████████████████████. Differences in treatment effect for the concomitant IST optimized cohort versus the remainder of the FAS for change from baseline to week 26 in the QMG score were assessed using an MMRM. The results showed that there ███████████████████████ between the cohort and the remainder of the FAS (the LSM treatment difference was █████████████████, supporting the ██████████████████████████████████ determined by visual inspection.

In the refractory concomitant IST optimized cohort, the LSM (SEM) change from baseline to week 26 in the QMG total score was █████████ in the ravulizumab group and -████████ in the placebo group. The LSM treatment difference was ████████████████████████. Differences in treatment effect for the refractory concomitant IST optimized cohort versus the remainder of the FAS for change from baseline to week 26 in the QMG score were assessed using an MMRM. The results showed that there was █████████████████████ between the cohort and the remainder of the FAS (the LSM treatment difference was ████████████████, supporting ████████████████████████████████████ determined by visual inspection.

Results of the 2 cohorts were consistent with those in the primary analysis conducted in the FAS of the CHAMPION-MG trial. A sensitivity analysis was performed in the concomitant IST optimized cohort by removing patients with increases in IST dosing during the RCP. Results of the sensitivity analysis were aligned with those in the primary analysis.

Table 17: Change From Baseline to Week 26 in QMG Total Score in 2 Cohorts and FAS

CI = confidence interval; FAS = full analysis set; IST = immunosuppressive therapy; LSM = least squares mean; MMRM = mixed model of repeated measures; QMG = Quantitative Myasthenia Gravis; SD = standard deviation; SEM = standard error of the mean.

^aEstimates were based on an MMRM that included treatment group, stratification factor region, and QMG total score at baseline, study visit and study-visit-by-treatment-group interaction. The P value was from a restricted maximum likelihood–based MMRM that tested whether the LSMs for the 2 treatments were equal. An unstructured covariance structure was used. P values were unadjusted for multiplicity; as a result, there was an increased risk of type I error (i.e., falsely rejecting the null hypothesis).

^bThe P value was from an MMRM that included change from baseline at postdosing visits as the response variable; fixed categorical effects of treatment, and study visit and treatment-by-study visit interaction; the randomization stratification variable geographic region; and the fixed covariate of baseline score.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

QMG 5-Point Response

Clinical responder was defined as patients who achieved a greater than 5-point improvement in the QMG total score; this is larger than the proposed MID of 2 points to 3 points.^26,27

In the concomitant IST optimized cohort, █████ of patients in the ravulizumab group compared to ████ in the placebo group achieved at least a 5-point improvement. The between-group difference was ███████████████████ and the OR was ████████████████████████████████████████ ███████ (Table 18).

In the refractory concomitant IST optimized cohort, the proportion of patients meeting the clinical responder definition for QMG improvement was █████ in the ravulizumab group versus ████ in the placebo group, with a between-group difference of ███████████████████ and an OR of ███████████████████████████████████████████████.

Results in the 2 cohorts aligned with those in the primary analysis: the QMG improvement was █████ in the ravulizumab group versus █████ in the placebo group (██████████████████████████████████████ █| █████████████████████████.

Table 18: Improvement of 5 or More Points in the QMG Score From Baseline to Week 26 in 2 Cohorts and FAS

CI = confidence interval; FAS = full analysis set; IST = immunosuppressive therapy; NR = not reported; OR = odds ratio; QMG = Quantitative Myasthenia Gravis.

^aBaseline was defined as the last available assessment value before the first study drug infusion. In the event that a cholinesterase inhibitor was not withheld for at least 10 hours before administration of the QMG assessment, the screening visit assessment was to be used as baseline. Estimates were based on a generalized linear mixed model that included treatment group, stratification factor region and QMG total score at baseline, study visit and study-visit-by-treatment-group interaction. An autoregressive covariance structure was used. P values were unadjusted for multiplicity; as a result, there was an increased risk of type I error (falsely rejecting the null hypothesis).

^bBaseline was defined as the last available assessment value before the first study drug infusion. In the event that a cholinesterase inhibitor was not withheld for at least 10 hours before administration of the QMG assessment, the screening visit assessment was to be used as baseline. Estimates were based on a generalized linear mixed model that included treatment group, stratification factor region and QMG total score at baseline, study visit and study-visit-by-treatment-group interaction. An unstructured covariance structure was used.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

MG-QoL15r Total Score

MG-QoL15r is used to measure a patient’s MG-specific HRQoL. The cumulative scores in the MG-QoL15r tool range from 0 to 60, with higher scores representing a worse quality of life.⁵² An MID for patients with MG has not been estimated.

This outcome did not reach statistical significance in the FAS population and therefore the P values for the subsequent secondary end points included in the prespecified hierarchical testing order (i.e., change from baseline in Neuro-QoL fatigue score at week 26 and an MG-ADL 3-point response at week 26) were considered nominal. In the concomitant IST optimized cohort, the LSM change from baseline to week 26 in the MG-QoL15r total score was ████████████████ in the ravulizumab group and ████████████████ in the placebo group during the RCP. The LSM treatment difference was ███████████████████████.

In the refractory concomitant IST optimized cohort, the LSM change from baseline to week 26 in the MG-QoL15r total score was ████████████████ in the ravulizumab group and ████████████████ in the placebo group during the RCP. The LSM treatment difference was ████████████████████████.

The results of the 2 cohorts are aligned with the primary analysis in the FAS (Table 19).

Table 19: Change From Baseline in MG-QoL15r Total Score at Week 26 in 2 Cohorts and FAS

CI = confidence interval; FAS = full analysis set; IST = immunosuppressive therapy; LSM = least squares mean; MG-QoL15r = Myasthenia Gravis Quality of Life 15-item Scale - Revised; MMRM = mixed model of repeated measures; NR = not reported; SD = standard deviation; SEM = standard error of the mean.

^aEstimates were based on an MMRM that included treatment group, stratification factor region and MG-QoL15r score at baseline, study visit and study-visit-by-treatment-group interaction. The P value was from a restricted maximum likelihood–based MMRM that tested whether the LSMs for the 2 treatments were equal. An unstructured covariance structure was used.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Neuro-QoL Fatigue Score

The Neuro-QoL fatigue scale is a generic 19-item patient-reported survey of fatigue with a score that ranges from 19 to 95. Higher scores represent greater fatigue and a greater impact of MG on a patient’s activities.^24,53 An MID of Neuro-QoL fatigue score has not been established in patients with MG. This outcome was tested after a prior nonsignificant result of the hierarchical testing procedure and therefore P values were considered nominal in the FAS population.

In the concomitant IST optimized cohort, the LSM change from baseline to week 26 in the Neuro-QoL fatigue score was ██████████████████ in the ravulizumab group and ████████████████ in the placebo group during the RCP. The LSM treatment difference was ████████████████████████.

In the refractory concomitant IST optimized cohort, the LSM change from baseline to week 26 in the Neuro-QoL fatigue score was █████████████████ in the ravulizumab group and █████████████████ in the placebo group during the RCP. The LSM treatment difference was ████████████████████████.

The results of the ██████████████████████████████████████████████████████████████████████████████████████████ (Table 20).

Table 20: Change From Baseline in Neuro-QoL Fatigue Score at Week 26 in 2 Cohorts and FAS

CI = confidence interval; FAS = full analysis set; IST = immunosuppressive therapy; LSM = least squares mean; MMRM = mixed model of repeated measures; Neuro-QoL = Quality of Life in Neurological Disorders; NR = not reported; SEM = standard error of the mean; vs. = versus.

^aEstimates were based on an MMRM that included treatment group, stratification factor region and Neuro-QoL fatigue score at baseline, and study visit and study-visit-by-treatment-group interaction. The P value was from a restricted maximum likelihood–based MMRM that tested whether the LSMs for the 2 treatments were equal. An unstructured covariance structure was used.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

MG-ADL 3-Point or Greater Response

Patients who achieved at least a 3-point improvement in their MG-ADL total score were considered responders; this exceeds the estimated MID of 2 points.^25,54 This outcome was tested after a prior nonsignificant result of the hierarchical testing procedure and therefore is at an increased risk of type I error (falsely rejecting the null hypothesis).

In the concomitant IST optimized cohort, the proportion of patients who met the clinical responder definition for MG-ADL improvement of at least 3 points was █████████████████ in the ravulizumab group compared to ██████████████████ in the placebo group, with a between-group difference of ███████████████████ and an OR of ███████████████████████████████████ ███████.

In the refractory concomitant IST optimized cohort, the proportion of patients meeting the clinical responder definition for MG-ADL improvement was █████████████████ in the ravulizumab group compared to █████████████████ in the placebo group, with a between-group difference of ███████████████████ and an OR of ████████████████, ███████████████████ (Table 21).

The results in the 2 cohorts were aligned with the primary analysis in the FAS.

Table 21: Improvement of 3 or More Points in the MG-ADL Total Score From Baseline at Week 26 in 2 Cohorts and FAS

CI = confidence interval; FAS = full analysis set; IST = immunosuppressive therapy; MG-ADL = Myasthenia Gravis Activities of Daily Living; OR = odds ratio.

^aEstimates were based on a generalized linear mixed model that included treatment group, stratification factor region and MG-ADL total score at baseline, study visit and study-visit-by-treatment-group interaction. An unstructured covariance structure was used. This analysis was unadjusted for multiplicity; as a result, there was an increased risk of type I error (falsely rejecting the null hypothesis).

^bEstimates were based on a generalized linear mixed model that included treatment group, stratification factor region and MG-ADL total score at baseline, study visit and study-visit-by-treatment-group interaction. A Toeplitz covariance structure was used.

^cThis analysis occurred after a failure of the testing hierarchy. There was an increased risk of type I error (falsely rejecting the null hypothesis).

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Incidence of Clinical Deterioration and MG Crisis

During the RCP, in the concomitant IST optimized cohort, || patients (████) in the ravulizumab group and ██ patients (█████) in the placebo group reported clinical deterioration. In the refractory concomitant IST optimized cohort, || patients (█████) in the ravulizumab group and || patients (█████) in the placebo group reported clinical deterioration.

In each cohort, an MG crisis was reported by || patient from the placebo group. No patients from the ravulizumab group reported an MG crisis (Table 22).

Table 22: Incidence of Clinical Deterioration and MG Crisis During RCP in 2 Cohorts and FAS

FAS = full analysis set; IST = immunosuppressive therapy; MG = myasthenia gravis; RCP = randomized controlled period.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Harms

Results for the safety end points in the 2 cohorts are presented in Table 23. The incidence of treatment-emergent AEs and SAEs during the RCP in the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort were ██████████ with the harm results from the primary analysis.

Adverse Events

The percentage of patients with AEs was similar between the treatment groups (████████ of placebo-treated patients and ████████ of ravulizumab-treated patients in the concomitant IST optimized cohort and ████████ of placebo-treated patients and ████████ of ravulizumab-treated patients in the refractory concomitant IST optimized cohort). The most commonly reported AEs in these 2 cohorts included ████████ (concomitant IST optimized cohort = █████ in the ravulizumab group versus █████ in the placebo group; refractory concomitant IST optimized cohort = █████ in the ravulizumab group versus █████ in the placebo group) and ████████ (concomitant IST optimized cohort = █████ in the ravulizumab group versus █████ in the placebo group; refractory concomitant IST optimized cohort = █████ in the ravulizumab group versus █████ in the placebo group). Most AEs were █████ ███ █████████.

In the safety set of the CHAMPION-MG trial, during the RCP, most patients (90.7% of ravulizumab-treated patients and 86.5% of placebo-treated patients) experienced AEs. The most common AEs were headache (25.8% of patients in the ravulizumab group and 18.6% of patients in the placebo group), diarrhea (12.4% of patients in the ravulizumab group and 15.1% of patients in the placebo group) and nausea (10.1% of patients in the ravulizumab group and 10.5% of patients in the placebo group).

Serious Adverse Events

The percentage of patients with SAEs was higher in the ravulizumab group compared to the placebo group. In the concomitant IST optimized cohort, █████ of ravulizumab-treated patients and █████ of placebo-treated patients experienced at least 1 SAE. In the refractory concomitant IST optimized cohort, █████ of ravulizumab-treated patients and █████ of patients in the placebo group experienced at least 1 SAE. Commonly reported SAEs were not reported for these 2 cohorts.

Withdrawals Due to Adverse Events

In both cohorts, there was || patient treated with ravulizumab who discontinued the study drug due to an AE.

Mortality

In both cohorts, there were || patients treated with ravulizumab who died, and ██ death was reported for the placebo group.

Notable Harms

Meningococcal infection was considered a notable harm for treatment with ravulizumab. No events were reported during the RCP in the entire study population.

Table 23: Overview of All Treatment-Emergent Adverse Events and Serious Adverse Events During the Randomized Controlled Period in the Concomitant IST Optimized, Refractory Concomitant IST Optimized Cohorts and Full Safety Set

AE = adverse event; IST = immunosuppressive therapy; SAE = serious adverse event; WDAE = withdrawal due to adverse event.

Sources: Clinical Study Report for the CHAMPION-MG trial²⁴ and cohort analyses of the CHAMPION-MG trial.²⁸ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Critical Appraisal

Internal Validity

This is a resubmission of the initial ravulizumab review. The CHAMPION-MG trial (N = 175) was the pivotal study. This was a phase III, double-blind, multicentre, placebo-controlled RCT with an OLE period of up to 4 years. The primary objective of the CHAMPION-MG trial was to evaluate the safety and efficacy of ravulizumab compared with placebo in complement inhibitor–naive adult patients with anti-AChR antibody–positive gMG. CDEC issued a negative reimbursement recommendation in 2023 (refer to the Submission History section in this report for details). The current review focuses on 2 post hoc subgroups that were identified from the FAS in the CHAMPION-MG trial: the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort.

According to the critical appraisal of the CHAMPION-MG trial in the initial ravulizumab submission, there were no major validity concerns regarding this trial. █████ baseline imbalances between study arms were viewed as unlikely to be prognostic or to significantly affect the study results. Study discontinuations before completing the RCP were relatively ██████████ (████ of patients) and missing data for reasons other than discontinuation were █████████████ (███ ████████ in each arm, depending on outcome). During the study, concomitant MG therapy was generally ███████ in both study arms, apart from ███████████████ █████████████████████████████ █████████████. Important protocol deviations occurred █████████ in both arms and were not viewed as likely to impact the study results. The instruments used to evaluate the primary and secondary efficacy outcomes (MG-ADL, QMG, MG-QoL15r, and Neuro-QoL fatigue) were appropriate and their psychometric properties have been investigated in patients with MG, although no MIDs have been estimated for the MG-QoL15r and Neuro-QoL fatigue scores. Statistical tests were appropriate overall, power was adequate for the primary analysis, and multiplicity was controlled using a hierarchical testing strategy. Statistical testing for a 3-point or greater improvement in the MG-ADL score and changes from baseline in the Neuro-QoL fatigue score at 26 weeks occurred after a prior nonsignificant result (MG-QoL15r total scores) in the statistical hierarchy, so there is an increased risk of type I error. Subgroup analyses identified were not adjusted for multiplicity nor were they specifically powered to detect differences among strata. In addition, an MID for MG-QoL15r or Neuro-QoL fatigue in patients with MG has not been estimated. Definite conclusions regarding the effect of the study drug on HRQoL cannot be drawn, due to the small sample size in the cohorts (HRQoL results were not available in 11% to 15% of patients treated with ravulizumab), and 95% CIs that were wide and crossed the null.

Post hoc analyses that are not prespecified in a trial’s analysis plan can be at risk of bias due to selective reporting (i.e., there is a risk that the presented results are selected from multiple analyses of the data based on their direction, magnitude, or statistical significance). However, the sponsor furnished clear justification for the selected subpopulations using criteria informed by Study ADAPT and Study REGAIN to provide directly relevant information related to a previous negative reimbursement recommendation. Additionally, the same prespecified analysis methods of the CHAMPION-MG trial were used to analyze the subgroup data and results for all relevant end points were presented. As a result, any concern for selective reporting was minimized.

Whether patients had received an optimized IST was not a stratification factor in the randomization of the CHAMPION-MG trial; therefore, there is a risk that the randomization was not fully upheld in the subgroups. Nevertheless, in these 2 subgroups, patients’ baseline demographic and disease characteristics were generally ███████████ between treatment groups and were similar to the FAS population. Also ███████ to the FAS, ██████████████ were observed for ████ ███████ █████████████████ ███████████████████████████████████████████ █████████████████████████ ██████████████████████) and ████████████████████████. The imbalances in the FAS could be due to the small sample size, which would have been ███████████ in the smaller subgroups. The clinical expert consulted by the review team noted that these may not have a significant impact on result interpretation.

Tests for subgroup differences between the subgroup populations and the rest of the FAS were performed for the MG-ADL score and QMG score. However, these analyses were post hoc and the CHAMPION-MG trial was not powered to find a difference between the groups. Otherwise, results from these post hoc subgroups can be interpreted for consistency with the main analyses of the CHAMPION-MG trial as they share the same limitations of those analyses. In addition, it should be considered that the small sample size in these 2 subgroups (the concomitant IST optimized cohort accounted for ███ of the full population and the refractory concomitant IST optimized cohort was ████ of the full population) could have resulted in insufficient power to detect true between-group differences, and multiplicity was not controlled using hierarchical testing. Therefore, there was an increasing risk of type I error (i.e., falsely rejecting the null hypothesis) rate for the investigated outcomes that achieved statistical significance at a conventional alpha of 0.05.

Efficacy data during the OLE period for the 2 cohorts were ██████████ with those from the RCP for the 2 cohorts and the FAS. Patients who switched from placebo to ravulizumab showed █████████ improvements in daily activities and other efficacy outcomes at week 164. However, further interpretation of these data was limited by the open-label and descriptive nature of the extension study. Attrition bias is also a concern in OLE studies; it increases the uncertainty in the study results.

External Validity

The drug was resubmitted for review by the sponsor on the basis of the availability of new evidence and a revised reimbursement request for ravulizumab as an add-on therapy for adult patients with anti–AChR antibody–positive gMG whose symptoms persist despite adequate treatment with AChE inhibitors, corticosteroids, and/or NSISTs. The new evidence provided by the sponsor consisted of 2 post hoc subgroups in the CHAMPION-MG trial: the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort. These cohorts aligned with the reimbursement request in that patients in these cohorts had anti-AChR antibody–positive gMG and were symptomatic despite having had “adequate” ISTs (i.e., corticosteroids and/or NSISTs). However, the reimbursement request also includes patients who have had an adequate trial of AChE inhibitors and no ISTs; these patients were not included in the 2 post hoc cohorts but were studied in the FAS in the CHAMPION-MG trial. A similar proportion of patients who received AChE inhibitors but no ISTs were enrolled in the CHAMPION-MG trial and Study ADAPT (i.e., approximately 10% of the overall trial populations). The review team noted that the conditions for reimbursement for efgartigimod alfa align with the sponsor’s reimbursement request for this resubmission. CDEC noted in the final CDA-AMC recommendation for efgartigimod alfa that it would be reasonable to consider efgartigimod alfa after a trial of at least 1 of AChE inhibitors, corticosteroids, and/or NSISTs in the previous 12 months. The NMD4C clinician group, which provided input to this resubmission for ravulizumab, agreed that the place in therapy for ravulizumab would also include patients who had received AChE inhibitors but not ISTs. However, both the clinician group and the clinical expert consulted for this review noted that it would be reasonable for patients with gMG to try other treatments before initiating ravulizumab, which requires periodic IV infusions over an extended period, is likely to be expensive, may not be available in smaller cities or nonspecialized centres, and requires extensive expertise. The clinical expert consulted for this review noted that most clinicians would likely continue offering AChE inhibitors and ISTs before ravulizumab.

Patient selection for the 2 post hoc cohorts was based on the inclusion criteria from other RCTs: Study ADAPT and Study REGAIN. According to the clinical expert consulted for this review, the definitions used to identify patients are reasonable and acceptable in clinical practice. These 2 subgroups reflect patients who did not experience improvement on a prior gMG treatment (known as patients with refractory gMG) and patients whose disease responded on a prior gMG treatment but had responded inadequately to existing standards of gMG therapy. According to the clinical expert consulted for this review, the population included in the 2 cohorts is adequately reflective of the patients who experience unmet needs in the treatment of gMG in clinical settings in Canada.

According to Table 14 and Table 15 in this report, the median duration of prior corticosteroid and NSIST use exceeded the typical IST treatment duration for maximal responses (2 months to 6 months for corticosteroids, and 9 months to 18 months for NSISTs) as outlined in the clinician input received for the initial ravulizumab submission. In addition, the median treatment durations for IST treatment surpassed the minimum criteria for inclusion in the concomitant IST optimized cohort. The clinical expert consulted for this review had no concerns about suboptimal management with IST treatment (i.e., inadequate dosing or duration of therapy) in patients at the time of enrolment in the CHAMPION-MG trial. The clinical expert agreed that median durations of corticosteroid and NSIST treatment appeared adequate and allowed sufficient time to achieve maximal responses with IST. Therefore, the patient population in the 2 cohorts aligned with the reimbursement request in that patients with anti-AChR antibody–positive gMG were symptomatic despite “adequate” ISTs (i.e., corticosteroids and/or NSISTs).

Concomitant conventional therapy such as AChE inhibitors and ISTs (e.g., corticosteroids, azathioprine, mycophenolate mofetil, tacrolimus) were allowed during the CHAMPION-MG trial. As per the protocol of the CHAMPION-MG study, the dosage of these drugs was not allowed to be changed and no new ISTs were allowed to be added throughout the RCP, unless deemed medically necessary. Rescue therapy (e.g., a high-dose corticosteroid, PLEX, IVIg) was allowed if a patient experienced protocol-defined clinical deterioration. Similar protocol requirements for concomitant ISTs are common across trials in patients with gMG to ensure consistency and to prevent the confounding of trial results. Notably in Study ADAPT (efgartigimod alfa versus placebo) and Study REGAIN (eculizumab versus placebo), a change in the type or dose and/or regimen of standard of care (replacing, adding, or removing standard of care, or adjusting the standard of care dose and/or frequency) was not allowed unless deemed medically necessary (standard of care referred to NSISTs, steroids, and AChE inhibitors). Because the concomitant conventional therapy was required to remain stable except in the case of rescue therapy, ravulizumab was not compared to any individual or combination conventional therapy as it would typically be used in clinical practice (i.e., altering doses or adding additional medications to suit patients’ current symptoms or other needs). According to the sponsor for the present submission, a change of ISTs (their introduction or discontinuation, or an increased or decreased dosage) occurred in 17% of patients overall (12% in the placebo group and 20% in the ravulizumab group) during the RCP of the CHAMPION-MG trial. The clinical expert consulted for this review was not concerned about discouraging dose changes in concomitant medications over the course of the 26 weeks in the RCP. The clinical experts noted that patients were on optimized and stable concomitant therapy entering the trial and were allowed to continue those therapies. Rescue therapy, including high-dose corticosteroids, PLEX, or IVIg, was permitted throughout the study for patients experiencing clinical deterioration, which was considered appropriate by the clinical expert consulted for this review. During the OLE (OLE data were available for up to 3.5 years of follow-up), physicians could adjust ISTs and changes to MG medication occurred in the majority of patients (68.3%). The most common reason for change was due to improvements in MG symptoms (43.5% of patients). By the end of the study, 30.1% (34 of 113) of patients had decreased their daily dose of corticosteroid therapy and 12.4% (14 of 113) of patients had stopped corticosteroid therapy.

According to the clinical expert consulted for this review, the eligibility criteria for patients of an MGFA class II to class IV and an MG-ADL total score of 6 or more would select patients with symptomatic gMG most in need of intervention. Patients with MGFA class I and class V were excluded. Very few patients with class IV gMG were included in the CHAMPION-MG trial (11 patients in the FAS). Whether the findings of the CHAMPION-MG trial can be generalized to patients with MGFA class I or MGFA class V is uncertain. The clinical expert consulted for this review indicated that a subset of patients with an MGFA class I (ocular MG) or class V and MG-ADL scores of less than 6, who were excluded from the trial, 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 patients with MGFA class V (on a ventilator) who have no contraindications would potentially benefit from ravulizumab. However, the results of the trial cannot be directly generalized to these groups of patients.

The median baseline MG-ADL score was approximately 9 points and the median QMG score of 14 points to 15 points across study groups in the FAS and the 2 post hoc cohorts. In addition, the IST treatment history of the cohort populations (Table 15 of this report) reflected a heavily pretreated patient population that, despite a lengthy duration of IST treatment, continued to experience significant gMG symptoms. The MG-ADL total score cut-off value was 6 or more points as an inclusion criterion in the CHAMPION-MG trial and in previous gMG clinical trials (i.e., in Study REGAIN). However, the review team noticed that the inclusion criteria for recruiting patients in Study ADAPT was 5 or more points on the MG-ADL total score. Patients in Study ADAPT who had 5 points on the MG-ADL total score at baseline were relatively few. The clinical expert consulted for this review indicated that severity status based on the MG-ADL was subjective; a difference of 1 point may not be clinically important in terms of disease severity and is unlikely to have a meaningful impact on the patient’s response to the treatment. Whether the results of the CHAMPION-MG trial can be generalized to patients who have an MG-ADL total score of less than 6 despite a stable dose of conventional therapies remains uncertain.

The RCP of the CHAMPION-MG trial was 26 weeks. The longer-term treatment effect of ravulizumab can only be assessed in the OLE period of this study. Moreover, the CHAMPION-MG trial did not provide evidence for comparisons between ravulizumab and other currently available active treatments for gMG.

Long-Term Extension Studies

Content in this section has been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

Description of Studies

The CHAMPION-MG OLE, with the last patient’s last visit being on May 25, 2023, represents up to 3.5 years of ravulizumab treatment exposure. At the end of the 26-week RCP, all patients were eligible to enter the OLE and receive open-label ravulizumab. Following the 26-week RCP, patients in the placebo group received a blinded loading dose of ravulizumab and patients in the ravulizumab group received a blinded ravulizumab dose of 900 mg. Starting week 28, all patients began open-label ravulizumab maintenance doses every 8 weeks. The OLE period for each patient started when the patient received their dose of ravulizumab on week 26 (day 183) and was intended to continue for up to 4 years or until the product is registered or approved and available by prescription.

Populations

There were no additional criteria to continue from the CHAMPION-MG study to the extension study (refer to section 2 eligibility criteria). All patients were eligible to receive open-label ravulizumab during the OLE period.

Interventions

During the OLE period, patients were given a blinded dose of ravulizumab at week 26 (day 183). As of 2 weeks following the blinded dose (week 28), all patients began receiving open-label ravulizumab maintenance doses every 8 weeks. Supplemental ravulizumab (or placebo) dosing was required following PLEX or IVIg rescue treatment on nondosing days. While all patients who entered the OLE period received ravulizumab, patients randomized to the ravulizumab group had previously received 26 weeks of ravulizumab treatment during the RCP (the ravulizumab-ravulizumab treatment group). Patients randomized to receive placebo during the RCP had their first dose of ravulizumab at the start of the OLE period (the day 183 visit in the placebo-ravulizumab treatment group).

Outcomes

The efficacy and safety outcomes assessed in the OLE were the same as the outcomes assessed during the RCP.

The primary efficacy end point was the MG-ADL total score. The key secondary end points were the QMG total score, a QMG 5-point or greater improvement, the MG-QoL15r total score, the Neuro-QoL fatigue score, and an MG-ADL 3-point or greater improvement. Data on clinical deterioration and MG crisis were not available.

Safety outcomes included infusion interruptions, SAEs, treatment-emergent AEs, common AEs, withdrawals due to adverse events (WDAEs), deaths, and AESIs.

Statistical Analysis

For continuous variables, the change from baseline, defined as the last available assessment value before the first study drug infusion during the RCP, was analyzed using an MMRM. The RCP estimates were based on an MMRM that included treatment group, stratification factor region, baseline score, and study visit and study-visit-by-treatment-group interaction. Visits up to week 26 were included in the model. The OLE estimates were based on an MMRM that included stratification factor region, baseline score, and study visit. A model was fit for the ravulizumab and placebo arm of the OLE set separately.

Summary statistics included the number of observations (n), the mean, the standard deviation, and median, minimum, and maximum values. Frequencies and percentages were calculated for categorical variables.

Long-term analyses of change from the OLE baseline in the MG-ADL and QMG total scores through to the end of the OLE period were performed post hoc for both the concomitant IST optimized cohort and the refractory concomitant IST optimized cohort, and were analyzed using an MMRM. Visits with assessments performed on at least 10 patients were used. Missing data were not imputed. A 2-sided 95% CI at each visit for each treatment sequence was calculated.

Of the 175 patients randomized in the RCP, 161 were included in the OLE set (patients who had received 1 dose or more of ravulizumab in the OLE period) and 169 were included in the RTS (all patients who received ≥ 1 dose of ravulizumab either in the RCP or the OLE period; the primary population for assessment of safety during ravulizumab treatment was this patient set). Additional cohort analyses were performed using a subset of the FAS concomitant IST optimized cohort (the OLE set concomitant IST optimized cohort). These patients received 1 dose or more of ravulizumab beginning from week 26 onward and were included in all efficacy analyses during the OLE period.

Results

Patient Disposition

Of the 175 patients randomized in the CHAMPION-MG trial, a total of 161 (92%) patients from the RCP entered the OLE study and received open-label treatment. Of these patients, 123 completed the study, with 38 patients discontinuing. Reasons for discontinuation were withdrawal by patient (9.9%), physician decision (5.0%), death (3.7%), and “other” (5.0%). Of the 6 deaths, 3 patient deaths were assessed as COVID-19–related. All 8 patients designated as “other” were patients in South Korea who transitioned to alternative treatment by the investigator when the study finished. Patient disposition for all study phases was previously illustrated in Figure 4.

Details of patient disposition are provided in Table 24.

Table 24: Patient Disposition During the OLE Period

IST = immunosuppressive therapy; PBO = placebo; RAV = ravulizumab.

Sources: Additional information provided by the sponsor: Response to CDA request on August 6, 2024 (concomitant IST optimized cohort),⁷³ Response to CDA request on August 6, 2024 (refractory concomitant IST optimized cohort),⁷⁴ and Study 306 End of Study Tables and Figures.⁷⁵

Baseline Characteristics

The baseline characteristics are summarized in Table 25. Baseline was defined as the last available assessment value before the first study drug infusion.

Table 25: Demographics and Baseline Characteristics (Open-Label Extension Set)

IST = immunosuppressive therapy; MGFA = Myasthenia Gravis Foundation of America; PBO = placebo; RAV = ravulizumab; SD = standard deviation.

Sources: Clinical Study Report Addendum (End of Study),⁴⁷ and additional information from the sponsor (Response to CDA request on August 15, 2024 (refractory concomitant IST optimized cohort),⁷⁶ and Response to CDA request on August 15, 2024 (concomitant IST optimized cohort). ⁷⁷

Exposure to Study Treatments

The ravulizumab-ravulizumab treatment group had a longer mean exposure to ravulizumab due to the prior 26 weeks of exposure in the RCP. In total, the exposure to ravulizumab at the end of the study was ███ patient-years included █████ patient-years in the placebo-ravulizumab group and █████ patient-years in the ravulizumab-ravulizumab group. The mean duration of ravulizumab treatment by the end of the study (during the RCP and the OLE period) was █████ days and the maximum exposure was ██████ days. From the first dose of ravulizumab through to the end of the study, █████ of patients had 100% compliance. Overall, ████████ patients missed 1 or more scheduled infusion during the ravulizumab treatment period and 19 (11.2%) patients had infusion interruptions.

During the long-term extension period in the concomitant IST optimized cohort, the mean treatment duration (SD) was ████████████ in the ravulizumab-ravulizumab group and ████████████ in the placebo-ravulizumab group. In the refractory concomitant IST optimized cohort, the mean treatment duration (SD) was ████████████ in the ravulizumab-ravulizumab group and ████████████ in the placebo-ravulizumab group. The drug compliance in the 2 cohorts has not been provided.

The patient-years of OLE exposure to ravulizumab in the concomitant IST optimized cohort was ████ in the ravulizumab-ravulizumab group and ████ in the placebo-ravulizumab group. In the refractory concomitant IST optimized cohort, patient-years of OLE exposure to ravulizumab was █████ in the ravulizumab-ravulizumab group and █████ in the placebo-ravulizumab group.

Table 26: Exposure to Treatments (Ravulizumab-Treated Set)

PBO = placebo; RAV = ravulizumab; SD = standard deviation.

Note: Percentages were based on the total number of patients in each group. Study duration equalled the date of study completion or discontinuation minus (the date of informed consent + 1). Treatment duration equalled the date of study completion or discontinuation minus (the first ravulizumab infusion date + 1). Percentage compliance equalled 100% minus (total percentage of time participants were noncompliant with scheduled ravulizumab doses in the ravulizumab treatment period).

Source: Clinical Study Report Addendum (End of Study).⁴⁷

Concomitant Medications and Cointerventions

All patients in the RTS used concomitant MG medications during the ravulizumab treatment period (i.e., concomitant MG medications were those received by the patients on or after the first ravulizumab infusion in the RCP or the OLE period). Overall, concomitant MG medications (other than IVIg) used by more than 25% of patients in the RTS were pyridostigmine bromide (███████ prednisone (███████ and mycophenolate mofetil (███████ The most commonly used IST in the RTS was corticosteroids (███████ In the RTS, █████ of patients used IVIg and ████ of participants used any PLEX.

In the OLE period, physicians had the option to adjust IST therapies. A ███████████████████ of patients had a change in concomitant MG medication (i.e., concomitant MG medications received by the patients on or after the first study drug infusion in the OLE period) during the OLE period. The most common reason for change in corticosteroid therapies was improvement in MG symptoms in █████ of patients. At the end of the OLE period, ███████████ of patients decreased their daily dose of corticosteroid therapy and ███████████ of patients stopped corticosteroid therapy. Similar data were not provided for the cohorts separately.

Efficacy

MG-ADL Total Score

The MG-ADL is an 8-item survey of patient-reported MG symptom severity and the symptoms’ functional impact on the patient, with each response graded from 0 (normal) to 3 (most severe); the total MG-ADL score ranges from 0 to 24.⁴⁹ A reduction in score indicates improvement and the MCID on the MG-ADL scale is estimated to be 2 points based on the change within an individual patient.²⁵

Primary Analysis

In the OLE set, in the ravulizumab-ravulizumab group, the LSM change from RCP baseline in the MG-ADL total score was −4.0 (95% CI, −5.3 to −2.8) at week 164. In the placebo-ravulizumab group, the LSM change from RCP baseline was −3.3 (95% CI, −4.2 to −2.4) at week 28 and −3.6 (95% CI, −4.8 to −2.3) at week 164.

OLE Set Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab concomitant IST optimized cohort, the LSM change from the OLE baseline (OLE period baseline is defined as the last available assessment before administration of the first study drug in the OLE period) in the MG-ADL total score was ██████████████) at week 28 and ███████████████) at week 164.

In the placebo-ravulizumab concomitant IST optimized cohort, the LSM change from the OLE baseline in the MG-ADL total score was ████ (95% ██████████) at week 28 and ████████████████) at week 148.

OLE Set Refractory Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab refractory concomitant IST optimized cohort, the LSM change from the OLE baseline in the MG-ADL total score was ██████████████) at week 28 and ███████████████) at week 164.

In the placebo-ravulizumab refractory concomitant IST optimized cohort, the LSM change from the OLE baseline in the MG-ADL total score was ████████████████) at week 28 and ████████████████) at week 148.

QMG Total Score

The QMG is a 13-item clinician-determined assessment of MG disease status, with each item graded for severity from 0 (none) to 3 (severe); total scores range from 0 to 39.⁵¹ A higher score is indicative of more severe disease. A definite MID for the QMG scale has not been established, although a threshold ranging between 2 points and 3 points depending on disease severity has been suggested in previous studies enrolling patients with MG.^26,27

Primary Analysis

In the OLE set in the ravulizumab-ravulizumab group, the LSM change from the RCP baseline in the QMG total score was −2.9 (95% CI, −4.1 to −1.8) at week 28 and −4.3 (95% CI, −6.0 to −2.7) at week 164.

In the placebo-ravulizumab group, the LSM change from the RCP baseline was −3.0 (95% CI, −4.3 to −1.7) at week 28 and −3.7 (95% CI, −5.5 to −1.9) at week 164.

OLE Set Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab concomitant IST optimized cohort, the LSM change from the OLE baseline in the QMG total score was ██████████████) at week 28 and ███████████████) at week 164.

In the placebo-ravulizumab concomitant IST optimized cohort, the LSM change from the OLE baseline in the QMG total score was ████████████████) at week 28 and ████████████████) at week 148.

OLE Set Refractory Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab refractory concomitant IST optimized cohort, the LSM change from the OLE baseline in the QMG total score was ███ (███████████) at week 28 and ██████████████) at week 148.

In the placebo-ravulizumab refractory concomitant IST optimized cohort, the LSM change from the OLE baseline in the QMG total score was █████████████████ at week 28 and ████████████████) at week 148.

QMG 5-Point or Greater Response

Based on a 5-point or greater improvement in the QMG total score from the RCP baseline, the proportion of clinical responders in the ravulizumab-ravulizumab group was 38.0% (95% CI, 26.8% to 50.3%) at week 28 and 61.5% (95% CI, 31.6% to 86.1%) at week 164.

The proportion of clinical responders in the placebo-ravulizumab group was 31.9% (95% CI, 21.2% to 44.2%) at week 28 and 58.3% (95% CI, 27.7% to 84.8%) at week 164.

Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab group, the proportion of patients with a 5-point or greater improvement in the QMG total score from the RCP baseline was █████████████████) at week 28 and █████████████████) at week 164.

In the placebo-ravulizumab group, the proportion of patients with a 5-point or greater improvement in the QMG total score from the RCP baseline was █████████████████) at week 28 and █████████████████) at week 164.

OLE Set Refractory Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab group, the proportion of patients with a 5-point or greater improvement in the QMG total score from the RCP baseline was █████████████████) at week 28 and █████████████████) at week 164.

In the placebo-ravulizumab group, the proportion of patients with a 5-point or greater improvement in the QMG total score from the RCP baseline was █████████████████) at week 28 and █████████████████) at week 164.

MG-QoL15r Total Score

The MG-QoL15r questionnaire (a revised version of the Myasthenia Gravis Quality of Life 15-item Scale tool) is a 15-item HRQoL instrument for MG, with each item having 3 response options that include 0 (not at all), 1 (somewhat), and 2 (very much); total scores range from 0 to 30, with higher scores indicating a greater extent of dissatisfaction with MG-related dysfunction.⁵² An MID has not been estimated for patients with MG.

In the OLE set, in the ravulizumab-ravulizumab group, the LSM change from the RCP baseline in the MG-QoL15r total score was −3.7 (95% CI, −5.4 to −2.0) at week 26 and −5.2 (95% CI, −7.3 to −3.1) at week 164.

In the placebo-ravulizumab group, the LSM change from the RCP baseline was −5.9 (95% CI, −8.0 to −3.7) at week 30 and −5.4 (95% CI, −7.8 to −3.0) at week 164.

OLE Set Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab group, the LSM change from the OLE baseline in the MG-QoL15r total score was ████████████████) at week 30 and ████████████████) at week 164.

In the placebo-ravulizumab group, the LSM change from the OLE baseline in the MG-QoL15r total score was ████████████████) at week 30 and ████████████████) at week 164.

OLE Set Refractory Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab group, the LSM change from the OLE baseline in the MG-QoL15r total score was ███████████████) at week 30 and ██████████████) at week 132.

In the placebo-ravulizumab group, the LSM change from the OLE baseline in the MG-QoL15r total score was ████████████████) at week 30 and ████████████████) at week 164.

Neuro-QoL Fatigue Score

The Neuro-QoL fatigue subscale is an instrument evaluating the effect of fatigue on the quality of life of patients with neurologic disorders.⁵³ It comprises 19 patient-reported items assessing the frequency of different aspects of fatigue over the previous 7 days, answered using a 5-point scale ranging from 1 (never) to 5 (always); total scores range from 19 to 95. Higher scores indicate greater fatigue and a greater impact of MG on activities. An MID has not been estimated for patients with MG.

In the ravulizumab-ravulizumab group, the LSM change from the RCP baseline in the Neuro-QoL fatigue score was −7.7 (95% CI, −12.2 to −3.2) at week 26 and −10.4 (95% CI, −16.4 to −4.7) at week 164.

In the placebo-ravulizumab group, the LSM change from the RCP baseline was −16.9 (95% CI, −21.4 to −12.5) at week 30 and −14.7 (95% CI, −20.1 to −9.3) at week 164.

OLE Set Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab group, the LSM change from the OLE baseline in the Neuro-QoL fatigue score was ███████████████) at week 30 and ███████████████) at week 164.

In the placebo-ravulizumab group, the LSM change from the OLE baseline was ██████████████████) at week 30 and █████████████████) at week 164.

OLE Set Refractory Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab group, the LSM change from the OLE baseline in the Neuro-QoL fatigue score was █████████████████) at week 30 and ████████████████) at week 132.

In the placebo-ravulizumab group, the LSM change from the OLE baseline was ███████████████████ at week 30 and █████████████████) at week 164.

MG-ADL 3-Point or Greater Response

Based on a 3-point or greater improvement in the MG-ADL total score from the RCP baseline, the proportion of clinical responders in the ravulizumab-ravulizumab group was 64.0% (95% CI, 52.1% to 74.8%) at week 28 and 71.4% (95% CI, 41.9% to 91.6%) at week 164.

The proportion of clinical responders in the placebo-ravulizumab group in the MG-ADL total score from the RCP baseline was 62.5% (95% CI, 51.0% to 73.1%) at week 28 and 75.0% (95% CI, 42.8% to 94.5%) at week 164.

OLE Set Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab group, the proportion of patients with a 3-point or greater improvement in the MG-ADL total score from the RCP baseline was ████ █████████████) at week 28 and ██████████████████ at week 164.

In the placebo-ravulizumab group, the proportion of patients with a 3-point or greater improvement in the MG-ADL total score from the RCP baseline was █████████████████) at week 28 and ██████████████████ at week 164.

OLE Set Refractory Concomitant IST Optimized Cohort Analysis

In the ravulizumab-ravulizumab group, the proportion of patients with a 3-point or greater improvement in the MG-ADL total score from the RCP baseline was █████████████████) at week 28 and ████████████████ at week 164.

In the placebo-ravulizumab group, the proportion of patients with a 3-point or greater improvement in the MG-ADL total score from the RCP baseline was █████████████████) at week 28 and ██████████████████ at week 164.

Harms

Results for the safety end points in this section are presented for the 169 patients in the RTS — all 86 patients in the ravulizumab-ravulizumab group and 83 patients in the placebo-ravulizumab group. A summary of harms is reported from the first ravulizumab treatment exposure during the RCP through to the end of the OLE period in the OLE set and in both cohorts. In all cases of AEs, SAEs, and death, the results for the 2 cohorts were similar to the OLE set. WDAEs were rare in the full population and only 1 case was reported in the ravulizumab-ravulizumab concomitant IST optimized cohort.

Adverse Events

During ravulizumab treatment, 96.4% of patients experienced at least 1 AE. The most commonly reported AEs (≥ 10% of total patients) in the RTS were COVID-19 (36.1%), headache (23.1%), diarrhea (17.2%), arthralgia (13.6%), back pain (13%), nausea (13%), urinary tract infection (12.4%), nasopharyngitis (11.8%), fatigue (10.7%), and dizziness (10.1%).

Serious Adverse Events

SAEs reported by 1 patient or more included COVID-19 (6 patients), MG (5 patients), COVID-19 pneumonia (4 patients); SAEs reported by 3 patients each included cellulitis and pneumonia; and SAEs reported by 2 patients each included erysipelas, urinary tract infection, spinal compression fracture, intervertebral disc protrusion, transient ischemic attack, congestive cardiac failure, pyrexia, dyspnea, dysphagia, dehydration, and nephrolithiasis.

There were 8 patient deaths throughout the entirety of the study period. There were 2 patient deaths in the RCP (due to COVID-19 pneumonia and cerebral hemorrhage). Six patients died in the OLE period due to the following reasons: COVID-19 (3 patients), toxicity due to various drugs (1 patient), dehydration (1 patient), and an unknown reason (1 patient).

Withdrawals Due to Adverse Events

Four patients had AEs that led to WDAEs. Two AEs occurred during the RCP (infected skin ulcer and COVID-19 pneumonia) and 2 AEs occurred during the OLE period (COVID-19 and MG). All 4 AEs occurred in the ravulizumab-ravulizumab treatment group. No patients in the placebo-ravulizumab group discontinued the study drug. The total number of patients with ravulizumab infusion interruption due to an AE was 8 (4.7%) patients and 74 (43.8%) patients reported potential infusion reactions during ravulizumab treatment.

Adverse Events of Special Interest

There were no meningococcal infections reported during the study, although there was 1 event of meningitis of unknown etiology reported in the OLE period in the placebo-ravulizumab group (Table 27 and Table 28).

Table 27: Summary of Harms Results From Long-Term Extension Phase of CHAMPION-MG Study

AE = adverse event; IST = immunosuppressive therapy; NR = not reported; PBO = placebo; RAV = ravulizumab; SAE = serious adverse event; SD = standard deviation.

Sources: Additional information provided by sponsor: Response to CDA request on August 6, 2024 (concomitant IST optimized cohort) ⁷³, and Response to CDA request on August 6, 2024 (refractory concomitant IST optimized cohort).⁷⁴

Table 28: All Treatment-Emergent Adverse Events and Serious Adverse Events During Ravulizumab Treatment Period in Ravulizumab-Treated Set

AE = adverse event; SAE = serious adverse event.

Source: The sponsor’s Summary of Clinical Evidence.⁷²

Critical Appraisal

Internal Validity

The lack of a control group precludes causal statements about benefit and harm compared with any comparator. The open-label nature of the study may increase the risk of bias in determining the magnitude of the safety outcomes and the efficacy end points that include more subjective assessments, because the lack of blinding may affect patients’ expectations of treatment. The direction and magnitude of this potential bias remains unclear. Patients of the OLE were patients who did not drop out of the placebo-controlled study (92%), which puts the results at some risk of selection bias, likely to be favouring ravulizumab. Of the 175 patients randomized in the CHAMPION-MG trial, 123 patients completed the OLE. There is therefore a risk of bias due to missing outcome data in both the placebo-ravulizumab and the ravulizumab-ravulizumab arms, with the impact on the direction of treatment effect over time not clear. The limitations of the post hoc analyses of the subgroups mirror those discussed in the systematic review section; however, these concerns have been minimized as the findings for these groups were generally consistent with the overall population.

External Validity

The appraisal of the external validity of the RCP, presented in the Critical Appraisal subsection regarding the CHAMPION-MG study in the Clinical Evidence section, is also applicable to the OLE period.

Indirect Evidence

Content in this section has been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

Objectives for the Summary of Indirect Evidence

The objective of this section is to summarize and appraise the ITCs that provide evidence for the relative efficacy and safety of ravulizumab versus relevant comparators for the treatment of adult patients with anti–AChR antibody–positive gMG in the concomitant IST optimized cohort as well as the refractory concomitant IST optimized cohort in an NMA analysis, and for the treatment of patients with gMG, although not for a specific subgroup, in an MAIC analysis.

Description of Indirect Comparison

Study Selection Methods

To inform the NMA, a systematic literature review was performed in February 2022 and updated in January of 2024 to identify literature published from 2000 to 2024. Published literature was identified by searching the following databases: MEDLINE All (1946–), Embase (1974–), and Evidence-Based Medicine Reviews. The search strategies used medical subject headings and keywords related to the study population of interest. Study design filters for clinical trials were also used. Comparators and outcomes were not prespecified in the search strategy. Instead, these characteristics were screened for in the abstract and full-text review steps to ensure that all potentially relevant studies were identified.

Study selection was conducted by 2 independent reviewers. All abstracts identified by the search strategy were reviewed and compared against the predetermined Population, Intervention, Comparison, Outcomes, and Study (PICOS) tool. If any discrepancies occurred between the studies selected for inclusion by the 2 reviewers, a third reviewer provided arbitration. Following abstract screening, the full texts of included citations were accessed. Similar to the abstract screening process, full-text review was conducted by 2 reviewers independently and decisions were made to include or exclude full-text articles depending on whether the PICOS criteria were met. It is important to note that studies describing patient populations comprising both anti–AChR antibody–positive MG and anti-MuSK antibody–positive MG were considered for inclusion. In general, 80% to 85% of patients with gMG have AChR, nearly 15% of patients are anti-AChR antibody–negative, and approximately 1% to 10% of patients have MuSK or LRP4.³ Further details regarding PICOS, data extraction, and the assessment of study quality are described in Table 29.

Table 29: Study Selection Criteria and Methods for Network Meta-Analyses

AChR = acetylcholine receptor; AE = adverse event; EBM = Evidence-Based Medicine; gMG = generalized myasthenia gravis; IVIg = IV immunoglobulin; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA-PIS = Myasthenia Gravis Foundation of America Post-intervention Status; MMS = minimal manifestation status; MSE = minimal symptom expression; PLEX = plasma exchange or plasmapheresis; QMG = Quantitative Myasthenia Gravis; RCT = randomized controlled trial; SAE = serious adverse event.

Sources: Sponsor-submitted network meta-analysis report.²⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

The sponsor also submitted an MAIC to assess the relative efficacy of ravulizumab compared to efgartigimod alfa for the management of patients with gMG and anti–AChR antibody–positive. In this MAIC, change in the MG-ADL total score at select follow-up time points (week 4, week 8, week 10, and week 26) was the only outcome that was examined. There was no description of study selection, data extraction, or assessment on the risk of bias of the included studies.

ITC Analysis Methods

Network Meta-Analyses

An NMA was performed for the concomitant IST optimized population. Given the limited data available for the refractory concomitant IST optimized population, comprising only 2 studies and 2 treatments, an ITC was conducted using placebo as the common comparator. To integrate the convergence diagnosis and output analysis into the economic model, the ITC was performed with a Bayesian approach. Supplementary analyses were performed using the frequentist method (i.e., Bucher).

Description of Model

The NMA was performed using a Bayesian framework. The results were calculated with Markov Chain Monte Carlo (MCMC) simulations, which allow reproducing the model several times until convergence.⁶³ Using the software gemtc (version 1.0 to 2) package of the R, a fixed-effect (FE) model and a random-effect (RE) model with a normal likelihood and an identity link function were performed for the continuous outcomes, and with a binomial likelihood and a log link function were performed for the binary outcomes.⁶⁴

The FE model assumes homogeneity between studies while the RE approach assumes that the studies were drawn from different populations, which could impact the estimated treatment effect. With the RE approach, between-study heterogeneity (tau-squared) is considered.⁶⁵ Due to the limited number of studies (n = 7) and in accordance with the Cochrane Scientific Committee (2018) — recommendations for meta-analyses that included fewer than 10 studies — informative priors for the tau-squared were used in the RE model.⁶⁶ The choice between the FE model and the RE model was based on the lowest deviance information criterion (DIC) score.⁶⁷ However, with similar DICs (i.e., a difference of less than 2), the FE model was preferred and selected.⁶⁸ As a result, the RE model was provided as supplementary analysis.

Vague priors (noninformative priors [uniform distribution]) were assigned to the parameters mu (mean effect) and d (treatment effect). The RE model was fitted with informative priors for the heterogeneity parameter (i.e., log-normal ([mean = −3.02, standard deviation = 1.85]² for semiobjective outcomes). The sponsor noted that due to the limited number of studies (n = 7) and in accordance with the Cochrane Scientific Committee (2018) recommendations for meta-analyses that included fewer than 10 studies, informative priors for the tau-squared were used in the RE model.⁶⁶ A sensitivity analysis was performed using vague priors (i.e., noninformative).

MCMC sampling was done with 4 chains, with an initial number of iterations discarded as the “burn-in” period. The burn-in was initially 20,000 samples, but this was increased if there was evidence that the MCMC did not converge. Once convergence was achieved, another minimum of 20,000 samples was generated from the posterior distribution to estimate treatment effects and the 95% Crls. The Potential Scale Reduction Factor was verified to be under a threshold of 1.05 and the convergence was assessed with the Brooks-Gelman-Rubin method using the same package. Specifically, Gelman and Rubin (1992) proposed a general approach to monitor the convergence of MCMC outputs in which 2 or more parallel chains are run with starting values that are overdispersed relative to the posterior distribution. The convergence is assessed by comparing the estimated between-chain and within-chain variances for each model parameter. Large differences between these variances indicate nonconvergence.

The model provided the mean difference for the efficacy outcomes or the risk ratio for the safety outcome for each comparison of treatments. Results of pairwise treatment comparisons were considered statistically significant when the 95% CrI did not cross the line of equality (mean difference = 0 or risk ratio = 1).

Heterogeneity Assessment

Because of the multiple comparisons involved in the NMA, which may generate inconsistency or incoherence in the model, these 3 assumptions were made: similarity, homogeneity, and consistency.⁶³

Heterogeneity between baseline characteristics and between study characteristics among trials were evaluated by conducting subgroups analyses, when relevant and possible. The statistical assessment of heterogeneity was based on the I² statistic, derived from a direct head-to-head meta-analysis of the treatment comparisons in each network that were evaluated in more than 1 study.⁶⁹ All comparisons with I² higher than 50% (indicating substantial or considerable heterogeneity) were investigated. The studies included in these treatment nodes were examined to determine the cause of the observed heterogeneity. If needed, sensitivity analyses without the studies at the origin of the observed heterogeneity were performed.

To verify the consistency of the direct and indirect comparisons, a node-splitting approach was considered. With this approach, the direct evidence for a specific treatment comparison is contrasted with the indirect evidence coming from the rest of the network. When direct and indirect evidence differed substantially, that indicated that inconsistency was present. However, due to the limited number of studies, the node-split analysis was not feasible for any outcomes.

Since outcomes were not prespecified, no preplanned analyses were excluded. An NMA was conducted at the end of the study (the longest follow-up time point) for the 2 efficacy outcomes (i.e., the mean change from baseline in the MG-ADL and QMG total scores). Additionally, a sensitivity analysis was performed at 8 weeks plus or minus 2 weeks (range, 6 weeks to 10 weeks) to ensure a maximum inclusion of studies at a similar time point. An NMA for WDAEs was also completed.

Table 30: Analysis Methods Used in Sponsor-Submitted NMA

AChR = acetylcholine receptor; CFB = change from baseline; FAS = full analysis set; FE = fixed-effect; gMG = generalize myasthenia gravis; IST = immunosuppressive therapy; MG-ADL = Myasthenia Gravis Activities of Daily Living; NMA = network meta-analysis; QMG = Quantitative Myasthenia Gravis; RE = random-effect; WDAE = withdrawal due to adverse event.

Sources: Sponsor-submitted network meta-analysis report.²⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Matching-Adjusted Indirect Comparison

The MAIC was conducted following the recommendations of the National Institute for Health and Care Excellence Decision Support Unit Technical Support Document⁶⁷ on population-adjusted indirect comparisons. Baseline characteristics of the study populations were compared before undertaking the MAIC. Imbalances across those baseline characteristics were identified, as were prognostic factors or treatment-effect modifiers. The subset of baseline characteristics available in both studies that were identified as potentially being prognostic or a treatment-effect modifier were selected for adjustment.

Mean changes from baseline in the MG-ADL score were compared at the following time points: week 4, week 10, and time of best response (week 26 for ravulizumab versus week 4 for efgartigimod alfa).

The MAIC was based on 2 data sources:

• Ravulizumab data were drawn from individual patient data (IPD) from the CHAMPION-MG trial (n = 86 for the ravulizumab group; n = 89 for the placebo group). These were adults diagnosed with MG, anti-AChR antibody–positive, and an MG-ADL score at baseline of 6 or more. They had no requirement for prior treatment failure but stable standard of care was allowed.

• Efgartigimod alfa data were drawn from aggregate data reported for the anti-AChR antibody–positive subgroup of Study ADAPT (n = 65 for the efgartigimod alfa group; n = 64 for the placebo group). These were adults diagnosed with MG and an MG-ADL score of 5 or more. Patients were also required to be on a stable dose of at least 1 treatment for gMG (e.g., AChE inhibitors, corticosteroids, NSISTs) before screening and throughout the trial

The analyses were anchored on the placebo groups of Study ADAPT and the CHAMPION-MG trial. The MAIC was conducted by matching the IPD from the CHAMPION-MG trial to baseline summary characteristics from Study ADAPT; this was performed at the arm-level (active arm to active arm, placebo arm to placebo arm) to account for differences between the treatment groups of Study ADAPT in characteristics such as mean age and mean time since diagnosis. Adjustment for differences in selected baseline characteristics was achieved by reweighting the influence of patients in each CHAMPION-MG trial arm’s IPD such that the weighted average of the characteristics matched the profile of the cohort reported by the respective arm in Study ADAPT.

Results of ITCs

Summary of Included Studies

Network Meta-Analyses

In total, 7 studies were included in the NMA, as summarized in Table 31.

Table 31: Studies Included in NMA Analyses