CADTH Reimbursement Review

Efgartigimod Alfa (Vyvgart)

Sponsor: argenx Canada Inc.

Therapeutic area: Generalized myasthenia gravis

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

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

Table 1: Background Information of Application Submitted for Review

AChEI = acetylcholinesterase inhibitor; AChR = acetylcholine receptor; gMG = generalized myasthenia gravis; NOC = Notice of Compliance.

Source: The sponsor’s submission.¹

Introduction

Myasthenia gravis (MG) is a rare, chronic, neuromuscular autoimmune disease mediated by pathogenic autoantibodies that target structural components of the neuromuscular junction, impairing neuromuscular transmission and leading to muscle weakness and fatigue.^2-4 Many patients initially present with symptoms affecting only the eye muscles (i.e., ocular MG). Approximately 85% of patients go on to develop generalized myasthenia gravis (gMG), with weakness affecting the neck, trunk, limbs, and bulbar and respiratory muscles. Patients with gMG experience symptoms that negatively impact health-related quality of life (HRQoL).⁵ The disease has a fluctuating natural history; MG exacerbation (an increase in symptoms in patients who were previously asymptomatic or minimally symptomatic, defined as a ≥ 3-point worsening in Quantitative Myasthenia Gravis [QMG] score versus baseline) and myasthenic crisis (muscle weakness causing life-threatening difficulties with breathing and swallowing and requiring ventilator support) can occur gradually or without warning.⁶

Approximately 85% of patients with gMG are acetylcholine receptor (AChR) antibody (Ab) seropositive (Ab+); as many as 15% of patients with gMG are seronegative for AChR-Ab (Ab–).^7,8 An estimated 1% to 10% of patients do not have AChR antibodies but do have autoantibodies against muscle-specific kinase (MuSK) antibody seropositive (MuSK-Ab+) or autoantibodies against lipoprotein receptor-related protein-4 (LRP4) Ab seropositive (LRP4-Ab+), which also lead to a decrease in AChRs.³ The Myasthenia Gravis Foundation of America (MGFA) classification system is a tool used to categorize gMG based on clinical features and/or disease severity.⁹ The classification ranges from Class I (i.e., ocular weakness only), through Class II, Class III and Class IV (representing patients with gMG with mild, moderate, and severe muscle weakness, respectively⁹), to Class V (defined by intubation, with or without mechanical ventilation, except when employed during routine postoperative management, or myasthenic crisis). The incidence of MG in Canada is estimated at 23 cases per 1 million person-years, with a prevalence of 32 cases per 100,000 adults (0.032%) in Canada.^10-12 Thus, there are approximately 8,121 patients with MG across the CADTH-participating drug programs (0.032% × 25,376,703 adult patients in CADTH-participating drug programs in 2023). Approximately 85% of adults with MG are anticipated to progress to gMG, which corresponds to approximately 6,903 adult patients with gMG in Canada.

The clinical experts that CADTH consulted for this review indicated that the goal of treatment in patients with gMG is to reduce disease symptoms and adverse effects of MG therapy and to allow the patient to function as they would normally with good HRQoL. Other goals of treatment include avoiding MG exacerbations and myasthenic crisis, minimizing hospitalizations and intensive care unit (ICU) admissions, and reducing the numbers and doses of therapies (e.g., especially corticosteroid use) required for symptom control. The available main therapies for gMG include acetylcholinesterase inhibitors (AChEIs), corticosteroids, nonsteroidal immunosuppressive therapies (NSISTs), rituximab, intravenous immunoglobulin (IVIg), plasma exchange (PE) or plasmapheresis (PP), and terminal complement inhibitors (i.e., ravulizumab and eculizumab). According to the clinical experts consulted by CADTH for this review, the first-line standard of care (i.e., the conventional therapy) for MG are AChEIs, corticosteroids, and NSISTs (e.g., azathioprine, mycophenolate mofetil, cyclophosphamide, cyclosporine, tacrolimus, methotrexate). Mild to moderate gMG (MGFA Class II or IIIa) is initially treated symptomatically with AChEIs (usually pyridostigmine);¹³ the onset of benefit occurs in hours to days. If this provides insufficient symptom relief, immunosuppressive therapy (IST) with corticosteroids (usually prednisone) is administered;¹⁴ maximal responses typically occur 2 to 6 months later, after which slow tapering of corticosteroids is begun. In patients who do not respond to corticosteroids, who have significant comorbidities such that long-term corticosteroid treatment is contraindicated, or whose doses of corticosteroids cannot be tapered, treatment with NSISTs¹⁵ and/or immunomodulatory drugs, including rituximab, may be initiated.¹⁶ The clinical experts stated that the onset of benefit from NSISTs occurs in months to years (approximately 9 to 18 months for azathioprine and mycophenolate mofetil). While rituximab has not been approved as a gMG treatment by Health Canada, it is considered a treatment option in Canada for patients with refractory gMG who are AChR-Ab+, according to surveys conducted with clinical experts.¹⁶ According to the clinical experts, in patients with moderate to severe gMG, especially those who have respiratory or bulbar weakness, IVIg, PE, or PP may be administered^17,18 in addition to rituximab, either at the time of IST initiation or to treat MG exacerbation or myasthenic crisis.

As MG symptoms improve, doses of AChEIs, corticosteroids, and NSISTs are reduced and the frequency of IVIg, PE, or PP are reduced until the minimal maintenance therapy required for remission is identified. Patients with refractory gMG who are AChR-Ab+ may be candidates for the complement inhibitor eculizumab. While eculizumab received a recommendation for reimbursement with conditions in 2020,¹⁹ price negotiations concluded without an agreement in December 2022.²⁰ A survey of 7 expert clinicians from across 6 provinces indicated that ravulizumab would be another option, if it were to be approved and funded. Ravulizumab would be a treatment option for patients who have an inadequate response to conventional therapy.¹⁸ In April 2023, CADTH issued a draft “do not reimburse” recommendation for ravulizumab in this indication.²¹ The clinical experts consulted by CADTH for this review emphasized that most patients with gMG (more than 80%) respond well to currently available treatments; although these cannot cure the disease, excellent symptom control is achieved in most patients and prognosis is generally good in terms of muscle strength and function as well as HRQoL. Despite treatment with conventional therapy (AChEIs, corticosteroids, and/or NSISTs), many patients continue to experience disease burden and symptoms that impact their HRQoL^22-27 and experience treatment-related side effects that may be severe.

Efgartigimod alfa is a first-in-class human immunoglobulin G1 (IgG1) antibody fragment crystallizable (Fc)-fragment that blocks the neonatal Fc receptor (FcRn).^28,29 Efgartigimod alfa is supplied as a 20 mg/mL solution, with 10 mg/kg administered as an IV infusion over 1 hour once weekly for 4 doses (i.e., weeks 0 to 3). In patients weighing 120 kg or more, the recommended dose of efgartigimod alfa is 1,200 mg (3 vials) per infusion. Efgartigimod alfa reduces the levels of pathogenic IgG autoantibodies.²⁹ Efgartigimod alfa received a Health Canada Notice of Compliance for the treatment of adult patients with gMG who are AChR-Ab+ on September 19, 2023. The sponsor’s reimbursement request is that efgartigimod alfa be reimbursed as an add-on therapy for adult patients with AChR-Ab+ gMG whose symptoms persist despite adequate treatment with AChEIs, corticosteroids, and/or NSISTs, which is a subgroup of the approved Health Canada indication.¹

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient and clinician groups who responded to CADTH’s call for input and from clinical expert(s) consulted by CADTH for the purpose of this review.

Patient Input

CADTH received 1 patient group submission from the Muscular Dystrophy Canada (MDC). MDC identified and contacted adults living with MG and invited them to participate in a survey and semistructured interviews. Respondents indicated that MG has a significant impact on their productivity, levels of fatigue, energy levels, quality of sleep, respiratory health, mobility, strength, independence, relationships and social participation, eyes and vision, speech, and swallowing. They also explained that the impact of MG extends beyond physical symptoms and affects their mental health, quality of life, and the well-being of their families.

Some of the respondents indicated they feel that their lungs are weaker, they had to go on a ventilator in ICU, choking on food or saliva interferes with breathing, they cannot even walk inside their own house, they always keep a walker or cane nearby because they never knew when the MG would flare up, they cannot sleep at night because of aches, and they are unable to drive. Some indicated that they had slurred speech, frequently go cross-eyed, their double vision interferes with reading, and they have experienced multiple acute hospitalizations.

When asked about how their MG is being managed with available treatments, 3 main themes emerged from the analysis: negative experiences with steroids (e.g., adverse effects; costs), the slow onset of medication effects, and a feeling of trial and error with medications. Regarding improved outcomes, the patient group identified 3 aspects of MG that they wanted better controlled: decreased intensity of exacerbations and side effects, maintenance of independence, and fewer serious hospital admissions. The method, duration, and frequency, and convenience of treatment, as well as the cost are very important to the patients and caregivers. They preferred less travel, fewer hospital visits, and less invasive methods of treatment. HRQoL was noted as a key priority over convenience of a drug.

Respondents stated that they would be willing to deal with side effects of medications if these aspects of MG were better controlled. They also stated that although their current medications decreased the number of exacerbations, they do not have an impact on overall quality of life. They expected new treatments to help them become independent, stop the myasthenic crises, address the respiratory and general weakness, be easier to swallow (for pills), reduce pain, not lead to diabetes, be target treatment for MG instead of general immunosuppression, be less expensive, work quickly, be a single daily dose in the morning.

One respondent had received Vyvgart as a participant in a clinical trial and explained that this medication replaced their need for IVIg, the effects appeared quicker compared with other therapies, the infusion time was less than expected, treatment was received less frequently compared with other therapies, and they experienced fewer side effects compared with other therapies. This patient respondents highlighted that while diarrhea was a problem and not unique to Vyvgart, it was manageable after the first cycle of treatment.

All the respondents had experienced diagnostic blood testing, and many had undergone single-fibre electromyography to confirm diagnosis. A total of 80% of the respondents reported difficulty receiving a diagnosis. Based on early findings of the MG journey mapping project, MDC reported 7 years from time of first bothersome symptom to diagnosis, with a range up to 23 years. According to MDC, the majority of respondents found the process of testing and diagnosis cost-effective but lengthy, with many missed opportunities, delayed diagnosis, misdiagnosis (such as stroke and Bell palsy), and costs incurred. Those who were diagnosed during a crisis or hospitalization reported a smooth diagnosis (25%).

Clinician Input

Input From Clinical Experts Consulted by CADTH

The following input was provided by 2 clinical specialists with expertise in the diagnosis and management of gMG.

The clinical experts indicated that approximately 90% of patients respond to current treatments; however, response is often partial, meaning that there are still symptoms that affect QoL and function. According to the clinical expert, besides prednisone and rescue treatments, ISTs take very long to act (e.g., azathioprine takes at least a year). The clinical experts explained that this means that patients may be exposed to higher doses of steroids for longer periods, and experience persistent symptoms for longer, before even knowing whether this medication will be effective or not. Current treatments are nontargeted, causing overall more diffuse immunosuppression, and there is an increased risk of cancer with long-term use. Also, the risk of all steroid-related adverse events (AEs) increases with prolonged doses. The clinical experts also indicated that the fraction of patients with refractory gMG varies according to definition, although a reasonable estimate would be 10% to 20% of the total population. According to the clinical experts consulted by CADTH for this review, patients with gMG usually start with pyridostigmine, but most patients will need disease-modifying treatment with immunosuppressants, most commonly prednisone. Depending on severity, age, and comorbidities, an NSIST (e.g., azathioprine, mycophenolate mofetil, and tacrolimus) may be started early after diagnosis, or later (for example, unable to reduce dose of steroids). Some patients with severe disease at onset (e.g., crisis or severe symptoms) may receive a rescue treatment such as IVIg or PE early to have fast improvement while immunosuppression begins. Few patients receive chronic IVIg or PE, and some of these patients are dependent on these treatments. According to the clinical experts, the treatment goals is to achieve minimal symptoms or remission, with the fewest AEs from treatments. Patients express a need to improve the ability to perform their daily life activities, reduce fatigue, and improve their ability to care for their family and work or home obligations.

The clinical experts stated that efgartigimod alfa has a specific mechanism of action related to the pathophysiology of MG (reduction of IgG levels, including AChR antibodies). Efgartigimod alfa reduces levels of IgG, but it does not affect the process of producing AChR antibodies. The clinical experts indicated that they do not foresee efgartigimod alfa being used as first-line treatment; rather, efgartigimod alfa would be suitable for individuals without adequate response to available treatments, those who are dependent on IVIg or PE, or those with very severe disease to bridge the gap of delayed action of standard immunosuppressive therapies. The clinical experts noted that patients should have received standard conventional treatments first because standard conventional treatments will be satisfactory in a large number of patients. The clinical experts also stated that most treatments for patients with gMG are given off label because of the lack of randomized controlled trial (RCT) evidence (e.g., prednisone, azathioprine, mycophenolate mofetil, tacrolimus, and rituximab). The clinical experts noted that IVIg or PE used as a rescue treatment is different from IVIg or PE used chronically (i.e., maintenance therapies, once a month); very few patients use IVIg or PE chronically. According to the clinical experts, eculizumab or ravulizumab are used too rarely at present to include them as comparators. One clinical expert indicated that, if cost was not an issue, one could argue that efgartigimod alfa could be tried as an initial therapy in patients for whom pyridostigmine alone was ineffective; however, they noted that the cost is likely to be a major barrier.

The clinical experts indicated that efgartigimod alfa will provide a new treatment for patients with gMG who are AChR-Ab+ and probably for patients who are MuSK-Ab+. However, whether patients with gMG who are AChR-Ab– would respond to efgartigimod alfa is unknown because few patients who are AChR-Ab– were included in the ADAPT trial.

The clinical experts indicated that, in most clinics (not academic), patients are not given standardized assessments. In academic settings, clinicians use validated measures. The clinical experts noted that the Myasthenia Gravis Activities of Daily Living (MG-ADL) scale used in trials is easy to use and can be easily incorporated into routine clinical practice and all settings (community, hospital, and academic). The clinical experts recommended using the MG-ADL for patients with active treatment at all visits, to be able to follow the clinical course. The clinical experts stated that an improvement (reduction) of 2 points is considered significant. Apart from symptom scores, overall function, and ability to return to work are also assessed. The clinical experts expressed that the ability to reduce or stop chronic use of corticosteroids, IVIg, or PE is an important outcome when considering the use of efgartigimod alfa. They explained that some patients are dependent on chronic IVIg or PE, weaning off these treatments is important. The clinical experts also highlighted that reduction or avoidance of hospitalization due to MG is an important outcome.

Frequency of assessments depends on symptoms and patient stability. Patients whose symptoms are well-controlled are typically seen by clinicians every 6 months, but can be seen more frequently (e.g., every 2 to 3 months) in case of worsening health, new medications, and so on. For efgartigimod alfa, most responses were fast, but a small proportion of responders lagged, and response could be seen at the second cycle. Therefore, the clinical experts suggested 2 to 3 months may be needed to assess response to cycle 1 and/or to assess for need of another cycle. An assessment at 6 months may be needed to determine which patients do not respond to treatment. For those who do respond, subsequent assessments could then be done every 3 to 6 months to determine if new cycles are needed.

The clinical experts indicated that efgartigimod alfa should be discontinued if a patient has no response to treatment (i.e., no improvement in symptoms and function), experiences a severe AE (e.g., severe infusion reaction), needs rescue treatment (IVIg or PE, or increased dose of steroids), or is unable to reduce chronic use of corticosteroids, IVIg, or PE. The clinical experts indicated that patients should be under the care of a neurologist with experience in diagnosing and treating MG, usually a neuromuscular specialist. The infusion itself can be arranged at infusion clinics.

Clinician Group Input

CADTH received 1 clinician group submission from the Neuromuscular Disease Network for Canada (NMD4C).

NMD4C stated that conventional treatment options for gMG have been based on symptomatic therapy, short-term rescue immunotherapy and long-term IST. Moreover, nonspecific immunosuppressants have been only partially effective and many patients do not attain stable remission, with 10% to 20% of patients not responding or intolerant to these drugs.

According to NMD4C, some of the unmet needs of the standard treatments are side effects, lack of effectiveness for all patients, long periods of treatment, and transient effectiveness. Another unmet need in this field is the lack of therapeutic options for seronegative patients.

NMD4C noted that patients who are AChR-Ab+ will most likely respond to the drug under review. Patients with MuSK antibodies and those who are double seronegative (i.e., AChR-Ab– and MuSK-Ab–) might respond. Patients who worsen quickly, particularly those who experience a myasthenic MG crisis are most in need of an intervention that works quickly, but patients who have symptoms restricted to only the ocular muscles are unlikely to require such rapid intervention with the drug under review. To identify the patients best suited for treatment with the drug under review, clinician examination and judgment supplemented by assessment of MG activities of daily living using scales that reflect severity of disease, such as the QMG score, Myasthenia Gravis Impairment Index (MGII), and the single simple question (SSQ). If scales are not available, then antibody testing needs to be done; but according to the clinician group, this can be delayed. It is not clear at this point how to predict which patients are more likely to respond to treatment, except by determining the presence of AChR antibodies.

NMD4C indicates that diagnosis of double seronegative patients is an issue since cluster antibodies to both acetylcholine and MuSK may be present but need to be tested specifically, which can take weeks.

The clinician group indicated that to determine patients’ response to therapy, scales such as the MG-ADL, QMG, MGII, and SSQ at 2 and 4 weeks are required and after that the assessment should be based on the patient’s status. The clinician group noted that a clinically meaningful response to treatment used in the clinical trials is 2 or more points on the ADL and 3 or more points on the QMG. For the SSQ, the clinician group suggested that levels above 72% indicate general satisfaction. In case of lack of response, discontinuation of treatment should be considered.

The clinician group mentioned that the usual Ig treatment for MG can be effective, but it places a significant burden on the Canadian health care system and supplies can be at risk in situations such as during a pandemic. They think the drug under review is likely to replace Ig therapies.

In summary, the clinician group’s input is aligned with the input provided by the clinical experts consulted by CADTH.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s Reimbursement Review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 4.

Clinical Evidence

Systematic Review

Description of Studies

One phase III, double-blind, placebo-controlled RCT (ADAPT, N = 167)³⁰ is included in the systematic review.³¹ The objective of the ADAPT trial was to evaluate the efficacy and safety of efgartigimod alfa added on to conventional therapy versus placebo added to conventional therapy in adult patients with gMG whose symptoms persisted despite a stable dose of standard-of-care treatment (concomitant gMG treatment) with AChEIs, corticosteroids, and/or NSISTs. All patients had MGFA Class II to IV gMG and MG-ADL total score of 5 or more. The mean age was 44.7 years in efgartigimod alfa group and 49.2 years in placebo group, and most patients were white (83.1% in efgartigimod alfa group and 87.5% in placebo group). In the patient population who were AChR-Ab+, 129 (100%) patients had received 1 prior therapy, 124 (96.1%) had received 2 prior therapies, and 102 (79.1%) had received 3 or more prior therapies. The majority had previously received 2 or more (93.4%) or 3 or more (77.2%) different classes of conventional therapy medication (any combination of AChEIs, corticosteroids, and/or NSISTs at the physician’s discretion). In the patient population who were AChR-Ab+, patients who received 3 classes of prior therapy (steroid plus NSIST plus AChEI), 41 (63%) were in the efgartigimod alfa group and 37 (58%) were in the placebo group. Eleven (16.9%) patients in the efgartigimod alfa group and 23 (36%) in the placebo group had received any 2 of the 3 prior therapies (i.e., steroid, NSIST, and/or AChEI). In addition, of the patients who were AChR-Ab+, 63% had prior gMG treatment failures (also known as refractory gMG)³² and 37% had did not have treatment failure but responded inadequately to the existing standard-of-care gMG therapy.

Patients were randomized 1:1 to receive efgartigimod alfa or a matching placebo in cycle 1 (i.e., for first 8 weeks) followed by an individualized treat-as-needed regimen based on the patients’ MG-ADL response. All patients received a stable concomitant treatment during the trial. The primary outcome of the study was the percentage of patients with AChR-Ab+ who were “MG-ADL responders” (defined in the study as a patient with a ≥ 2-point improvement [reduction] in MG-ADL score) in the first treatment cycle. Key secondary outcomes included percentage of MG-ADL responders in cycle 1 in the overall population (i.e., AChR-Ab+ and AChR-Ab–), percentage of time patients with AChR-Ab+ showed a clinically meaningful improvement (CMI) in MG-ADL score (≥ 2-point reduction) up to day 126, time from week 4 to qualify for re-treatment in the AChR-Ab+ population; percentage of early MG-ADL responders in cycle 1 in the AChR-Ab+ population (i.e., MG-ADL ≥ 2 points occurred by week 2), and change from cycle baseline in MG-ADL total score in cycle 1 and cycle 2. Changes from cycle baseline in HRQoL (MG-QoL15r and EQ-5D visual analogue scale [VAS]) in cycle 1 and cycle 2 were assessed as tertiary or exploratory outcomes. Post hoc analysis was performed for gMG hospitalization, gMG exacerbation, and gMG crisis. It should also be noted that, although the ADAPT trial duration was designed for 26 weeks, the primary, key secondary, and the HRQoL outcomes at the end of the study (i.e., week 26) were not assessed. Instead, the outcomes were assessed at the end of cycle 1 and cycle 2.

Efficacy Results

Patients Who Are AChR-Ab+

Myasthenia Gravis Activities of Daily Living

MG-ADL responders during cycle 1 and cycle 2: MG-ADL responders during cycle 1 in patients who were AChR-Ab+ was the primary outcomes in the ADAPT trial. There were 38% (95% confidence interval [CI], 22% to 56%) more patients in the efgartigimod alfa group (those with AChR-Ab+ randomized to receive efgartigimod alfa) than in the placebo group achieved an improvement in MG-ADL of 2 points or more during cycle 1. The between-group difference was considered clinically meaningful by the clinical experts consulted by CADTH.

Various post hoc subgroup analyses were conducted for MG-ADL responders during cycle 1. Consistent with the primary analysis, these results demonstrated that efgartigimod alfa produces improvements in MG-ADL response compared to placebo, regardless of prior therapies, concomitant therapies, disease duration, thymectomy, and prior treatment failure;^1,30,32,33 however, the trial was not powered to detect subgroup differences. In terms of the MG-ADL responders, similar benefit was observed in cycle 2.

Early MG-ADL responders: Early MG-ADL responders (i.e., those who responded at week 2 of cycle 1) in patients who were AChR-Ab+ was assessed as a fifth key secondary outcome. Because the statistical testing hierarchy was broken at the fourth secondary end point (i.e., time to qualify for re-treatment), the percentage of patients in the AChR-Ab+ population who were early MG-ADL responders was not statistically tested based on statistical plan in the protocol. Nevertheless, a higher proportion of patients in the efgartigimod alfa group than in the placebo group achieved an MG-ADL improvement of 2 or more points at week 2 (between-group difference = 31.9%; 95% CI, not reported [NR]) The between-group difference was considered clinically meaningful by the clinical experts consulted by CADTH. The percentage of MG-ADL early responders during cycle 2 was not assessed and not reported in the sponsor’s evidence summary.

Percentage time of the MG-ADL CMI up to day 126: Among patients who were AChR-Ab+, the percentage of time with a CMI in the MG-ADL total score up to day 126 was assessed as a third key secondary outcome and was included in the hierarchy test to control for type I error. According to the clinical experts, the percentage of time with a CMI in the MG-ADL total score in the efgartigimod alfa group was clinically meaningfully longer (between-group difference = 22.07%; 95% CI, 10.94% to 33.18%; P = 0.0001) than that in the placebo group.

MG-ADL change from cycle baseline: During cycle 1 and cycle 2, the changes from cycle baseline in MG-ADL score were assessed as exploratory outcomes for patients who were AChR-Ab+. At week 4 of cycle 1, the reduction (improvement) of MG-ADL total score in the efgartigimod alfa group was larger than that in the placebo group (between-group difference = –2.84; 95% CI, –3.8 to –1.9; P < 0.0001). This was assessed as an exploratory outcome with no multiplicity adjustment (i.e., it was not included in the hierarchy test); therefore, there is an increased risk of type I error. However, the findings were aligned with the early responder analysis and considered clinically meaningful by the clinical experts consulted by CADTH. It should be noted that the maximum MG-ADL change from cycle baseline with efgartigimod alfa appeared to occur at approximately week 4 of the cycle. The magnitude of the improvement and the comparative benefit of efgartigimod alfa compared with placebo tended to smaller at the end of the cycle. Similar results were observed in cycle 2.

Time to re-treatment: Time to qualify for re-treatment for patients who were AChR-Ab+ was assessed as the fourth key secondary outcome. The median time to qualify for re-treatment in the efgartigimod alfa group (median = 35 days; 95% CI, 29 to 43 days) was numerically but not significantly greater than the time in the placebo group (median = 8 days; 95% CI, 1 day to 30 days; log-rank P = 0.2604). The statistical hierarchy test was broken at this point. Since the week 4 visit, the proportion of patients who were qualified for re-treatment appeared to be similar in both groups (between-group difference: 1.4%; 95% CI, NR). The clinical experts that CADTH consulted for this review indicated that the results likely show that approximately half of patients would need re-treatment around week 6 of the treatment cycle.

Disease Severity (Assessed With QMG)

QMG responder during cycle 1: The percentage of QMG responders among patients who were AChR-Ab+ was assessed as the first key secondary outcome. It was reported that 49.0% (95% CI, 34.5% to 63.5%) more patients in the efgartigimod alfa group compared with the placebo group achieved a QMG response in cycle 1 (odds ratio [OR] = 7.1, 95% CI, 3.24 to 16.49; P < 0.0001). According to the clinical experts, this benefit of treatment with efgartigimod alfa compared with placebo was considered clinically meaningful.

Health-Related Quality of Life

The HRQoL (i.e., MG-QoL15r and EQ-5D VAS) was assessed as an exploratory outcome among patients who were AChR-Ab+. The change from cycle baseline in MG-QoL15r and EQ-5D VAS scores were assessed for cycle 1 and cycle 2. At week 4 of cycle 1, the reduction (improvement) in MG-QoL15r score in the efgartigimod alfa group was greater than that in the placebo group (between-group difference = –5.45; 95% CI, –7.221 to –3.685; P < 0.0001). The increase (improvement) in EQ-5D VAS score in the efgartigimod alfa group was greater than that in the placebo group (between-group difference = 13.28; 95% CI, 8.32 to 18.24; P < 0.0001). Because HRQoL was assessed as an exploratory outcome with no multiplicity adjustment (i.e., it was not included in the statistical hierarchy test), there is an increased risk of type I error; however, the results provide supportive evidence. Although there is no known minimally important difference (MID) for either the MG-QoL15r or EQ-5D VAS among patients with gMG, the clinical experts considered the results to be clinically meaningful. It should be noted that both maximum MG-QoL15r and EQ-5D VAS improvement with efgartigimod alfa occurred at approximately week 4 of the cycle. The magnitude of the improvement and the comparative benefit of efgartigimod alfa compared with placebo tended to be smaller at the end of the cycle. Similar results were observed in cycle 2.

Other Clinical Outcomes (MG Hospitalizations, MG Exacerbations, and MG Crisis)

In patients who were AChR-Ab+, during the 26-week double-blind period, the event rates for MG hospitalization and MG crisis were low in both groups. MG exacerbations were identified in 17 patients (26.2%) in the efgartigimod alfa group and 27 patients (44.3%) in the placebo group. The between-group absolute risk difference was –18.2 (95% CI, NR). The results of MG hospitalization, MG exacerbations, and MG crisis were based on post hoc analyses. Therefore, the results for these outcomes were inconclusive.

Harms Results

Reduction of side effects was identified in the patient input for this review as of interest for patients with gMG. The ADAPT trial, including its randomized controlled period and open-label extension, provided relevant information regarding the safety profile of efgartigimod alfa in the treatment of gMG. However, it did not provide direct comparative evidence regarding the adverse effects of efgartigimod alfa versus other active gMG therapies. In the AChR-Ab+ population, during the randomized controlled period, the proportion of patients with treatment-emergent adverse events (TEAEs) in the efgartigimod alfa group appeared to be similar to that in the placebo group (75.4% versus 84.4%, respectively). The proportion of patients with serious adverse events (SAEs) was low in both groups and appeared lower in the efgartigimod alfa group than in the placebo group (4.6% versus 9.4%) in the ADAPT trial. Withdrawals due to AEs occurred in similar proportions in both the efgartigimod alfa and placebo groups (3.1% and 4.7%, respectively) in the ADAPT trial. No deaths were reported during the double-blind period; however, the length of follow-up in the trial may not have been long enough to assess this outcome with certainty. The main notable harms (i.e., AEs of special interest [AESIs] for this review) were in the Medical Dictionary for Regulatory Activities (MedDRA, version 23.0) system organ class of “infections and infestations,” which were reported in a higher proportion of patients in the efgartigimod alfa group than in the placebo group (44.6% versus 34.4%). No meningococcal infections were reported. According to the clinical experts CADTH consulted for this review, the TEAEs reported in the ADAPT trial were expected and commonly seen in existing immunosuppressive treatments, such as complement C5-inhibitor treatment of gMG.

Critical Appraisal

Appropriate methods of randomization, blinding, and allocation concealment were reported. Outcomes were assessed using validated scales incorporating physician and patient assessments, and end points requiring a combination of CMI and sustained effect. However, minimally important between-group differences, that is, the thresholds used for the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) for all outcomes, were not available. Therefore, clinical experts’ opinion informed the thresholds to determine whether the between-group differences observed for each outcome were clinically meaningful. Appropriate statistical methods were used in the ADAPT trial. Multiplicity adjustments were made for the primary and 5 key secondary outcomes to control for family-wise type I error (the probability of making more than 1 type I error). Overall, the ADAPT trial was relatively well designed; however, a potentially key limitation of the ADAPT trial includes notable imbalance of baseline disease characteristics between groups. For example, the proportions of patients who had a total MG-ADL score of 10 or higher, prior combination use of steroid plus AChEI, and prior thymectomy were imbalanced between groups. Furthermore, the proportions of patients who used concomitant AChEI or concomitant steroid plus AChEI were also imbalanced between the 2 treatment groups. Whether these baseline imbalances introduced bias is uncertain. However, the clinical experts consulted by CADTH for this review indicated that these observed imbalances were unlikely to have significantly affected the study results. The efficacy outcome assessments for patients who were AChR-Ab– were assessed as sensitivity or subgroup analysis. In addition, the ADAPT trial was a placebo-controlled trial. The comparative efficacy information comparing efgartigimod alfa with existing gMG therapies (e.g., AChEIs, steroids, NSIST, IVIg, PE, complement C5 inhibitors) are unknown. Furthermore, MG-ADL total score change from cycle baseline, HRQoL, and all outcomes examined in cycle 2 were assessed as either tertiary or exploratory outcomes, which were not included in the statistical hierarchy test and were not controlled for type I error. Therefore, the results of all those tertiary and exploratory analysis should interpreted with the consideration of the lack of control for the type I error. Finally, reduction of steroid use and reduction of high-dose steroid use are treatment goals for efgartigimod alfa; however, these outcomes could not be assessed because of the study design — the concomitant treatments were not allowed to change unless it was used for rescue. The impact of efgartigimod alfa on changes in MG medications could not be evaluated because this was not allowed as per the study protocol.

According to clinical experts CADTH consulted for this review, the population included in the ADAPT trial well reflects the patients who experience unmet needs in the treatment of gMG in Canadian clinical settings. However, patients with gMG MGFA Class I (ocular MG) and Class V were excluded in the ADAPT trial. Whether the ADAPT trial findings can be generalized to patients with MGFA Class I (ocular MG) or Class V is uncertain. The clinical experts CADTH consulted for this review indicated that efgartigimod alfa will provide a new treatment for patients with gMG who are AChR-Ab+ and probably for those who are MuSK-Ab+. The clinician group input for this review also indicated that patients with MuSK antibodies might respond to efgartigimod alfa. It is uncertain whether the findings from the ADAPT trial can be generalized to patients who are AChR-Ab–, MuSK-Ab–, or double negative (AChR-Ab– and MuSK-Ab–). The number of patients who were AChR-Ab– was relatively small, and the ADAPT trial was not powered for testing the statistically significant between-group difference. Therefore, the comparative efficacy of efgartigimod alfa versus placebo for patients who are AChR-Ab– is inconclusive. In addition, although 6 (3.6%) patients in the ADAPT trial were MuSK-Ab+, there was no sensitivity or subgroup analysis for patients who were MuSK-Ab+. The comparative efficacy of efgartigimod alfa versus placebo in patients who are MuSK-Ab+ is also unknown. Therefore, findings for the overall population were mainly driven by the patients who were AChR-Ab+.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered most relevant to informing CADTH’s expert committee deliberations. A final certainty rating was determined as outlined by the GRADE Working Group.^34,35 Following the GRADE approach, evidence from RCTs was initially rated as high-certainty evidence and could be rated down for concerns related to study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias.

The selection of outcomes for GRADE assessment was based on the sponsor’s Summary of Clinical Evidence, consultation with clinical experts, and input received from patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with expert committee members: activities of daily living (proportion of MG-ADL responders in cycle 1 and cycle 2; proportion of early MG-ADL responder in cycle 1; mean proportion of time with a clinically meaningful (≥ 2-point) improvement in MG-ADL (follow-up: 126 days); time to qualify for re-treatment (up to 168 days); MG-ADL total score change from cycle baseline at week 4 of cycle 1 and cycle 2; disease severity (measured using the QMG); HRQoL (MG-QoL15r and EQ-5D-5L VAS) change from cycle baseline at week 4 in cycle 1 and cycle 2); and other clinical outcomes (MG hospitalization, MG exacerbation, and MG crisis by week 26) and notable harms (i.e., infections and infestations).

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. For this review, the target of the certainty of evidence assessment was based on the presence of absence of a clinically important effect, as informed by MIDs suggested by the sponsor and agreed upon by the clinical experts consulted by CADTH for this review (for change from baseline in MG-ADL score and QMG score), or by thresholds suggested by the clinical experts (for all other outcomes).

Results of GRADE Assessment

Table 2: Summary of Findings for Efgartigimod Alfa Versus Placebo for Patients With AChR-Ab+

AChR-Ab+ = acetylcholine receptor antibody positive; CI = confidence interval; CMI = clinically meaningful improvement; HRQoL = health-related quality of life; LSM = least squares mean; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MG-QoL15r = Revised Myasthenia Gravis Quality of Life 15-item; NA = not applicable; NR = not reported; OR = odds ratio; QMG = Quantitative Myasthenia Gravis; RCT = randomized controlled trial; VAS = visual analogue scale.

^a–1 level for serious imprecision. The 95% CI excludes the threshold of a 20% difference between groups, as informed by the clinical experts; however, the sample size and number of events does not meet the optimal information size.

^bUpon request, the sponsor did not provide the 95% CI for the between-group difference (indicated that it was not calculable).

^c–1 level for serious imprecision. No CI was available for judging the precision of the comparative effect estimate. The number of events does not meet the optimal information size.

^d–1 level for serious imprecision. The 95% CI includes the possibility of a trivial effect, based on a suggested minimally important difference of 2 points, as defined in the trial and agreed upon by the clinical experts.

^eIn the trial, statistical testing for these efficacy outcomes was not adjusted for multiplicity. The results are considered as supportive evidence.

^f–1 level for serious risk of bias. Not all patients completed the second treatment cycle, so prognostic balance between the groups cannot be ensured. –1 level for serious imprecision. No CI was available for judging precision of the comparative effect estimate. The number of events does not meet the optimal information size.

^g–1 level for serious risk of bias. Not all patients completed the second treatment cycle, so prognostic balance between the groups cannot be ensured.

^h–1 level for serious imprecision. The 95% CI includes the possibility of a trivial effect, based on a threshold of a 10% to 15% difference between groups, as informed by the clinical experts.

ⁱ–1 level for serious imprecision. The 95% CI excludes the threshold of a 20% difference between groups, as informed by the clinical experts; however, the sample size and number of events does not meet the optimal information size.

^j–1 level for serious imprecision. No CI was available for judging the precision of the comparative effect estimate. The number of events does not meet the optimal information size.

^k–1 level for serious imprecision. The 95% CI includes the possibility of a trivial effect, based on a threshold of a difference of 5 points between groups, as informed by the clinical experts.

^l–1 level for serious imprecision. The 95% CI includes the possibility of a trivial effect, based on a threshold of a difference of 10 points between groups, as informed by the clinical experts.

^m–1 level for serious risk of bias. Not all patients completed the second treatment cycle, so prognostic balance between the groups cannot be ensured. –1 level for serious imprecision. The 95% CI includes the possibility of a clinically important effect favouring efgartigimod alfa, based on a threshold of a difference of 5 points between groups, as informed by the clinical experts.

ⁿ–1 level for serious risk of bias. Not all patients completed the second treatment cycle, so prognostic balance between the groups cannot be ensured. –1 level for serious imprecision. The 95% CI includes the possibility of a trivial effect, based on a threshold of a difference of 10 points between groups, as informed by the clinical experts.

^o–1 level for serious risk of bias. The analyses of these outcomes were undertaken post hoc, so there is risk of bias in the selection of the reported result. –2 levels for very serious imprecision. No CI was available for judging the precision of the comparative effect estimate. The number of events does not meet the optimal information size; there were very few or no events in either group.

^p–1 level for serious risk of bias. The analyses of these outcomes were undertaken post hoc, so there is risk of bias in the selection of the reported result. –1 level for serious imprecision. No CI was available for judging the precision of the comparative effect estimate. The number of events does not meet the optimal information size.

^q–1 level for serious imprecision. The 95% CI includes the possibility of a trivial effect, based on a threshold of a 10% difference between groups, as informed by the clinical experts.

Sources: Clinical Study Report;³⁰ the sponsor’s submission^1; and sponsor-provided additional information.^36,37 Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Long-Term Extension Studies

Description of Studies

The ADAPT+ study (ARGX-113-1705) is a long-term, single-arm, open-label, multicentre, phase III follow-on study of patients who enrolled in the ADAPT study (ARGX-113-1704; NCT03669588). The primary objective was to evaluate the long-term safety and tolerability of efgartigimod alfa in the AChR-Ab+ subgroup, and the secondary objective was to evaluate safety and tolerability in the overall population (AChR-Ab+ and AChR-Ab–). Efficacy data were collected as exploratory end points. The ADAPT+ study was conducted at 51 sites, including 41 sites in 14 countries or regions that had 1 or more patient roll over from the ADAPT study. Data were collected over a 3-year period in 2 sequential parts (part A: 1 year, part B: 2 years maximum). Part B was added as a protocol amendment to ensure accessibility to efgartigimod alfa until it became commercially available or available through an expanded access program. Results of the long-term extension phase up to 14 cycles (ADAPT+) are also presented in this report. The ADAPT+ trial was still ongoing at the time of this review; therefore, the long-term efficacy and safety outcomes of ADAPT+ were based on interim analysis (IA4 and IA5). MG-ADL, QMG, and safety outcomes were assessed in the long-term extension study. At the data cut-off (June 30, 2022), 151 patients had rolled over from the ADAPT trial, regardless of treatment or placebo group, into ADAPT+ and 145 patients had received at least 1 partial or complete dose of efgartigimod alfa in ADAPT+.

Efficacy Results

In terms of MG-ADL response (up to 14 cycles) and QMG response (up to 7 cycles), evidence from the long-term, open-label, extension (ADAPT+) trial appeared consistent with those from the randomized controlled period. Patients who switched from placebo to efgartigimod alfa experienced numeric improvements from baseline in MG-ADL and QMG in each cycle. However, interpretation of these data was limited by the open-label and descriptive nature of the extension study.

Harms Results

Safety data from the long-term extension phase appeared consistent with that observed in the double-blind phase with no new safety signals reported.

Critical Appraisal

Internal Validity

The ADAPT+ study was limited by its open-label and noncomparative design because it was uncertain whether the results observed were attributable to the effects of the drugs including other treatments or the natural history of disease. Furthermore, the missing outcome data and the small sample size toward the end of ADAPT+ made it difficult to draw any firm conclusions on the efficacy and safety of efgartigimod alfa. Due to its open-label nature of ADAPT+, the subjective outcomes (e.g., rates of self-reported AEs) were at risk of bias and potentially in favour of the intervention (i.e., efgartigimod alfa). It is noteworthy that ADAPT+ had fewer scheduled visits for outcome assessments compared with the ADAPT study. ADAPT+ only collected MG-ADL data at week 3 of each study; however, the maximum clinical effect in ADAPT was observed at weeks 4 to 5 of a cycle. Furthermore, the longer-term safety and tolerability profile of efgartigimod alfa treatment was hard to determined due to the rates of AEs and SAEs may be underestimated with the less frequent assessment schedule. In ADAPT+, efficacy was assessed as an exploratory outcome using the patient-reported MG-ADL and physician-reported QMG scales. Patients were to remain on their stable dose and regimen of concomitant gMG treatment during part A of ADAPT+. Given the presence of rollover effects, efficacy results related to part B may be difficult to interpret due to that changes were permitted in part B, including changes in the type, dose, or regimen of the concomitant gMG treatment as well as the additional use of other treatments. Therefore, the confounding effects of other therapies cannot be eliminated in part B. In terms of outcome measures, some important long-term outcomes reported by patients and clinicians were not measured (e.g., HRQoL, exacerbations) in the ADAPT+.

External Validity

Because the patients who took part in the open-label, long-term safety extension phase were originally from the pivotal ADAPT trial, it is reasonable to expect that the same limitations to generalizability are relevant to the extension phase. Given the nature of noncomparative study design, it is not possible to compare the effectiveness and tolerability of efgartigimod alfa as an add-on treatment of gMG against other add-on treatments (e.g., IVIg).

Indirect Comparisons

Description of Studies

To date, there have been no clinical trials directly comparing the efficacy of efgartigimod alfa with other treatments in patients diagnosed with gMG. Due to this gap in evidence, the sponsor submitted an indirect treatment comparison (ITC) that included a systematic literature review (SLR)³⁸ and a network meta-analysis (NMA)¹ that provide comparative evidence of the efficacy of efgartigimod alfa relative to ravulizumab and IVIg. The eligible interventions for the ITC were limited to those used in Canada for the treatment of gMG to ensure that the comparators were relevant to the Canadian settings. After feasibility assessment, the following 5 studies were considered eligible to be included in the NMA: 2 studies comparing efgartigimod alfa with placebo,^28,39 2 studies comparing IVIg with placebo,^40,41 and 1 study comparing ravulizumab with placebo.⁴² All NMAs were performed using a Bayesian framework. Placebo was chosen as the reference treatment for all analyses, given its presence as an anchor treatment in all studies and the outcomes assessed in the network. The clinical end points used for ITC estimates included change from baseline in MG-ADL and QMG as these were the most consistently reported outcomes in all studies included in the NMA. Primary analyses were performed at the primary assessment time points for all included studies, ranging from 4 to 26 weeks, and sensitivity analyses were performed at or plus or minus 2 weeks of week 4, which was the primary assessment time point in the ADAPT trial.

As the sponsor’s reimbursement request is limited to patients in the AChR-Ab+ subpopulation, comparators relevant to that group were used in the primary ITC analysis.

Efficacy

Primary Analyses

The mean differences for change from baseline in MG-ADL were –2.64 (95% credible interval [CrI], –4.16 to –1.12) for efgartigimod alfa versus IVIg, and –0.91 (95% CrI, –2.25 to 0.39) for efgartigimod alfa versus ravulizumab. The mean differences for change from baseline in QMG were –4.39 (95% CrI, –6.95 to –1.81) for efgartigimod alfa versus IVIg, and –2.89 (95% CrI, –4.72 to –1.12) for efgartigimod alfa versus ravulizumab. A change of 2 points in the MG-ADL score and 3 points in the QMG score was estimated to be the threshold of clinical significance in patients with MG.

Sensitivity analyses results for change from baseline in MG-ADL and QMG at or plus or minus 2 weeks of week 4 were consistent with results of the primary analyses.

Additional Analyses

The mean differences for change from baseline in MG-ADL were –2.64 (95% CrI, –4.18 to –1.12) for efgartigimod alfa versus IVIg, –0.92 (95% CrI, –2.25 to 0.43) for efgartigimod alfa versus ravulizumab, and –1.93 (95% CrI, –3.87 to 0.07) for efgartigimod alfa versus rituximab. The mean differences for change from baseline in QMG were –4.39 (95% CrI, –7.01 to –1.83) for efgartigimod alfa versus IVIg, –2.89 (95% CrI, –4.72 to –1.06) for efgartigimod alfa versus ravulizumab, and –2.71 (95% CrI, – 5.56 to –0.2) for efgartigimod alfa versus rituximab.

Harms

No analysis of harms was reported in the sponsor-submitted ITC report.

Critical Appraisal

The SLR used to identify relevant studies was methodologically sound in terms of the sponsor using a comprehensive literature search strategy as well as performing study selection, data extraction, risk-of-bias assessments in duplicate, and providing a list of excluded studies and justifying the exclusions. However, it was unclear in the ITC report whether the feasibility assessment was carried out by a single or multiple assessors. By conducting a feasibility assessment, the sponsor excluded all head-to-head trials, including those comparing efficacy of IVIg treatment versus PE, which may have reduced the information to inform the NMA. The risk of bias of included studies in the SLR was assessed per individual study; however, this may differ according to the outcomes of each study. Analyses were run using a Bayesian framework with placebo as the reference treatment, which was deemed appropriate. Change from baseline in MG-ADL and QMG scores were considered the best source of comparative efficacy data for this NMA, although these outcomes were not primary or secondary end points of the ADAPT trial. The studies that did not report on MG-MDL or QMG were excluded even if they reported other relevant outcomes, which may have biased the results, although the extent of bias is uncertain. All trials included in the ITC had sufficiently similar study designs and a common comparison group (placebo). However, there were some important differences between the trials included in the NMA that increase the uncertainty of the analyses. All included studies employed a dosing schedule involving spaced infusions, but only ADAPT used individual patient response to determine subsequent cycles of treatment. The studies included in the ITC analyses ranged in follow-up time from 4 to 26 weeks. All studies allowed the use of concomitant standard-of-care treatments (e.g., corticosteroids, NSISTs), but detailed information on the breakdown of actual concomitant medications used was not available. In many studies, baseline data were not reported consistently, such as for MGFA at baseline, use of steroids or NSISTs at baseline, disease duration, and history of thymectomy. The primary analyses conducted at the primary assessment time point for all trials could be biased against ADAPT, as they could exclude the best responders to efgartigimod alfa, whereas ITCs conducted at week 4 only could be biased against any treatments that demonstrated improved responses over time. Therefore, sensitivity analyses were performed at or plus or minus 2 weeks of week 4 to improve the robustness of the ITCs and align with the primary assessment time point of the ADAPT trial.

The results were reported as mean differences and 95% Crls. The evidence is imprecise in the effect estimates from the NMA due to the sparseness of data, with wide Crls. In addition, heterogeneity between the included studies can potentially introduce bias into the study estimates observed between the comparators. Because all comparator studies were performed with patients who were exclusively AChR-Ab+, all ITC analyses included only patients from the AChR-Ab+ subpopulation, which aligns with the reimbursement request submitted by the sponsor and the approved Health Canada indication. Another important limitation of the presented ITC is the lack of safety and HRQoL data. The results of this ITC are highly uncertain given the inconsistency between trials with respect to dosing regimen (individualized dosing for efgartigimod alfa versus continuous dosing for the comparators), variability in eligibility criteria, and study follow-up times. The ITC estimates were too imprecise to draw a conclusion about the comparative effect of efgartigimod alfa relative to alternative treatments on change from baseline in MG-ADL and QMG.

Conclusions

One double-blind, RCT of patients with gMG was included in this review.

Evidence from the ADAPT trial showed that, compared with placebo, treatment with efgartigimod alfa as an add-on to standard conventional therapy likely results in a clinically meaningful benefit in terms of the proportion of MG-ADL and QMG responders, and HRQoL (as assessed using the MG-QoL15r and EQ-5D VAS) after cycle 1 of treatment relative to placebo among adult patients with AChR-Ab+ gMG whose symptoms persist despite a stable dose of standard-of-care (concomitant gMG treatment) treatment with AChEIs, corticosteroids, and/or NSISTs (moderate certainty). Similar benefit was observed in cycle 2, although there is less certainty in these results because not all randomized patients participated in this cycle. Efgartigimod alfa likely results in a clinically important increase in the percentage of time with a meaningful MG-ADL improvement compared with placebo (moderate certainty). Maximum benefit appeared to occur at approximately 4 weeks of each cycle. Efgartigimod alfa may result in a clinically important reduction in MG exacerbations relative to placebo (low certainty); the results were inconclusive for MG-related hospitalizations and MG crises due to low numbers of events reported for these outcomes. The safety profile of efgartigimod alfa reported in the ADAPT trial was considered as expected and commonly seen in existing gMG therapies. In terms of MG-ADL and QMG scores and safety profile, evidence from the long-term, open-label, extension (ADAPT+) trial appeared consistent with those from the randomized controlled period in the patients who were AChR-Ab+. However, interpretation of the long-term data was limited by the open-label and descriptive nature of the extension study. The results of the sponsor-submitted NMA suggest that relative to IVIg and ravulizumab, efgartigimod alfa may provide a benefit with respect to change in MG-ADL and QMG scores; however, the 95% CrIs for the effect estimates included the possibility of trivial effects (i.e., only small, nonclinically important differences between groups) and no difference (in the case of change in MG-ADL score relative to ravulizumab). No difference in efficacy in terms of change from baseline in MG-ADL and QMG scores could be concluded for efgartigimod alfa relative to rituximab due to wide 95% CrIs (which included the possibility of clinically important benefit favouring efgartigimod alfa), and methodological limitations. There is no evidence for the effect of efgartigimod alfa on HRQoL or harms outcomes relative to any other active treatment.

Introduction

Disease Background

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

MG is a rare, chronic, neuromuscular autoimmune disease mediated by pathogenic autoantibodies that target structural components of the neuromuscular junction, impairing neuromuscular transmission, and leading to muscle weakness and fatigue,^2-4 extensively disrupting the ability to perform normal daily activities and profoundly impairing HRQoL. Many patients initially present with symptoms affecting only the eye muscles (i.e., ocular MG). Approximately 85% of patients go on to develop gMG, with generalized weakness affecting the neck, trunk, limbs, and bulbar and respiratory muscles. The characteristic feature of gMG is fluctuating fatigable muscle weakness, although there is heterogeneity in the specific muscles that are affected. Patients with gMG experience symptoms that negatively impact HRQoL.⁵ The disease has a fluctuating natural history, with MG exacerbation (an increase in symptoms in patients who were previously asymptomatic or minimally symptomatic, defined as a ≥ 3-point worsening in QMG score versus baseline) and myasthenic crisis (muscle weakness causing life-threatening difficulties with breathing and swallowing that requires ventilator support) can occur gradually or without warning.⁶ The serological profile of gMG can be defined by the presence of autoantibodies to particular receptors, which can affect treatment decisions and disease prognosis.⁴³ Approximately 85% of patients with gMG are AChR-Ab+ and as many as 15% are AChR-Ab–.^7,8 An estimated 1% to 10% of patients do not have AChR antibodies but do have autoantibodies against MuSK-Ab+ or autoantibodies against LRP4-Ab+, which also lead to a decrease in AChRs.³ The MGFA classification system is a tool used to categorize gMG based on clinical features and/or disease severity.⁴⁴ The classification ranges from Class I (i.e., ocular weakness only) to Class V (i.e., patients require intubation, with or without mechanical ventilation, except when employed during routine postoperative management or myasthenic crisis); MGFA Class II to V are used to characterize gMG. Class II, Class III, and Class IV represent patients with mild, moderate, and severe muscle weakness, respectively.⁹ The incidence of MG in Canada is estimated at 23 cases per 1 million person-years, with a prevalence of 32 cases per 100,000 adults (0.032%) in Canada.^10-12 Thus, there are approximately 8,121 patients with MG across the CADTH-participating drug programs (0.032% × 25,376,703 adult patients in CADTH-participating drug programs in 2023). Among adults with MG, approximately 85% are anticipated to progress to gMG, which corresponds to approximately 6,903 adult patients with gMG in Canada.

Standards of Therapy

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

The clinical experts CADTH consulted for this review indicated that the goal of treatment in most patients with gMG is to reduce disease symptoms as well as adverse effects of MG therapy to allow the patient to function as they would normally with good HRQoL. Other goals of treatment include avoiding MG exacerbations and myasthenic crises, minimizing hospitalizations and ICU admissions, and reducing the numbers and doses of therapies (especially corticosteroid use) required for symptom control. The available main therapies of gMG include AChEIs, corticosteroids, NSIST, rituximab, IVIg, PE or PP, and terminal complement inhibitors (i.e., ravulizumab and eculizumab). According to the clinical experts consulted by CADTH for this review, the first-line standard of care (i.e., the conventional therapy) for MG are AChEIs, corticosteroids, and NSISTs (e.g., azathioprine, mycophenolate mofetil, cyclophosphamide, cyclosporine, tacrolimus, and methotrexate). Mild to moderate gMG (MGFA Class II or IIIa) is initially treated symptomatically with AChEIs (usually pyridostigmine);¹³ the onset of benefit occurs in hours to days. If this provides insufficient symptom relief, IST with corticosteroids (usually prednisone) is administered;¹⁴ maximal responses typically occur 2 to 6 months later, after which slow tapering of corticosteroids is begun. In patients who do not respond to corticosteroids, who have significant comorbidities such that long-term corticosteroid treatment is contraindicated, or whose doses of corticosteroids cannot be tapered, treatment with NSISTs¹⁵ and/or immunomodulatory drugs, including rituximab, may be initiated.¹⁶ The clinical experts stated that the onset of benefit from NSISTs occurs in months to years (approximately 9 to 18 months for azathioprine and mycophenolate mofetil). Rituximab is a monoclonal antibody directed against the CD20 receptor on B-lymphocytes. Rituximab has not been approved as a gMG treatment by Health Canada. International guidelines have identified evidence for the use of rituximab in patients with MuSK-Ab+ gMG who have an insufficient response to other immunotherapies; and it is considered a treatment option in Canada for patients with refractory gMG who are AChR-Ab+, according to surveys conducted with clinical experts.¹⁶ According to the clinical experts, for patients with moderate to severe gMG, especially those who have respiratory or bulbar weakness, IVIg, PE, or PP may be administered^17,18 in addition to rituximab, either at the time of IST initiation or to treat MG exacerbation or myasthenic crisis. Critical care, including ICU admission and ventilator support, may be required for patients experiencing myasthenic crisis. Thymectomy may also be considered in a small group of patients.¹⁶ As MG symptoms improve, doses of AChEIs, corticosteroids, and then other ISTs are reduced and the frequency of IVIg, PE, or PP is also reduced until the minimal maintenance therapy required for remission is identified. Patients with refractory gMG who are AChR-Ab+ may be candidates for the complement inhibitor eculizumab. While eculizumab received a recommendation for reimbursement with conditions in 2020,¹⁹ price negotiations concluded without an agreement in December 2022.²⁰ A survey of 7 expert clinicians from across 6 provinces indicated that eculizumab would be another treatment option if it were to be approved and funded, while ravulizumab would be a treatment option for patients who have an inadequate response to conventional therapy.¹⁸ In April 2023, CADTH issued a draft “do not reimburse” recommendation for ravulizumab in this indication.²¹ The clinical experts consulted by CADTH for this review emphasized that most patients with gMG (more than 80%) will respond well to currently available treatments: although these cannot cure the disease, excellent symptom control is achieved in most patients and prognosis is generally good in terms of muscle strength and function as well as HRQoL. Despite treatment with conventional therapy (AChEIs, corticosteroids, and/or NSISTs), many patients continue to experience disease burden and symptoms that impact their HRQoL,^22-27 and treatment-related side effects may be severe. Approximately 25% of patients with gMG are able to achieve pharmacological remission (i.e., symptom control with conventional therapy), but only 8% achieve clinical remission (i.e., no symptoms off-treatment for more than a year),⁴⁵ which highlights the need for additional options to improve disease control. A survey of 7 clinicians across 6 provinces¹⁸ indicated that the characteristics of patients enrolled in the ADAPT trial were generally aligned with the characteristics of patients with gMG in their practice. Furthermore, the clinicians indicated that they saw the greatest benefit of efgartigimod alfa treatment for patients with AChR-Ab+ who require chronic IVIg or PE add-on treatment. Therefore, based on these opinions and the clinical evidence from ADAPT and ADAPT+, the sponsor proposes that efgartigimod alfa be used as an add-on treatment for adult patients with AChR-Ab+ gMG whose symptoms persist despite adequate treatment with conventional therapy (any combination of AChEIs, corticosteroids, and/or NSISTs, at the treating physician’s discretion).¹

Drug Under Review

Key characteristics of efgartigimod alfa and other treatments available for adult patients with gMG are summarized in Table 3. Efgartigimod alfa is a first-in-class human IgG1 antibody Fc-fragment that blocks FcRn.^28,29 Efgartigimod alfa is supplied as a 20 mg/mL solution, and 10 mg/kg is administered as an IV infusion over 1 hour once weekly for 4 doses (i.e., weeks 0 to 3). In patients weighing 120 kg or more, the recommended dose of efgartigimod alfa is 1,200 mg (3 vials) per infusion. Efgartigimod alfa binds to FcRn, resulting in the reduction of circulating IgG including autoantibodies. Efgartigimod alfa does not affect the levels of other immunoglobulins (IgA, IgD, IgE, or IgM) or of albumin. IgG autoantibodies are the underlying cause of the pathogenesis of MG. They impair neuromuscular transmission by binding to AChRs, MuSK, and LRP4. Efgartigimod alfa reduces the levels of pathogenic IgG autoantibodies.²⁹

Efgartigimod alfa received a Health Canada Notice of Compliance for the treatment of adult patients with gMG who are AChR-Ab+ on September 19, 2023. The sponsor’s reimbursement request is that efgartigimod alfa be reimbursed as an add-on therapy for adult patients with gMG who are anti-acetylcholine receptor (AChR) antibody positive, whose symptoms persist despite adequate treatment with AChEIs, corticosteroids, and/or NSISTs, which is a subgroup of the approved Health Canada indication.¹

Table 3: Key Characteristics of Efgartigimod Alfa and Other Drugs Used for the Treatment of gMG

Ab+ = antibody positive; AChEI = acetylcholinesterase inhibitor; AChR = acetylcholine receptor; gMG = generalized myasthenia gravis; FcRn = neonatal crystallizable fragment receptor; IgG1 = immunoglobulin G subclass 1; IVIg = intravenous immunoglobulin; NA = not applicable; PE = plasma exchange; PP = plasmapheresis.

^aRelevant Health Canada–approved indications.

^bSupplemental ravulizumab doses of 1,200 mg to 1,800 mg are given after PE or PP, and supplemental doses of 600 mg are given after IVIg.

^cSupplemental eculizumab doses of 300 mg to 600 mg are given after PE or PP.

Sources: Product monographs for efgartigimod,²⁹ ravulizumab,⁴⁶ eculizumab,⁴⁷ and pyridostigmine;⁴⁸ and the sponsor’s submission.¹

Stakeholder Perspectives

Patient Group Input

This section was prepared by the CADTH review team based on the input provided by patient groups. The full original patient input(s) received by CADTH have been included in the Stakeholder section of this report.

CADTH 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 searching for a cure through well-funded research.

MDC identified and contacted adults living with MG to participate in a survey and semistructured interviews. Surveys were shared with members via e-blasts, personalized invites, and online patient groups. MDC also conducted a journey mapping project for adults living with MG via virtual interviews, roundtable sessions, surveys, and HRQoL measures (EQ VAS, EQ-5D, MG-ADL, MG-QoL). MDC collected information from 108 individuals impacted by MG, including 39 males and 69 females between ages 21 to 76 from all provinces in Canada. MDC used a qualitative descriptive approach, employing a constant comparison technique, to produce a thematic analysis.

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

Some of the respondents explained the impact of MG using the following phrases:

• I had to retire early

• I am fearful of the next MG crisis

• I am on disability leave

• MG has financial impacts

• I feel useless at home

• feeling tired

• need a full day to recover if I do too much

• after 10 minutes of housework then have to rest

• feel my lungs are weaker

• had to go on a ventilator in ICU

• choking on food or saliva interferes with breathing

• can’t even walk inside my house

• always keep a walker or cane nearby because you never know when the MG will flare up

• can't sleep at night because it aches

• unable to drive

• on bad days I feel like a prisoner in my own house

• I miss out on socializing

• quite disabled and dependent on others

• depressing knowing that there is no cure and that this is my way of life now

• scared to be alone

• have to puree my foods

• my voice sounds very strained after a short conversation

• I had slurred speech as though I was intoxicated

• I frequently go cross-eyed

• double vision interferes with reading

• multiple acute hospitalizations.

When the respondents were asked how their MG was being managed with the available treatments, 3 main themes emerged from the analysis: negative experiences with steroids (e.g., adverse effects; costs); the slow onset of medication effects (e.g., example, 1 patient respondent indicated, “My doctor told me it could take 6, maybe even 9, months for the treatment to take effect”; another patient respondent stated, “Imagine living half the year waiting for a drug to show benefit and then to find out you need to be switched to something else”); and a feeling of going through a process of trial and error with medications. Although the treatment had a positive impact on health outcomes, the participants had concerns about the long-term and sustained benefits of supportive treatments.

Regarding the improved outcomes, the patient group identified 3 aspects of MG that they wanted better controlled: decreased intensity of disease exacerbations and medication side effects, maintenance of independence, and fewer serious hospital admissions. The method, duration, and frequency of medication administration and the convenience and cost of treatment were considered very important to the patients and caregivers. They preferred less travel, fewer hospital visits, and less invasive methods of treatment. HRQoL was noted to be a key priority over convenience of the drug. Respondents stated that they would be willing to deal with side effects of medications if these aspects of MG were better controlled. They also stated that although their current medications decreased the number of exacerbations, they do not have an impact on overall quality of life. They expected new treatments to help them become independent, stop the myasthenic crises, address the respiratory and general weakness, be easier to swallow (for pills), reduce pain, not lead to diabetes, be target treatment for MG instead of generally immunosuppressive, be less expensive, work quickly, be a single daily dose in the morning.

One respondent had received Vyvgart as a participant in a clinical trial and explained that this medication replaced the need for IVIg, the effects appeared to be quicker compared with other therapies, the infusion time was shorter than expected, treatment was received less frequently compared with other therapies, and they experienced fewer side effects compared with other therapies. This respondent highlighted that although diarrhea was a problem and not unique to Vyvgart, it was manageable after the first cycle of treatment.

All the respondents had undergone diagnostic blood testing, and many had undergone single-fibre electromyography to confirm diagnosis. A total of 80% of the respondents reported difficulty receiving a diagnosis. Based on early findings of the MG journey mapping project, MDC reported 7 years to be the average time from first bothersome symptom to diagnosis, with the range up to 23 years. According to MDC, the majority of respondents found the process of testing and diagnosis cost-effective but lengthy, with many missed opportunities, delays in diagnosis, and misdiagnoses (such as stroke and Bell palsy), resulting in incurred costs. Those who were diagnosed during a MG crisis or hospitalization (25% of respondents) reported a smooth diagnosis.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise on the diagnosis and management of the condition for which the drug is indicated. Clinical experts are an essential part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 2 clinical specialists with expertise in the diagnosis and management of gMG.

Unmet Needs

The clinical experts indicated that approximately 90% of patients respond to current treatments, but response is often partial, meaning that there are still symptoms that affect quality of life and function. Besides prednisone and rescue treatments, immunosuppressive therapies take a very long time to act (e.g., azathioprine takes at least a year). That means that patients may be exposed to higher doses of steroids for longer periods, and experience persistent symptoms for long periods, before even knowing whether a medication will be effective. Current treatments are nontargeted, causing overall more diffuse immunosuppression, and there is an increased risk of cancer with long-term use. Also, there is an increased risk of steroid-related AEs with prolonged doses, which requires periodic monitoring. The clinical experts also indicated that the fraction of patients considered to have refractory gMG varies according to the definition, but that 10% to 20% of the total population is a reasonable estimate. The risk of generalized immunosuppression is principally related to infection (whether azathioprine affects risk of malignancy is debated, and the long-term data are limited).

Place in Therapy

According to the clinical experts consulted by CADTH for this review, patients with gMG usually start with pyridostigmine (symptomatic treatment) but most patients will need disease-modifying treatment with immunosuppression, most commonly prednisone. Depending on severity, age, and comorbidities, an NSIST (e.g., azathioprine, mycophenolate mofetil, tacrolimus) may be started soon after diagnosis or later (e.g., inability to reduce dose of steroids). Some patients with severe disease at onset (e.g., crisis or severe symptoms) may receive a rescue treatment such as IVIg or PE early for fast improvement while immunosuppression kicks in. A few of patients receive chronic IVIg or PE, but some are dependent on these. The treatment goals are to achieve minimal symptoms or remission, with the fewest AEs from treatments. Patients express the need to improve their ability to perform their daily life activities, reduce fatigue, and be able to care for their family, work, or carry out home obligations.

The clinical experts stated that efgartigimod alfa has a specific mechanism of action related to the pathophysiology of MG (reduction of IgG levels, including AChR antibodies), but that the same could be said about pyridostigmine. Efgartigimod alfa reduces levels of IgG but does not affect the process of producing AChR antibodies. The clinical experts indicated that they do not foresee efgartigimod alfa being used as first-line treatment. Rather it would be suitable for individuals without response to available treatments, for those dependent on IVIg or PE, or for those with very severe disease, to bridge the delayed action of standard immunosuppressive therapies. Patients should receive standard treatments first, as these will be satisfactory in a large number of patients. The clinical experts also stated that most treatments for patients with gMG (for example, prednisone, azathioprine, mycophenolate mofetil, tacrolimus, rituximab, and others) are given off label given the lack of RCT evidence. IVIg or PE used for rescue therapy is different from IVIg or PE used chronically (i.e., maintenance therapies, once a month); only a small number of patients use IVIg or PE chronically. Eculizumab or ravulizumab are used too rarely at present to include them as comparators. One clinical expert indicated that it could be argued that efgartigimod alfa be tried as an initial therapy by patients for whom pyridostigmine alone was ineffective; however, the cost is likely to be a major barrier.

Patient Population

The clinical experts indicated that efgartigimod alfa will provide a new treatment for patients with gMG who are AChR-Ab+ and probably for patients who are MuSK-Ab+. However, whether patients with gMG who are seronegative (e.g., AChR-Ab–, MuSK-Ab–) respond to efgartigimod alfa is unknown because few seronegative patients were included in the ADAPT trial. One clinical expert indicated that, based on efgartigimod alfa’s mechanism of action, AChR-Ab+, MuSK-Ab+, and likely “seronegative” patients would be expected to respond (“seronegative” patients are generally assumed to have as-yet-unidentified autoantibodies).

Assessing the Response Treatment

The clinical experts indicated that in most clinics (not academic) patients are not given standardized assessments. In academic settings, clinicians use validated measures. The MG-ADL used in trials is easy to use, and can be easily incorporated into routine clinical practice and all settings. The clinical experts recommended using the MG-ADL for patients receiving active treatment at all visits, as this enables following the clinical course. An improvement (reduction) of 2 points is considered significant. Apart from using symptom scores to assess improvements, overall function, ability to return to work, and so on, are also assessed. The ability to reduce or stop chronic use of corticosteroids, IVIg, or PE is an important outcome when considering the use of efgartigimod alfa. Some patients are dependent on chronic IVIg or PE, so it is important to be able to wean off these. Some patients have multiple hospitalizations due to MG, so reducing or avoiding these is also important. The frequency of assessments depends on patients’ symptoms or stability. Patients whose symptoms are well-controlled are typically seen by clinicians every 6 months but can be seen more frequently (e.g., every 2 to 3 months) in case of worsening health, new medications, and so ono. For efgartigimod alfa, most responders were fast, but a small proportion of responders lagged, and response could be seen at the second cycle. Therefore, 2 to 3 months to assess response to cycle 1 or the need for a new cycle is reasonable. Patients would then need to be seen at 6 months to determine nonresponders. For responders, subsequent assessments every 3 to 6 months could be done to determine if new cycles are needed.

Discontinuing Treatment

The clinical experts indicated that efgartigimod alfa should be discontinued if there is no response to treatment (no improvement in symptoms/function); if a severe AE (e.g., severe infusion reactions) occurs; if rescue treatment (IVIg or PE, or increased dose of steroids) is needed; or if chronic use of corticosteroids, IVIg, or PE cannot be reduced.

Prescribing Considerations

The clinical experts indicated that patients should be under the care of a neurologist with experience diagnosing and treating MG, usually a neuromuscular specialist. The infusion itself can be arranged at infusion clinics.

Additional Considerations

None.

Clinician Group Input

This section was prepared by the CADTH review team based on the input provided by clinician groups. The full original clinician group input(s) received by CADTH have been included in the Stakeholder section of this report.

CADTH received 1 clinician group submission from NMD4C, which was launched in January 2020 with funding from the Canadian Institutes of Health Research and MDC. The mission of NMD4C is to improve the care, research, and treatment of neuromuscular diseases for all people living in Canada. Its vision is to be a comprehensive, inclusive, open, and enduring network through which stakeholders can share expertise and data and collaborate on joint activities and research for the benefit of all patients in Canada.

Clinicians with experience in treating gMG were asked to contribute to this submission. The information presented in this submission was gathered from one-to-one discussions with the lead author and through group discussions.

NMD4C stated that conventional treatment options for gMG have been based on symptomatic therapy, short-term rescue immunotherapy, and long-term immunosuppressive therapy. Moreover, nonspecific immunosuppressants have been only partially effective and many patients do not attain stable remission, with 10% to 20% not responding or intolerant to these drugs.

According to NMD4C some of the unmet needs of the standard treatments are side effects, lack of effectiveness for all patients, long periods of treatment, and transient effectiveness. Another unmet need in this field is the lack of therapeutic options for seronegative patients.

NMD4C noted that patients with AChR antibodies in their system will most likely respond to the drug under review. Patients with MuSK antibodies and those who are double seronegative might respond. Patients who get worse quickly, particularly patients with MG crisis, are most in need of an intervention that works quickly, but patients who have symptoms restricted to their ocular muscles are unlikely to require such rapid intervention with the drug under review. The patients best suited for treatment with the drug under review are identified by clinician examination and judgment supplemented by assessment of activities of daily living using scales that reflect severity of disease, such as the QMG, MGII, and SSQ. If these are not available, then antibody testing needs to be done, although this can be delayed, according to the clinician group. Currently, there is no clear way at this point to predict which patients are more likely to respond except the presence of AChR antibodies.

NMD4C indicates that diagnosis of double seronegative patients is an issue because cluster antibodies to both acetylcholine receptor and MuSK may be present but need to be tested specifically, which can take weeks.

The clinician group indicated that to reduce the risk of underdiagnosis, serological testing as well as single-fibre electromyography and repetitive nerve stimulation studies are necessary. To determine patients’ response to therapy, scales such as the MG-ADL, QMG, MGII, and SSQ at 2 and 4 weeks are required; after that, the assessment should be based on the patient’s status. The clinician group noted that a clinically meaningful response to treatment used in the clinical trials is 2 or more points on the ADL and 3 or more points on the QMG scale. For the SSQ, the clinician group suggested that levels above 72% indicate general satisfaction. In case of lack of response, discontinuation of treatment should be considered.

The clinician group mentioned that usual Ig treatment for MG can be effective but place a significant burden on the Canadian health care system and that supplies can be at risk in situations such as a pandemic. They think the drug under review is likely to replace Ig therapies.

In summary, the clinician group’s input is aligned with the input provided by the clinical experts consulted by CADTH.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s Reimbursement Review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 4.

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

AChEI = acetylcholinesterase inhibitor; AChR-Ab+ = AChR antibody positive; CDEC = Canadian Drug Expert Committee; CMI = clinically meaningful improvement; CS = corticosteroid; Fc = fragment crystallizable; gMG = generalized myasthenia gravis; IgG = immunoglobulin G; IMP = investigational medicinal product; IVIg = intravenous immunoglobulin; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGII = Myasthenia Gravis Impairment Index; NOC = notice of compliance; NSIST = nonsteroidal immunosuppressive therapy; PE = plasma exchange; QMG = Quantitative Myasthenia Gravis.

Figure 1: Anticipated Place in Therapy of Efgartigimod Alfa for Anti-AChR Antibody-Positive gMG

AChEI = acetylcholinesterase inhibitor; CS = corticosteroid; NSIST = nonsteroidal immunosuppressive therapy.

Clinical Evidence

The objective of CADTH’s Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of efgartigimod alfa (VYVGART), 20 mg/mL solution, IV infusion in the treatment of adult patients with gMG. The focus is on comparing efgartigimod alfa to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence provided by the sponsor is presented in 3 sections, with CADTH’s 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. CADTH’s assessment of the certainty of the evidence in this first section using the GRADE approach follows the critical appraisal of the evidence. The second section includes sponsor-submitted long-term extension studies. The third section includes indirect evidence provided by the sponsor.

Included Studies

Clinical evidence from the following is included in the CADTH review and appraised in this report:

• 1 pivotal phase III double-blind RCT (ADAPT) identified in the systematic review

• 1 long-term extension study (ADAPT+)

• 1 ITC.

Systematic Review

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

Description of Studies

Key characteristics of the ADAPT trial are shown in Table 5, and a summary of its design is shown in Figure 1. The ADAPT trial (ARGX-113-1704; NCT03669588; N = 167)³⁰ was a randomized, double-blind, placebo-controlled, multicentre, phase III trial designed to evaluate the efficacy, safety, and tolerability of efgartigimod alfa and its impact on HRQoL and normal daily activities in patients with gMG. The study has been completed and the data cut-off (last patient completed) was April 6, 2020. Patients were recruited from 56 centres in the US, Europe, Japan, and Canada, including 4 patients were enrolled at 3 sites in Canada. The primary objective was to evaluate the efficacy of efgartigimod alfa as assessed by the percentage of MG-ADL responders in the AChR-Ab+ population after the first treatment cycle. The secondary objectives were to evaluate the safety and tolerability of efgartigimod alfa in the overall population and in subgroups and to evaluate the efficacy of efgartigimod alfa based on:

• the percentage of QMG responders after the first treatment cycle in the AChR-Ab+ population

• the percentage of MG-ADL responders after the first treatment cycle in the overall population (patients with AChR-Ab+ and AChR-Ab–)

• the percentage of time that patients show a CMI in the MG-ADL total score during the study (up to and including day 126) in the AChR-Ab+ population

• the time to qualify for re-treatment in the AChR-Ab+ population (up to 168 days)

• the percentage of early MG-ADL responders after the first treatment cycle in the AChR-Ab+ population.

After a 2-week screening period, patients were randomized in a 1:1 ratio to receive 4 IV infusions of efgartigimod alfa 10 mg/kg or a matching placebo administered at weekly intervals in addition to their ongoing treatment for gMG. Randomization was based on a permuted block strategy with interactive response technology, and stratification was based on Japanese or non-Japanese status, AChR-Ab+ or AChR-Ab– status, and use of NSISTs or non-NSISTs as concomitant therapy for gMG. The patients, investigators, study staff, and sponsor were blinded to patient treatment assignment. Unblinding of treatment was permitted in the event of a medical emergency, in which case the patient was discontinued from the study. The total study duration was up to 28 weeks divided into the 2-week screening period and a 26-week treatment period. Patients started the 26-week treatment period with an initial 8-week treatment cycle that comprised a 3-week treatment period (4 infusions on weeks 0 to 3) and a 5-week follow-up period. At the end of the first treatment cycle, patients started an intertreatment cycle of variable length depending on the patient’s clinical response to efgartigimod alfa or placebo. If a patient was an MG-ADL responder during a previous treatment cycle and lost response, that patient could qualify for re-treatment. Loss of response was defined as a less than 2-point reduction in the MG-ADL total score during the cycle compared to the baseline value for that cycle). Re-treatment in subsequent cycles was permitted if patients met all the following criteria:

• completed the prior treatment cycle (i.e., the 3-week treatment period and 5-week follow-up period).

• had an MG-ADL total score greater than or equal to 5 points with more than 50% of the total score attributed to non-ocular symptoms.

• the subsequent cycle did not start after day 127 and could be completed within the 26-week treatment period.

Based on these criteria, a maximum of 3 treatment cycles was possible during the 26-week study; patients who completed the study were eligible to enrol in a long-term extension study, ADAPT+ (refer to Section 3).

Table 5: Details of the ADAPT Trial

Ab+ = antibody positive; Ab– = antibody negative; AChEI = acetylcholinesterase inhibitor; AChR = acetylcholine receptor; C1 = cycle 1; C1B = baseline of cycle 1; CMI = clinically meaningful improvement; CnB = baseline of cycle n; gMG = generalized myasthenia gravis; IgG = immunoglobulin G; IMP = investigational medicinal product; IVIg = intravenous immunoglobulin; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA = Myasthenia Gravis Foundation of America; MG-QoL15r = Revised Myasthenia Gravis Quality of Life 15-item; NSIST = nonsteroidal immunosuppressive therapy; PE = plasma exchange; q.7.d. = every 7 days; QMG = Quantitative Myasthenia Gravis; SEB = study entry baseline; VAS = visual analogue scale.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Sources: ADAPT Clinical Study Report (ADAPT 2020),³⁰ and the sponsor’s submission.¹

Figure 2: ADAPT Study Design

D = study day; EoS = end of study; gMG = generalized myasthenia gravis; ITC₁V1 = intertreatment cycle 1 visit 1; ITCnVn = intertreatment cycle n visit n; Prim. EP = primary end point; SEB = study entry baseline; TC = treatment cycle; TCB = treatment cycle baseline; V = visit.

Note: A study cycle is defined as a treatment cycle and an intertreatment cycle.

^a The cycle sequence could have been repeated as many times as necessary as long as the last treatment cycle did not begin after day 127.

^b The intertreatment cycle period varied from patient to patient.

^c Time windows for clinic visits were ± 1 day for treatment cycle visits and ± 2 days for intertreatment cycle visits.

Source: ADAPT Clinical Study Report.³⁰

Populations

Inclusion and Exclusion Criteria

The eligibility criteria for the ADAPT trial are shown in Table 5. Patients were eligible for inclusion in the ADAPT study if they were adults (aged ≥ 18 years), had a diagnosis of Class II to IV gMG based on the MGFA system, had any serotype of gMG (AChR-Ab+, AChR-Ab–, MuSK, or LRP4), and had an MG-ADL score of greater than or equal to 5 (with > 50% of the score caused by non-ocular symptoms) despite being on a stable dose of concomitant conventional therapy before screening (AChEIs, steroids, and/or NSISTs), which was continued throughout the trial. There was no requirement to have received or discontinued the use of any specific treatment. A maximum of 20% of patients with AChR-Ab–were to be enrolled. Patients were required to be on a stable dose of standard of care before screening. For AChEIs, this meant no dose change for 2 weeks before screening; for steroids, 3 months or longer of treatment and no dose change for 1 month before screening; and for NSISTs, 6 months or longer of treatment and no dose change for 3 months before screening. The standard of care was allowed to be used alone or in combination. Patients were excluded if they received treatment with IVIg or PE within 1 month of screening, treatment with rituximab or eculizumab in the 6 months before screening or underwent thymectomy in the 3 months before screening. Patients were also excluded if they had MGFA Class I or V gMG, worsening muscle weakness secondary to concurrent infection/medication, seropositivity or tested positive for active viral infection (e.g., hepatitis B, hepatitis C, HIV), a known severe infection or major episode in last 8 weeks, a history of non-MG autoimmune disease that would interfere with an accurate assessment of clinical symptoms, and documented lack of clinical response to PE.

Interventions

Efgartigimod alfa IV 10 mg/kg or matching placebo was administered as a 1-hour infusion every 7 days for 4 infusions (days 1, 8, 15, and 22) in each cycle. The maximum permitted efgartigimod alfa dose per infusion was 1,200 mg. Patients were to maintain their stable concomitant conventional therapy during treatment with efgartigimod alfa or placebo. Efgartigimod alfa and the matched placebo were identical in physical appearance and supplied in identical containers. If a significant change (> 10%) in body weight was observed, the dose was recalculated to ensure that efgartigimod alfa dose was 10 mg/kg. When delayed for more than 3 days, a dose was not administered to ensure that 2 consecutive doses were administered 3 or more days apart. Treatment adherence per cycle was defined according to the following formula: (number of doses received ÷ 4) × 100%.

Subsequent treatment cycles were initiated as needed to permit personalized treatment based on the patient’s clinical response, which was measured by the validated MG-ADL scale. Loss of response was defined as a less than 2-point reduction in the MG-ADL total score in a specific cycle compared to the baseline for that cycle. After treatment cycle 1, the frequency of re-treatment (i.e., initiation of subsequent cycles) was based on clinical response as measured using the MG-ADL. To be re-treated, a patient must have met all of the following re-treatment criteria: completed the prior treatment cycle (i.e., the 3-week treatment period and the 5-week follow-up period), have an MG-ADL total score greater than or equal to 5 points with more than 50% of the total score attributed to non-ocular symptoms, and the subsequent cycle must have started no later than day 127 and have been completed within the 26-week study duration. If the patient was an MG-ADL responder in a previous cycle and lost the response, then the patient could qualify for re-treatment if the re-treatment criteria were met. Loss of response is defined as a less than 2-point reduction in the MG-ADL total score in a particular cycle compared to that cycle’s baseline.

Rescue therapy was permitted for patients with protocol-defined MG clinical deterioration (new or worsening of respiratory and/or bulbar symptoms or a ≥ 2-point increase in individual non-ocular items on the MG-ADL) and whose health the investigator considered to be in jeopardy if rescue therapy was not provided. Patients who were provided rescue therapy were discontinued from study treatment but were still followed according to the assessment schedule. Rescue therapy options included PE, IVIg, immunoadsorption, any new type of corticosteroid, or any increased dose of current corticosteroid used as a standalone therapy or combined with another treatment.

Prohibited medications during the study, which would result in study discontinuation, included any other IgG therapy; change in the type, dose, or regimen of concomitant treatment (replacing, adding, and/or removing treatment or adjusting the dose and/or frequency), even if used for non-gMG indications; any monoclonal antibody for immunomodulation; and vaccines.

Outcomes

A list of efficacy end points assessed in this Clinical Review Report is provided in Table 6, followed by descriptions of the outcome measures. Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence as well as any outcomes identified as important to this review according to the clinical experts consulted by CADTH and stakeholder input from patient and clinician groups and public drug plans. Using the same considerations, the CADTH review team selected end points that were considered to be most relevant to inform CADTH’s expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. All summarized efficacy end points were assessed using GRADE. Select notable harms outcomes considered important for informing CADTH’s expert committee deliberations were also assessed using GRADE. For the purpose of this review, only information on the patients with AChR-Ab+ is presented in this report.

Table 6: Outcomes Summarized From the Studies Included in the Systematic Review

AChR-Ab+ = acetylcholine receptor antibody positive; AESI = adverse event of special interest; CMI = clinically meaningful improvement; EQ-5D-5L = 5-Level EQ-5D; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MG-QoL15r = Revised Myasthenia Gravis Quality of Life 15-item; QMG = Quantitative Myasthenia Gravis.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Sources: ADAPT Clinical Study Report;³⁰ Howard et al. (2021).²⁸

The MG-ADL is an 8-item patient-reported outcome tool used to assess MG symptoms and their effects on daily activities. The scale includes 2 items on daily life activities (ability to brush teeth or comb hair and limitations in the ability to rise from a chair) and 6 MG symptom categories (diplopia, chewing, voice and speech, ptosis, swallowing, and respiratory). Each item was graded on a 4-point symptom severity scale (0 = normal to 3 = most severe), with the total score from 0 to 24; higher scores indicate worse symptoms (Table 7). Scoring of the MG-ADL was performed by a trained and certified evaluator. A CMI on the MG-ADL has been defined as a 2-point reduction in total score,^52,53 which was used to define CMIs for responders and nonresponders. Minimal symptom expression is a score of 0 or 1.⁵² End points in the ADAPT trial related to the MG-ADL that were relevant to this review included:

• The primary efficacy end point: Percentage of patients with AChR-Ab+ who were MG-ADL responders in the first treatment cycle (an MG-ADL responder was defined as a patient with a ≥ 2-point improvement [reduction] in MG-ADL score, sustained for ≥ 4 consecutive weeks, with the first improvement occurring by week 4 of the cycle [≤ 1 week after the fourth infusion]).

• Secondary end point: Percentage of MG-ADL responders in cycle 1 in the overall population (i.e., AChR-Ab+ and AChR-Ab–).

• Secondary end point: Percentage of time patients with AChR-Ab+ showed a CMI in their MG-ADL score (≥ 2-point reduction) up to day 126.

• Secondary end point: Time from week 4 to qualify for re-treatment in the AChR-Ab+ population (i.e., a < 2-point reduction in the MG-ADL total score and MG-ADL total score of ≥ 5 points with > 50% of the total score attributed to non-ocular symptoms).

• Secondary end point: Percentage of early MG-ADL responders in cycle 1 in the AChR-Ab+ population (MG-ADL responders with first MG-ADL improvement of ≥ 2 points occurring by week 2).

• Various tertiary and exploratory end points related to characterizing the time to onset of effect and magnitude of effect in MG-ADL scores, change over time (total score and individual items), duration and magnitude of MG-ADL response, and repeatability of the effect from the second cycle onward.

The QMG quantifies disease severity based on impairments of body functions and structures, as defined by the International Classification of Functioning, Disability and Health.⁵⁴ The QMG includes 13 investigator-rated items that measure endurance or fatigability, accounting for disease state fluctuations. The QMG score is composed of the following items: ocular (2 items), facial (1 item), bulbar (2 items), gross motor (6 items), axial (1 item), and respiratory (1 item).⁵⁵ A trained evaluator performed scoring on the QMG scale. The items were scored on a 4-point severity scale (0 = no symptoms to 3 = severe symptoms), and permissible scores range from 0 to 39, with higher scores indicative of greater disease severity (Table 7). A CMI on the QMG has been defined as a 3-point reduction in the QMG score.⁵⁶ End points in the ADAPT trial related to the QMG scale that were relevant to this review included:

• Secondary end point, the percentage of QMG responders (defined as a ≥ 3-point improvement in the total QMG score for ≥ 4 consecutive weeks with the first improvement occurring by week 4 of cycle 1) in the AChR-Ab+ population.

The MG-QoL15r is a 15-item survey of a patient’s perceived HRQoL that addresses attributes known to be meaningful to a patient with gMG, such as psychological well-being and social functioning. The MG-QoL15r helps inform the clinician about the patient’s perception of the extent of and dissatisfaction with MG-related dysfunction. The patient uses a 3-point (0 = not at all, to 2 = very much) Likert scale to assess statements regarding the domains of mobility (9 items), symptoms (3 items), general contentment (1 item), and emotional well-being (2 items). The maximum possible score is 30 points, and higher scores reflect greater MG-related dysfunction (Table 7). An MID or CMI on the MG-QoL15r has not been determined.⁵⁷ In the ADAPT trial, the only tertiary end point involving the MG-QoL15r was the change from baseline in the total MG-QoL15r score.

The EQ-5D-5L is a standardized instrument developed by the EuroQol Group to provide a simple, generic measure of health status for clinical and economic appraisal.⁵⁸ A VAS is included in the questionnaire, and respondents were asked to mark their health status from 0 to 100 on the day the interview was conducted, with a score of 0 corresponding to “the worst health you can imagine” and of 100 corresponding to “the best health you can imagine.” The sponsor is not aware of a reported MID or CMI value for EQ-5D VAS in patients with gMG.

gMG exacerbation was defined as a greater than or equal to 3-point worsening in QMG score versus baseline. In the overall population, a post hoc analysis was also conducted to estimate the proportions of patients with exacerbation (a ≥ 3-point worsening in QMG score versus baseline), which was compared between the treatment arms using the chi-square test.⁵⁹

Safety Outcomes

Safety was assessed through the incidence of TEAEs and changes in clinical laboratory values, vital signs, and electrocardiogram results.²⁸ The TEAEs were coded using MedDRA central coding dictionary and were graded using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0. Events within the MedDRA system organ class “infections and infestations” were defined as AESIs. All AEs were assessed, documented, and reported in accordance with International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) Guideline for Good Clinical Practice. The safety and disease severity follow-up visits continued until day 182.

Table 7: Summary of Outcome Measures and Their Measurement Properties in the ADAPT Trial

CI = confidence interval; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MG-QoL15 = Myasthenia Gravis Quality of Life 15-item; MG-QoL15r = Revised Myasthenia Gravis Quality of Life 15-item; MGFA = Myasthenia Gravis Foundation of America; MID = minimally important difference; MMT = Manual Muscle Test; QMG = Quantitative Myasthenia Gravis; SD = standard deviation; VAS = visual analogue scale.

Statistical Analysis

Clinical Trial End Points

The data cut-off date (last patient completed the study) was April 6, 2020. A summary of the statistical analyses of efficacy end points is presented in Table 8. The total MG-ADL, QMG, and MG-QoL15r changes from the baseline of cycle 1 between treatment group differences were analyzed using mixed models for repeated measures. The model included available data up to week 8 on treatment, visit and treatment-by-visit interaction terms as fixed effects, with baseline value and stratification factors as covariates. The within-subject correlation was modelled by assuming an unstructured covariance matrix for the error terms.

Sample Size and Power Calculation

The proportion of MG-ADL responders in the placebo group was hypothesized to be 30%. The treatment difference was assumed to be 29% in favour of the efgartigimod alfa group, representing a weighted average treatment difference of 35% for patients with AChR-Ab+ (80% of enrolled patients) and 5% for patients with AChR-Ab– (20% of enrolled patients). A 10% dropout rate was assumed.

Based on the assumptions above, a sample size of 150 patients would provide 96% power to detect a treatment difference of 35% (2-sided significance level of 5%) in the rate of responders among 120 patients in the primary population (patients with AChR-Ab+).²⁸

Statistical Testing

The statistical analysis methods are presented in Table 8. The results summarized in this submission are from the final analyses of the ADAPT trial.

The primary end point was tested using a 2-sided exact test using a logistic regression model with baseline MG-ADL total score as a covariate and the following 3 stratification factors: AChR-Ab status (seropositive versus seronegative); NSISTs (taking versus not taking); and Japanese nationality (yes versus no). The treatment effect was presented as an OR with a 95% CI and 2-sided P value (5% 2-sided alpha level).

If the primary end point met significance, the 5 secondary end points were tested in a strict hierarchical order (refer to Table 15) to control the type I error rate using a 5% 2-sided significance level for each analysis. If an end point did not meet significance at the 5% significance level, the subsequent end points in the hierarchical order were not evaluated.

The results of tertiary and exploratory end points were summarized descriptively.

A similar logistic regression model was used to analyze parameters related to the MG-ADL and QMG scales as for the primary efficacy end point.

The percentage of time that patients had a CMI in their MG-ADL total score was analyzed using an analysis of covariance (ANCOVA) model with treatment (as randomized) and baseline total score as covariates; the model was stratified for the stratification variables (race and concomitant gMG treatment).

The time to qualify for re-treatment, as monitored by the MG-ADL total score, was analyzed using the Kaplan-Meier time-to-event analysis (stratified log-rank test), stratified for the stratification variables.

Analysis Populations

The analysis sets are presented in Table 9. The ITT analysis set included all patients who were randomized. Efficacy analyses were performed in the modified intention-to-treat (mITT) population, which included all randomized patients who had a valid baseline MG-ADL assessment and at least 1 postbaseline MG-ADL assessment. The per-protocol analysis set was a subset of the mITT population that included patients with 3 or more of the 4 infusions (in any order) without any major protocol deviations. The per-protocol analysis set was used for sensitivity analyses of the primary and secondary end points. In both the mITT and per-protocol analysis sets, patients were analyzed in the group to which they were randomized. The AChR-Ab+ population was defined based on the stratification factor as randomized. The safety analysis set included all patients who received at least a partial dose of investigative product. Patients were analyzed based on the treatment received.

Data Imputation Methods

In general, the main reasons for missing data (i.e., due to missing visit or missing value(s) on specific questions) on an efficacy scale were classified as:

• Missing data due to MG disease worsening or missing data due to an AE

• Missing data not due to MG disease worsening (e.g., vacation) or an AE

• Missing data in an analysis window as visit was performed out of window

• Missing data due to a missing value(s) on an efficacy scale (due to technical reasons, patient not able to perform the test with reason not linked to MG disease status, missing answer to a question, and so on).

Reason a resulted in data missing not at random, whereas reasons b, c, and d can reasonably be considered as data missing at random. The following rules were therefore intended for the handling of missing visit(s) or missing value(s):

• A patient who dropped out or was lost-to-follow-up was treated as a nonresponder if they had not qualified as an MG-ADL responders before.

• Intermittent missing data for only 1 of 4 postonset (O) consecutive analysis windows:

  • If the missing data followed 1 of the following score patterns, then the patient was considered as having achieved an MG-ADL response if that the missing data “m” was not due to reason a):

    • OXmXX

    • OmXXX

    • OXXXmX

    • OXXmX

  • If no such patterns were present, then the patient was considered as not having achieved an MG-ADL response.

• Intermittent missing data following the onset of response at greater than or equal to 2 of 4 consecutive postonset analysis windows. Regardless of the reason for missing data, the patient was considered as not having a sustained response.

• Using a missing-is-failure imputation method as a sensitivity analysis to assess the imputation impact for missing values.

Subgroup Analyses

For MG-ADL and QMG response, the following subgroups were prespecified (but not adjusted for multiplicity), with response rates and differences in response rate (efgartigimod alfa – placebo) shown together with the 95% Wald confidence limits:

• age category at baseline (18 to < 65 years versus ≥ 65 years)

• sex (male versus female)

• race (Black or African American versus Asian versus white)

• region (US versus Japan versus rest of the world)

• baseline MG-ADL score (5 to 7 versus 8 to 9 versus ≥ 10)

• number of cycles (1 versus 2 versus 3).

Additional post hoc analyses were conducted to evaluate MG-ADL and QMG response rates in additional clinically meaningful subgroups:

• duration of disease (< 3 years versus 3 to < 6 years versus ≥ 6 years)

• concomitant therapies (AChEI only versus any steroid versus any NSIST)

• prior thymectomy (yes versus no)

• prior NSIST use (≥ 1 versus none)

• with and without prior treatment failure (prior exposure to ≥ 2 immunosuppressive therapies or treatment with ≥ 1 immunosuppressive therapy and requiring IVIg or PE multiple times within 1 year before study inclusion).

Sensitivity Analyses

Sensitivity analyses were conducted for the primary and secondary end points on the per-protocol analysis set. A second sensitivity analysis was conducted for the primary end point on the mITT population by using missing-is-failure imputation.

Table 8: Statistical Analysis of Efficacy End Points in the ADAPT Trial

Ab = antibody; Ab+ = antibody positive; Ab– = antibody negative; AChR = acetylcholine receptor; AE = adverse event; ANCOVA = analysis of covariance; C1B = baseline of cycle 1; CMI = clinically meaningful improvement; IMP = investigational medicinal product; m = missing data; MG-ADL = Myasthenia Gravis Activities of Daily Living; mITT = modified intention to treat; NSIST = nonsteroidal immunosuppressive therapy; QMG = Quantitative Myasthenia Gravis; vs. = versus.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Sources: ADAPT Clinical Study Report;³⁰ Howard et al. (2021).²⁸

Table 9: Analysis Populations in the ADAPT Trial

IMP = investigational medicinal product; ITT = intention to treat; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; mITT = modified intention to treat.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Sources: ADAPT Clinical Study Report;³⁰ Howard et al. (2021).²⁸

Results

Patient Disposition

Patient disposition in the AChR-Ab+ population is summarized in Table 10.

Of the 216 patients screened for inclusion, 129 patients were enrolled and randomized to the efgartigimod alfa group (65 patients) or the placebo group (64 patients), with 117 patients (90.7%) completing treatment and 121 patients (93.8%) completing the study. The primary reason for discontinuation from treatment was the occurrence of a TEAE; 2 patients (3.1%) in both efgartigimod alfa group and the placebo group discontinued treatment for this reason. Administration of rescue therapy resulted in protocol-mandated treatment discontinuation for 1 patient (1.5%) in the efgartigimod alfa group and 2 patients (3.1%) in the placebo group.

Table 10: Summary of Patient Disposition From the ADAPT Study Included in the Systematic Review (AChR-Ab+ Population, Safety Analysis Set)

AChR-Ab+ = acetylcholine receptor antibody positive; mITT = modified intention to treat.

^aData are only available for the full population as randomization happens after screening; available in Figure 1 of the Howard et al. (2021) pivotal clinical trial publication provided to CADTH as part of the original submission.

Sources: Sponsor’s submission;¹ sponsor-provided additional information.³⁶

Baseline Characteristics

In the AChR-Ab+ population, 129 patients were enrolled and randomized (65 patients in the efgartigimod alfa group and 64 patients in the placebo group). The mean age was 44.7 years (standard deviation [SD] = 14.9 years) and 49.2 years (SD = 15.5 years) in the efgartigimod alfa and placebo groups, respectively. Most patients were female (66.7%), and white (85.3%). Demographic characteristics were similar in both the efgartigimod alfa and placebo groups (Table 11). The baseline clinical characteristics were generally similar in both the efgartigimod alfa and placebo groups, although some baseline imbalances were noted (Table 12). There were baseline imbalances between the efgartigimod alfa and placebo groups by age category (≥ 65 years: 12.3% and 20.3%, respectively) and by sex (female: 70.8% and 62.5%, respectively), across the treatment groups.

The mean time since diagnosis was 9.31 years. The most frequently reported MGFA class at screening was Class III to IIIb (n = 37; 28.69%), which reflects a patient population with moderate weakness affecting muscles other than the ocular muscle and ocular muscle weakness of any severity. Most patients had previously undergone thymectomy for MG (n = 75; 58.14%), with mean time since that procedure 10.93 years. The proportion of patients who underwent thymectomy was higher in the efgartigimod alfa group than in the placebo group (69.2% versus 46.9%). The mean total MG-ADL score at baseline was 8.8 points (range, 5 to 16), and 54 patients (42%) had scores of 8 to 9 points, which reflects a moderate symptom burden. A total of 7 (11%) patients in the efgartigimod alfa group and 4 (6.0%) in the placebo group had MG-ADL total scores of 5, and 24 (37%) patients in the efgartigimod alfa group and 17 (27%) in the placebo group had MG-ADL total scores greater than or equal to 10.

Similar results were observed for the mean QMG score at baseline, which was 15.6 points in the overall population, which reflects a overall moderate severity of patients with gMG (based on MGFA QMG severity classification).⁶⁵ Patients also had impaired HRQoL based on median baseline MG-QoL15r scores of 16 to 17 points (out of a possible 30, higher scores indicate more severe impairment) across the efgartigimod alfa and placebo groups.

All patients had previously received 1 or more classes of conventional therapy for gMG, and large proportions had previously received 2 or more (96.1%) or 3 or more classes of conventional therapy (79.1%); no notable differences were observed between treatment arms. Based on the disease characteristics at baseline, these patients had persistent symptoms and impaired HRQoL despite previously receiving up to 3 classes of conventional therapy (AChEIs, corticosteroids, and/or NSISTs). Of note, despite their disease burden, these patients were still receiving stable doses of conventional therapies at baseline (i.e., they did not discontinue or progress beyond conventional therapy). These concomitant conventional therapies are summarized in the Exposure to Study Treatments section.

The baseline characteristics outlined in Table 11 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 11: Summary of Baseline Demographic Characteristics From the ADAPT Study (Safety Analysis Set)

AChR = acetylcholine receptor antibody positive; SD = standard deviation.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Source: ADAPT Clinical Study Report.³⁰

Table 12: Summary of Baseline Clinical Characteristics From the ADAPT Study (Safety Analysis Set)

Ab = antibody; Ab+ = antibody positive; AChR = acetylcholine receptor; AChEI = acetylcholinesterase inhibitor; gMG = generalized myasthenia gravis; IVIg = intravenous immunoglobulin; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MGFA = Myasthenia Gravis Foundation of America; MG-QoL15r = Revised Myasthenia Gravis Quality of Life 15-item; min = minimum; MuSK = muscle-specific kinase; NSIST = nonsteroidal immunosuppressive therapy; QMG = Quantitative Myasthenia Gravis; SD = standard deviation.

Notes: Scores range from 0 points to 24 points for the MG-ADL, 0 points to 39 points for the QMG, 0 points to 50 points for the MGC, and 0 points to 30 points for the MG-QoL15r; higher scores are indicative of worse quality of life.

Details included in the table are from the sponsor’s Summary of Clinical Evidence.

^aA post hoc analysis was conducted where “prior treatment failure” was defined as prior exposure to ≥ 2 immunosuppressive therapies or treatment with ≥ 1 immunosuppressive therapy and requiring PE or IVIg multiple times within 1 year before study inclusion.³²

Sources: ADAPT Clinical Study Report;³⁰ data on file.³²

Exposure to Study Treatments

Study Treatments

All patients received at least 1 dose of efgartigimod alfa or placebo during the study and the mean duration in the study (i.e., from the first dose until end of the study) was not reported in the sponsor’s evidence summary for the patients who were AChR-Ab+. The cumulative duration of treatment exposure was not reported for the patients who were AChR-Ab+ in the sponsor’s evidence summary either.

Patients in either group received a maximum of 3 cycles in the 26-week treatment period. Therefore, data regarding cycle duration after the first cycle would be cut-off at the end of the 26-week treatment period and thus not represent the true duration of those treatment cycles.

The number of patients per cycle was balanced between the efgartigimod alfa group (only 1 cycle: n = 14 [21.54%] patients; 2 cycles: n = 44 [67.7%] patients; 3 cycles: n = 7 [10.77%] patients) and the placebo group (only 1 cycle: n = 21 [32.82%] patients; 2 cycles: n = 42 [65.6%] patients; 3 cycles: n = 1 [1.59%] patients). An overview of the number of patients in each treatment group who received the treatment, and the cycle durations are summarized by cycle in Table 13.

Table 13: Summary of Patient Exposure From Studies Included in the Systematic Review

Ab+ = antibody seropositive; AChR = acetylcholine receptor; max = maximum; min = minimum; NC = not calculable; SD = standard deviation.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Source: ADAPT Clinical Study Report.³⁰

Concomitant Medications and Cointerventions

Concomitant gMG treatment was defined as any MG treatment received at a stable dosage during screening and maintained throughout the study (Table 14).

In the AChR-Ab+ population, the categories of concomitant gMG treatment were:

• Any AChEI: 57 patients (87.7%) in the efgartigimod alfa group and 57 patients (89.1%) in the placebo group.

• Any steroid: 46 patients (70.8%) in the efgartigimod alfa group and 51 patients (79.7%) in the placebo group.

• Any NSIST: 40 patients (61.5%) in the efgartigimod alfa group and 37 (57.8%) patients in the placebo group. Detailed information on the most common NSIST used for the AChR-Ab+ population was not presented in the sponsor’s evidence summary.

In the AChR-Ab+ subgroup, concomitant gMG treatment in the efgartigimod alfa and placebo groups often involved 2 classes of therapy (21.5% versus 34.4%, respectively) or 3 classes of therapy (49.2% versus 46.9%, respectively).

Table 14: Summary of Concomitant Medications in the ADAPT Study (Safety Analysis Set)

Ab+ = antibody positive; AChEI = acetylcholinesterase inhibitor; AChR = acetylcholine receptor; NSIST = nonsteroidal immunosuppressive therapy.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Source: ADAPT Clinical Study Report.³⁰

Rescue Medication Use

Treatment discontinuation due to rescue therapy was reported for 3 patients (2.3%) in the AChR-Ab+ population, 1 (1.5%) patient in the efgartigimod alfa group, and 2 (3.1%) patients in the placebo group.

Efficacy

Overall results from the primary and key secondary end points for AChR-Ab+ population are presented in Table 15.

Table 15: Summary of MG-ADL and QMG Efficacy Results in the AChR-Ab+ Population (mITT Population)

AChR-Ab+ = acetylcholine receptor antibody positive; CI = confidence interval; CMI = clinically meaningful improvement; gMG = generalized myasthenia gravis; LSM = least squares mean; MG-ADL = Myasthenia Gravis Activities of Daily Living; mITT = modified intention to treat; NR = not reported; OR = odds ratio; QMG = Quantitative Myasthenia Gravis; vs. = versus.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

^aThis outcome was not tested statistically due to earlier failure of the hierarchy.

^bThe hierarchy test failed at this end point.

Sources: ADAPT Clinical Study Report;^1,30 sponsor provided additional information.⁶⁶

Activities of Daily Living (MG-ADL)

Percentage of MG-ADL Responders During Cycle 1

Primary Analysis

Percentage of MG-ADL responders in the AChR-Ab+ population during cycle 1 was assessed as the primary outcome. In the AChR-Ab+ population, the MG-ADL responder criterion was met by 44 patients (67.7%) in the efgartigimod alfa group and 19 patients (29.7%) in the placebo group. The between-group difference was 38% (95% CI, 22.05% to 53.96%). The OR was 4.95 (95% CI, 2.21 to 11.53; P < 0.0001) (Table 15).

Subgroup Analysis

Various post hoc subgroup analyses were conducted to evaluate the primary end point in clinically relevant subgroups within the AChR-Ab+ population (Table 16).^1,33,67 Results among the subgroups analyzed were generally consistent with the main analysis.

Table 16: Subgroup Analyses of Cycle 1 MG-ADL Response in the AChR-Ab+ Population (Post Hoc Analyses, ADAPT)

Ab+ = antibody positive; AChEI = acetylcholinesterase inhibitor; AChR = acetylcholine receptor; CI = confidence interval; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; mITT = modified intention to treat; NSIST = nonsteroidal immunosuppressive therapy.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

^aPost hoc analyses of Argenx data on file (April 21, 2023).¹ Selected data presented by Karam et al. (2021).⁶⁷

^bAdapted from Bril et al. (2022)³³ and Sacca et al. (2023).³²

^cPrior treatment failure is defined as prior exposure to ≥ 2 immunosuppressive therapies or treatment with ≥ 1 immunosuppressive therapy and requiring plasma exchange or IV immunoglobulin multiple times within 1 year before study inclusion.

Sources: ADAPT Clinical Study Report;³⁰ sponsor’s submission.¹

Sensitivity Analysis

Per-protocol analysis was done as a sensitivity analysis of MG-ADL responders in the AChR-Ab+ population during cycle 1. The results of this per-protocol analysis were consistent with the results in the mITT set. In the AChR-Ab+ population, the MG-ADL responder criterion was met in || ||||||| patients in the efgartigimod alfa group compared to 18 (34.0%) patients in the placebo group (|| |||| |||||||| ||||||| |||||).³⁰

Percentage of MG-ADL Responders During Cycle 2

Percentage of MG-ADL responders during cycle 2 was assessed as exploratory outcome (Table 15 and Figure 2). In the AChR-Ab+ population, the proportion of MG-ADL responders in each group during cycle 2 appeared to be similar to that in cycle 1 in the AChR-Ab seropositive population. MG-ADL responders were reported in 36 of 51 patients (70.6%) in the efgartigimod alfa group and 11 of 43 patients (25.6%) in the placebo group in cycle 2. The between-group difference was 45.0% (95% CI, NR). The OR was 8.19 (95% CI, 2.88 to 25.73; P < 0.001).⁶⁶

Early MG-ADL Responders

Early MG-ADL responders (i.e., responded at week 2 of the cycle) in the AChR-Ab+ population was assessed as a key secondary outcome. Early MG-ADL response occurred in 37 patients (56.9%) in the efgartigimod alfa group and 16 patients (25.0%) in the placebo group. The between-group difference was 31.9% (95% CI, NR). The || |||| ||| was |||| ||||| || |||||| | ||||| | ||||||.⁶⁶

Percentage of Time with a CMI of MG-ADL Score

In the AChR-Ab+ population, the percentage of time with a CMI (reduction of ≥ 2 points versus study entry baseline) in the MG-ADL total score was assessed as a key secondary outcome. It was observed that the least squares mean (LSM) percentage of time with a CMI up to day 126 was 48.7% (standard error [SE] = 6.2) in the efgartigimod alfa group compared to 26.6% (SE = 6.3) in the placebo group. The between-group difference was 22.1% (95% CI, 10.9% to 33.2%; in favour of efgartigimod alfa; P = 0.0001) (Table 15).

Time to Qualify for Re-Treatment

Time to qualify for re-treatment in the AChR-Ab+ Population was assessed as a key secondary outcome. Since the week 4 visit, || ||||||| patients in the efgartigimod alfa group and || ||||||| patients in the placebo group qualified for re-treatment. The between-group difference was ||||| |||| ||| ||| ||||||||. The median time to qualify for re-treatment || |||| ||||||| ||||| |||| || |||| ||||| in the efgartigimod alfa group and 8 days (95% CI, 1.0 day to 30.0 days) in the placebo group, respectively (||| ||||| | ||||| | ||||||). The statistical testing hierarchy was broken at this end point (Table 15 and Table 17). The Kaplan-Meier plot for time to qualification for re-treatment in the AChR-Ab+ population is presented in Figure 2. The hazard ratio was not provided.¹

Table 17: Time to Qualify for Re-Treatment in the AChR-Ab+ Population (mITT Analysis Set)

AChR+ = acetylcholine receptor antibody positive; CI = confidence interval; mITT = modified intention to treat; NR = not reported; vs. = versus.

Sources: Sponsor’s submission;¹ and sponsor-provided additional information.³⁶

Figure 3: Time to Qualify for Re-Treatment in the AChR-Ab+ Population (mITT Analysis Set) [Redacted]

|||| || ||||||| ||| ||||||||||| || ||| |||||||| |||||||||| ||||| |||||||| |||||||| ||||| |||||| | |||||| |||| |||||||||| || |||||||| ||||||||| ||| ||||||||||| || || ||| ||||| || |||||| ||||| |||| | |||||| || ||||||| |||||||| || |||||||||||| |||| ||||| ||| || ||||||| ||||||||| ||| ||||||||||||||

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Sources: ADAPT Clinical Study Report;³⁰ sponsor’s submission.¹

MG-ADL Total Score Mean Change From Cycle Baseline During Cycle 1 and Cycle 2

The MG-ADL total score changes from baseline in cycle 1 and cycle 2 are presented in Table 18 and Figure 3. Figure 3 shows that the maximum improvement from study entry baseline in MG-ADL total score occurred at week 4 during cycle 1 and cycle 2 in the efgartigimod alfa group of the AChR-Ab+ population.

At week 4 of cycle 1, the LSM change from study entry baseline in the MG-ADL total score was –4.104 points (95% CI, –5.007 points to –3.201 points) in the efgartigimod alfa group and –1.269 (95% CI, –2.199 to –0.339) in the placebo group. The between-group difference in the LSM change from baseline at week 4 of cycle 1 was –2.84 (95% CI, –3.8 to –1.9; P < 0.0001) (Table 18). At the end of cycle 1, the LSM change from study entry baseline in the MG-ADL total score was –1.56 (95% CI, –2.52 to –0.60) in the efgartigimod alfa group and –1.19 (95% CI, –2.18 to –0.18) in the placebo group. The between-group difference in the LSM change from baseline at week 4 of cycle 1 was –0.37 (95% CI, –1.45 to 0.71; P = 0.4978) (Table 18).

Table 18: MG-ADL Total Score LSM Change From Cycle Baseline During Cycle 1 and Cycle 2 in the AChR-Ab+ Population (mITT Population)

AChR-Ab+ = acetylcholine receptor antibody positive; CI = confidence interval; LSM = least squares mean; MG-ADL = Myasthenia Gravis Activities of Daily Living; mITT = modified intention to treat.

^aP values were not adjusted for multiplicity.

^bThe treatment period is defined as a 26-week period divided into 8-week treatment cycles; therefore, the end of cycle n was defined as week 8

Sources: Sponsor’s submission¹ and the sponsor-provided additional information.⁶⁶

Figure 4: MG-ADL Total Score Mean Change From Cycle Baseline During Cycle 1 and Cycle 2 in the AChR-Ab+ Population (mITT Analysis Set)

Ab = antibody; AChR = acetylcholine receptor; MG-ADL = Myasthenia Gravis Activities of Daily Living; mITT = modified intention to treat; SE = standard error.

Notes: The MG-ADL was scored from 0 to 24, with higher scores indicating more severe disease.

The red dotted line represents changes over time in the efgartigimod alfa group; the blue solid line represents changes over time in the placebo group.

Sources: ADAPT Clinical Study Report;³⁰ sponsor’s submission.¹

Figure 5: MG-ADL Total Score Mean Change From Cycle Baseline During Cycle 1 and Cycle 2 in the Overall Population (mITT Analysis Set) [Redacted]

Ab = antibody; AChR = acetylcholine receptor; MG-ADL = Myasthenia Gravis Activities of Daily Living; mITT = modified intention to treat; SE = standard error.

Notes: The MG-ADL was scored from 0 to 24, with higher scores indicating more severe disease.

The red curve represents changes over time in the efgartigimod alfa group; the blue represents changes over time in the placebo group.

Sources: ADAPT Clinical Study Report;³⁰ sponsor’s submission.¹

Disease Severity (Assessed With QMG)

Quantitative Myasthenia Gravis

The percentage of QMG responders during cycle 1 within the AChR-Ab+ population was assessed as the first key secondary outcome. The QMG responder criterion was met by 41 patients (63.1%) in the efgartigimod alfa group and 9 patients (14.1%) in the placebo group. The between-group difference was 49% (95% CI, 34.5% to 63.5%). The OR was 10.84 (95% CI, 4.18 to 31.20; P < 0.0001) (Table 15, Figure 4).

Figure 6: Percentage of QMG Responders During Cycle 1 in the AChR-Ab+ Population (mITT Analysis Set)

AChR-Ab+ = acetylcholine receptor antibody positive; CI = confidence interval; mITT = modified intention to treat; QMG = Quantitative Myasthenia Gravis.

Sources: ADAPT Clinical Study Report³⁰ and the sponsor’s submission.¹

Figure 7: Percentage of QMG Responders During [Redacted] in the AChR-Ab+ Population (mITT Analysis Set) [Redacted]

AChR-Ab+ = acetylcholine receptor antibody positive; CI = confidence interval; mITT = modified intention to treat; QMG = Quantitative Myasthenia Gravis.

Sources: ADAPT Clinical Study Report³⁰ and the sponsor’s submission.¹

Health-Related Quality of Life

Revised Myasthenia Gravis Quality of Life 15-Item Scale

Mean Change From Baseline During Cycle 1 and Cycle 2 (mITT Analysis Set)

MG-QoL15r mean change from baseline during cycle 1 and cycle 2 was assessed as an exploratory outcome (Table 19).

The LSM change in MG-QoL15r score from study baseline to week 4 of cycle 1 was –7.21 (95% CI, –8.79 to –5.63) in the efgartigimod alfa group and –1.76 (95% CI, –3.38 to –0.14) in the placebo group. The between-group difference in LSM change from study baseline at week 4 of cycle 1 was –5.45 (95% CI, –7.22 to –3.69; P < 0.0001). At the end of cycle 1, the LSM change from study baseline was –4.07 (95% CI, –5.69 to –2.45) in the efgartigimod alfa group and –1.70 (95% CI, –3.37 to –0.04) in the placebo group. The between-group difference in LSM change from study baseline at week 4 of cycle 1 was –2.37 (95% CI, –4.21 to –0.53; P = 0.0123).

At cycle 2, at week 4, the LSM change from study baseline was –5.09 (95% CI, –6.73 to –3.47) in the efgartigimod alfa group and 0.35 (95% CI, –1.44 to 2.14) in the placebo group. The between-group difference in LSM change from study baseline at week 4 of cycle 2 was –5.45 (95% CI, –7.28 to –3.62; P < 0.0001). At the end of cycle 2, the LSM change from study baseline was –1.71 (95% CI, –3.34 to –0.08) in the efgartigimod alfa group and 0.35 (95% CI, –1.41 to 2.11) in the placebo group. The between-group difference in LSM change from study baseline at the end of cycle 2 was –2.06 (95% CI, –3.85 to –0.28; P = 0.0239).

Table 19: Summary of MG-QoL15r and EQ-5D VAS Results from the ADAPT Trial (AChR-Ab+)

AChR-Ab+ = AChR antibody positive; CI = confidence interval; EFG = efgartigimod alfa group; LSM = least squares mean; MG-QoL15r = Revised Myasthenia Gravis Quality of Life 15-item; PBO = placebo group; VAS = visual analogue scale.

^aP values were not adjusted for multiplicity.

^bThe treatment period is defined as a 26-week period divided into 8-week treatment cycles; therefore, the end of cycle n was defined as week 8.

Sources: The sponsor’s submission¹ and the sponsor-provided additional information.³⁶

Figure 8: MG-QoL15r Least Squares Mean Change From Baseline During Cycle 1 and Cycle 2 in the AChR-Ab+ Population (mITT Analysis Set)

AChR-Ab+ = acetylcholine receptor antibody positive; LSM = least squares mean; LSMD = least squares mean difference; MG-QoL15r = Revised Myasthenia Gravis Quality of Life 15-item; mITT = modified intention to treat; SE = standard error.

Note: The MG-QoL15r was scored from 0 to 30, with higher scores indicating more impaired quality of life and decreases indicating improved quality of life.

* Indicates treatment administration (efgartigimod alfa or placebo).

Source: Sacca F et al. J Neurol. 2023;270:2096-2105. Reprinted with accordance with Creative Commons Attribution 4.0 International License (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/⁵⁰

EQ-5D

EQ-5D VAS Mean Change From Baseline

EQ-5D VAS mean change from baseline during cycle 1 and cycle 2 was assessed as an exploratory outcome (Table 19).

At week 4 of cycle 1, the LSM change from baseline was 16.04 (95% CI, 11.37 to 20.71) in the efgartigimod alfa group and 2.76 (95% CI, –2.05 to 7.56) in the placebo group. The between-group difference in LSM change from baseline at week 4 of cycle 1 was 13.28 (95% CI, 8.32 to 18.24; P < 0.0001). At the end of cycle 1, the LSM change from baseline was 4.61 (95% CI, –0.01 to 9.24) in the efgartigimod alfa group and 3.92 (95% CI, –0.09 to 8.69) in the placebo group. The between-group difference in LSM change from baseline to the end of cycle 1 was 0.69 (95% CI, –4.19 to 5.59; P = 0.7784).

At week 4 of cycle 2, the LSM change from baseline was 14.16 (95% CI, 9.69 to 18.63) in the efgartigimod alfa group and 1.92 (95% CI, –3.03 to 6.87) in the placebo group. The between-group difference in LSM change from baseline at week 4 of cycle 2 was 12.24 (95% CI, 7.33 to 17.16; P < 0.0001). At the end of cycle 2, the LSM change from baseline was 0.16 (95% CI, –4.53 to 4.86) in the efgartigimod alfa group and 0.27 (95% CI, –4.87 to 5.41) in the placebo group. The between-group difference in LSM change from baseline to the end of cycle 2 was –0.10 (95% CI, –5.40 to 5.19; P = 0.9693).

Figure 9: EQ-5D-5L VAS Mean Change From Baseline in the AChR-Ab+ Population (mITT Analysis Set)

Ab+ = antibody seropositive; AChR = acetylcholine receptor; EQ-5D-5L = 5-Level EQ-5D; LSM = least squares mean; LSMD = least squares mean difference; mITT = modified intention to treat; SE = standard error; VAS = visual analogue scale.

Note: The cycle 2 panel contains a transposition error, where the values shown in the black line with triangles start under week 0 instead of week 1.

* Indicates treatment administration (efgartigimod alfa or placebo).

Source: Sacca F et al. J Neurol. 2023;270:2096-2105. Reprinted with accordance with Creative Commons Attribution 4.0 International License (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/⁵⁰

Other Clinical Outcomes (Post Hoc Analysis)

MG Hospitalizations

Post hoc analysis of hospitalizations was conducted using data from the ITT population of ADAPT.⁵⁹ In that analysis, the numbers of all-cause and gMG-related hospitalizations were combined with follow-up times to calculate the incidence rates of hospitalizations, which were compared between the efgartigimod alfa and placebo arms for the AChR-Ab+ population. By week 26, no patients in the efgartigimod alfa group and 3 (4.69%) patients in placebo group had MG hospitalizations. The between-group difference (efgartigimod alfa – placebo) was –4.69% (95% CI, NR) (Table 20). The incidence of gMG-related hospitalizations was 0 per 100 patient-years and 10.98 per 100 patient-years in patients in the efgartigimod alfa and placebo groups, respectively.^36,59 The between-group difference was 10.98 per 100 patient-years (95% CI, –23.42 per 100 patient-years to 1.47 per 100 patient-years).

Table 20: Results of Other Clinical Outcomes in the ADAPT Trial (AChR-Ab+ Population, Week 26) (mITT)

AChR-Ab+ = acetylcholine receptor antibody positive; CI = confidence interval; EFG = efgartigimod alfa group; mITT = modified intention to treat; MG = myasthenia gravis; NR = not reported; PBO = placebo group.

Notes: MG hospitalization, MG exacerbation, and MG crisis were assessed as post hoc analyses. P values were not adjusted for multiplicity for type I error control.

Sources: Sponsor’s submission¹ and the sponsor-provided additional information.³⁶

MG Exacerbation

gMG exacerbation was defined as a greater than or equal to 3-point worsening in QMG score versus baseline. Among patients who were AChR-Ab+, gMG exacerbations were identified in 17 (26.1%) in the efgartigimod alfa group and 27 (44.3%) in the placebo group. The between-group absolute risk difference was –18.2% (95% CI, NR; nominal P = 0.033).^36,59

MG Crisis

At week 26, In patients with AChR-Ab+, use of rescue therapy was reported | |||| in the efgartigimod alfa group | ||||||| ||||||| in the placebo group. The between-group difference was |||||| ||| ||| ||| ||||||||. The incidence rate per 100 follow-up years, was ||| in the efgartigimod alfa group and |||| in the placebo group. The between-group incidence rate difference was |||||| |||||| |||||| || ||||| ||||||||.

Harms

Only those harms identified in the sponsor’s evidence summary review protocol are reported below.

AEs in the population of patients who were AChR-Ab+ are presented in Table 21.

Adverse Events

During the randomized controlled period, among patients in the AChR-Ab+ group, TEAEs were reported in || ||||||| patients in the efgartigimod alfa group and || ||||||| patients in the placebo group (Table 21). The most common TEAEs (occurring in ≥ 10% patients in any arm) were headache (EFG versus PBO, ||||| ||| |||||); nasopharyngitis (EFG versus PBO, ||||| ||||||||), upper respiratory tract infection |||||||| ||| ||||||, and diarrhea (EFG versus PBO | |||| ||| |||||).

Serious Adverse Events

During the randomized controlled period, treatment-emergent SAEs in patients in the AChR-Ab+ population were reported in ||| |||||| patients in the efgartigimod alfa group and | |||||| patients in the placebo group (Table 21).

Withdrawal of Treatment Due to Adverse Events

During the randomized controlled period, TEAEs that led to discontinuation of treatment among patients in the AChR-Ab+ population were reported for | |||||| patients in the efgartigimod alfa group and | |||||| patients in the placebo group. No specific TEAE that led to discontinuation of the investigational medicine product was reported for more than 1 patient in either treatment group.

Mortality

During the randomized controlled period, no death was reported in either group (Table 21).

Notable Harms (Adverse Events of Special Interest)

In the ADAPT trial, AESIs were defined as infections and infestations. During the randomized controlled period, as shown in Table 21, AESIs were reported for || ||||||| patients in the efgartigimod alfa group and || ||||||| patients in the placebo group. The most frequently reported treatment-emergent AESIs (≥ 3% of patients in either group) among patients in the AChR-Ab+ population were: nasopharyngitis (EFG versus PBO: ||||| ||||||||); upper respiratory tract infection |||||| ||| ||||); urinary tract infection ||||| ||| |||||| bronchitis (|||| ||| |||||; and influenza (|||| ||| ||||| (Table 21).

Table 21: Adverse Events in the AChR-Ab+ Population (Safety Analysis Set)

AChR-Ab+ = acetylcholine receptor antibody positive; AESI = adverse event of special interest; NR = not reported; SAE = serious adverse event; TEAE = treatment-emergent adverse event.

Sources: Sponsor’s submission;¹ sponsor-provided additional information.³⁶

Critical Appraisal

Internal Validity

Appropriate methods of randomization, blinding, and allocation concealment were reported. Outcomes were assessed using validated scales incorporating physician and patient assessments, and end points requiring a combination of CMI and sustained effect. However, minimally important between-group differences, that is, the thresholds used for the GRADE for all outcomes, were not available. Therefore, clinical expert opinion was used to inform the thresholds to determine whether the between-group difference observed for each outcome are clinically meaningful. Appropriate statistical methods were used. Multiplicity adjustments were made for the primary and 5 key secondary outcomes to control for family-wise type I error (probability of making more than 1 type I error); the hierarchy failed at time to re-treatment (result was not statistically significant). Other outcomes (e.g., tertiary, exploratory, and outcomes assessed in cycle 2) were not adjusted for multiplicity, so there is an increased risk of false-positive conclusions for statistically significant results.

Not all patients participated in the second cycle; therefore, there is a risk of selection bias. The extent and direction of the bias on the outcomes assessed during the second treatment cycle cannot be determined.

In addition, although the overall baseline demographic and clinical disease characteristics were mostly balanced, there was some notable imbalance between the efgartigimod alfa and placebo groups in AChR-Ab+ population. For example, more patients in the efgartigimod alfa group had an MG-ADL score greater than or equal to 10 than in placebo group (EFG versus PBO: ||||| ||| ||||| in AChR-Ab+ population), fewer patients in the efgartigimod alfa group had prior combination use of steroid plus AChEI than in placebo group (EFG versus PBO | |||| ||| ||||| in AChR-Ab+ population); and the proportion of patients who underwent thymectomy was higher in the efgartigimod alfa group than in the placebo group (EFG versus PBO: 69.2% versus 46.9% in AChR-Ab+ population). Furthermore, more patients in the efgartigimod alfa group used concomitant AChEIs than in placebo group (EFG versus PBO: ||||| ||| |||| in AChR-Ab+ population); fewer patients in the efgartigimod alfa group used concomitant steroid plus AChEI than that in placebo group (EFG versus PBO: ||||| ||| ||||| in in AChR-Ab+ population). Whether these baseline imbalances introduced bias is uncertain. The clinical experts consulted by CADTH for this review indicated that these observed imbalances were unlikely to have significantly affected the study results.

The treatment discontinuations were relatively infrequent in the efgartigimod alfa group compared with the placebo group (4.6% versus 14.1%, respectively). Protocol violations, prohibited drug use, and rescue therapy were all infrequent in both arms, all of which are unlikely to impact the study results. There was no indication from the efficacy or harms data that unblinding of patients or study personnel had occurred.

Furthermore, the changes from the study baseline to the end of the study (i.e., by week 26) were not assessed. The clinical experts CADTH consulted for this review indicated that the assessment at cycle 1 is acceptable. The results of other clinical outcomes (MG hospitalization, MG exacerbations, and MG crisis) were based on post hoc analyses. The study was not powered to assess the between-group difference of these outcomes. Therefore, the results should be interpreted with consideration of limitations of the statistics.

External Validity

Of the patients who were AChR-Ab+, 63% had experienced prior gMG treatment failure (also known as refractory gMG).³² Thirty-seven percent of patients were classified as having responded inadequately to the existing standard-of-care gMG therapy. According to the clinical experts CADTH consulted for his review, the population included in the ADAPT trial is adequately reflective of the population of patients in Canadian clinical settings who experience unmet needs in the treatment of gMG.

According to the clinical experts consulted by CADTH for this review, the demographic and disease characteristics and prior treatment histories of the patients enrolled in the ADAPT trial were reflective of the population of adult patients with gMG in Canada. The eligibility criteria for patients with MGFA Class II to IV gMG and an MG-ADL total score of 5 or more would select appropriately for patients with symptomatic gMG most in need of intervention. Patients with MGFA Class V gMG were excluded from the ADAPT trial, and very few patients with MGFA Class IV gMG were included (3.9% in the AChR-Ab+ population). Whether the findings of the ADAPT trial can be generalized to patients with MGFA Class IV or MGFA Class V is uncertain. The clinical experts consulted by CADTH for this review indicated that a subset of patients with MGFA Class I (ocular MG) or V and MG-ADL scores less than 5, who were excluded from the trial, would be suitable for treatment. Specifically, the clinical experts 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 efgartigimod alfa. However, the results of the trial cannot be directly generalized to these groups of patients.

For patients with gMG, 1 of the treatment goals is to reduce or discontinue the use of steroids. However, changes to concomitant medications, including steroids, were not allowed per the study protocol. Therefore, there is no information available about whether efgartigimod alfa would have an impact on steroid use among patients with gMG who are eligible for this treatment.

Furthermore, It is uncertain whether the findings derived from the ADAPT trial can be generalized to patients who are MuSK-Ab+. The clinical experts CADTH consulted for this review indicated that efgartigimod alfa will provide a new treatment for patients with gMG who are AChR-Ab+ and probably for those who are MuSK-Ab+. The clinician group input for this review also indicated that patients with MuSK antibodies might respond to efgartigimod alfa; however, whether these patients would benefit could not be confirmed based on ADAPT trial evidence.

All patients in the AChR-Ab+ population had previously received 1 or more prior therapies, with a large proportion of patients having received 2 or more prior therapies (96.1% in the AChR-Ab+ only populations) or 3 prior or more therapies (79.1% in the AChR-Ab+ only populations). In addition, during the trial, in patients who were AChR-Ab+, use of concomitant medications was 23.3% for 1 class of concomitant medications, 27.9% for 2 class of concomitant medications, and 48.1% for 3 class of concomitant medications. The clinical experts CADTH consulted for this review indicated that both the prior therapies and the concomitant medications used in the trial are aligned with clinical practice.

One of the inclusion criteria for recruiting patients into the ADAPT trial was an MG-ADL total score of 5 or more points. Of the patients who were AChR-Ab+, a total of | ||||||| patients in the efgartigimod alfa group, and | |||||| patients in the placebo group had an MG-ADL total score of 5 at the baseline of the study. The sponsor’s rationale for using the cut-off of 5 or more points was to align with the expected real-world usage of efgartigimod alfa and the unmet medical need for an additional treatment alternative to IVIg. The sponsor indicated that at the time of designing the trial, it was understood that patients who maintained an MG-ADL score of less than 5 would not need efgartigimod alfa. However, CADTH noticed that an MG-ADL total score cut-off of 6 or more points was an inclusion criterion in previous gMG clinical trials (i.e., in the REGAIN trial for eculizumab for gMG and in the CHAMPION trial for ravulizumab for gMG). Whether lowering the cut-off from 6 to 5 points for the MG-ADL total score potential has an impact on the patient’s response to efgartigimod alfa is uncertain. The clinical experts CADTH consulted for this review indicated that severity status based on the MG-ADL was subjective and a difference of 1 point may not be clinically important in terms of disease severity. In addition, the proportion of patients with an MG-ADL total score of 5 points was relatively low. Therefore, it is unlikely to have a meaningful impact on patients’ response to the treatment. Furthermore, the ADAPT trial was a placebo-controlled trial, so it does not provide any comparative efficacy or harms information on efgartigimod alfa relative to existing gMG therapies (e.g., AChEIs, steroids, NSIST, IVIg, PE, complement C5 inhibitors). Finally, the change from baseline of MG-ADL total score, MG-QoL15r score, and EQ-5D VAS score were assessed at cycle 1 and cycle 2. However, the overall efficacy of these outcomes at the end of the 26-week trial is unknown. The long-term comparative efficacy of efgartigimod alfa versus placebo still needs to be further established.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered most relevant to inform CADTH’s expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group:^34,35

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

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

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

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

Following the GRADE approach, evidence from RCTs is initially considered to be of high certainty and can be rated down if there are concerns related to study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias.

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. For this review, the target of the certainty of evidence assessment was based on the presence of absence of a clinically important effect, as informed by MIDs suggested by the sponsor and agreed upon by the clinical experts consulted by CADTH for this review (for change from baseline in MG-ADL score and QMG score), or by thresholds suggested by the clinical experts (for all other outcomes).

Results of GRADE Assessments

Table 2 presents the GRADE summary of findings for efgartigimod alfa versus placebo for the AChR-Ab+ population.

Long-Term Extension Studies

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

Description of Studies

The ADAPT+ study (ARGX-113-1705) is a long-term, single-arm, open-label, multicentre, phase III follow-on study of patients who enrolled in the ADAPT study (ARGX-113-1704; NCT03669588). The primary objective was to evaluate the long-term safety and tolerability of efgartigimod alfa in the AChR-Ab+ subgroup and the secondary objective was to evaluate safety and tolerability in the overall population (AChR-Ab+ and AChR-Ab–). Efficacy data were collected as exploratory end points. A final Clinical Study Report is available for ADAPT+ (data cut-off: June 30, 2022).⁶⁸

The ADAPT+ study was conducted at 51 sites, including 41 sites in 14 countries or regions that had 1 or more patients who could roll over from the ADAPT study. Data were collected over a 3-year period in 2 sequential parts (part A: 1 year, part B: 2 years maximum) (Figure 7). Part B was added as a protocol amendment to ensure accessibility to efgartigimod alfa until it became commercially available or available through an expanded access program. At the data cut-off (June 30, 2022), 151 patients had rolled over from the ADAPT trial, regardless of whether they were in the treatment or placebo group, into ADAPT+, and 145 patients had received at least 1 partial or complete dose of efgartigimod alfa in ADAPT+.

Figure 10: Study Design for the ADAPT+ Trial

D = day; EoA = end of part A; gMG = generalized myasthenia gravis; ITP_nV_n = visit n of intertreatment period n; SEB = study entry baseline; TPnB = baseline of treatment period n; TP_nVn = visit n of treatment period n.

Note: As with the ADAPT study, a study cycle was defined as a treatment period plus the subsequent intertreatment period.

Source: ADAPT+ final Clinical Study Report (data cut-off: June 30, 2022).⁶⁸

Populations

As in the ADAPT trial, patients were required to be receiving a stable dose of concomitant gMG treatment before ADAPT+ entry. Consenting patients who could comply with the study procedures were eligible to roll over into ADAPT+ if they had participated in the ADAPT trial and:

• reached the end of study at day 182 in the ADAPT trial

• needed (re-)treatment while in the ADAPT trial, but could not complete a treatment cycle within the ADAPT study period (these patients could immediately roll over to ADAPT+ and receive efgartigimod alfa)

• discontinued early from randomized treatment in the ADAPT trial for reasons other than pregnancy, rescue therapy, or an SAE

• had a temporary interruption from randomized treatment in the ADAPT trial (these patients could be offered the option to roll over into ADAPT+).

The exclusion criteria for ADAPT+ were generally consistent with those of the ADAPT trial. Patients were also excluded from rolling over to ADAPT+ if they had discontinued from randomized treatment or the ADAPT study because of pregnancy, need for rescue therapy, or an SAE.

Interventions

Efgartigimod alfa treatment was administered in ADAPT+ using the dosing calculation and schedule from the ADAPT trial: efgartigimod alfa 10 mg/kg administered as a 1-hour IV infusion (total volume: 125 mL) every 7 days for 4 infusions (days 1, 8, 15, and 22) in each cycle. The maximum efgartigimod alfa dose per infusion was 1,200 mg (for patients weighing ≥ 120 kg).

Consistent with the ADAPT study criteria, subsequent treatment and re-treatment cycles in part A of ADAPT+ were started when the following criteria were met:

• the patient had completed the previous cycle’s treatment period (i.e., after visit 4)

• the patient had an MG-ADL total score of greater than or equal to 5 points (> 50% of score due to non-ocular symptoms)

• the patient had a less than 2-point reduction in MG-ADL total score compared to:

  • the last cycle’s baseline in ADAPT for patients receiving their first treatment period in ADAPT+

  • the previous cycle’s baseline for patients in any cycle of ADAPT+.

Subsequent re-treatment cycles in Part B of ADAPT+ were started when the following criteria were met:

• the patient had completed the previous cycle’s treatment period (i.e., after visit 4)

• the investigator judged that efgartigimod alfa treatment was in the best interest of the patient

• 4 weeks or longer (1 calendar month) had elapsed since the last efgartigimod alfa infusion.

Prior therapy, concomitant therapy, and prohibited medications were defined as in the ADAPT trial. Use of prohibited medications would result in discontinuation of efgartigimod alfa treatment only in part A; part B did not restrict concomitant gMG treatment if it was based on standard clinical practice. Patients who required rescue therapy for protocol-defined MG clinical deterioration (as in the ADAPT trial) were discontinued from the study.

Outcomes

Efficacy outcomes in ADAPT+ were assessed using the MG-ADL tool (part A and part B) and the QMG tool (part A only) in the AChR-Ab+ population, in the AChR-Ab– population, and in the overall population (seropositive and seronegative). For the purpose of this review, only information on the patients with AChR-Ab+ are presented in this report. Efficacy outcome definitions, as well as the cut-offs for CMIs, were consistent between the ADAPT and ADAPT+ studies. A CMI on the MG-ADL has been defined as a 2-point reduction in the total score,^52,53 and minimal symptom expression is a score of 0 or 1.⁵² A CMI on the QMG has been defined as a 3-point reduction in the QMG score.⁵⁶

Due to the slightly different visit schedules in ADAPT+ (no visits were scheduled at weeks 4 to 6 of a cycle) and the ADAPT study (every week for the first 8 weeks of a cycle), ADAPT+ only collected MG-ADL data at week 3 of each study cycle.

Safety outcomes in ADAPT+ were assessed in the same manner as in the ADAPT study, with coding according to MedDRA (version 23.0) and severity grading according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0.

Statistical Analysis

All analyses were descriptive and performed using the safety analysis set from ADAPT+ (all patients who received ≥ 1 dose or partial dose), with subgrouping according to AChR-Ab status (Ab+ versus AB–) and the treatments received in the ADAPT and ADAPT+ trials (efgartigimod alfa in ADAPT and efgartigimod alfa in ADAPT+ [EFG-EFG] versus placebo in ADAPT and efgartigimod alfa in ADAPT+ [PBO-EFG]) (Table 22). Summary statistics were calculated for absolute values and changes from ADAPT+ baseline (overall) or changes from baseline of each cycle (i.e., the change for each cycle).

The rollover patients set, which included all patients who rolled over from ADAPT, was not used for any analyses.

Table 22: Patients in the Safety Analysis Set (ADAPT+)

Ab+ = antibody positive; Ab– = antibody negative; AChR = acetylcholine receptor antibody.

Notes: The rollover set includes all patients who rolled over from the ADAPT study, while the safety analysis set includes all patients who rolled over and received ≥ 1 dose or partial dose. AChR-Ab serostatus was determined based on the stratification factor at the time of randomization in the ADAPT study.

Source: ADAPT+ Clinical Study Report.⁶⁸

Patient Disposition

Baseline Characteristics

Demographic characteristics of patients who were AChR-Ab+ were enrolled in ADAPT+ are shown in Table 23, stratified by the treatments received in the ADAPT and ADAPT+ trials. Patient characteristics were generally similar across the reported groups. Mean age was 47.0 years (SD = 14.8 years), and most patients were white (86.9%) and female (71.0%).

Table 23: Baseline Demographic Characteristics in ADAPT+ (Safety Analysis Set)

AChR-Ab+ = acetylcholine receptor antibody positive; SD = standard deviation.

Source: ADAPT+ Clinical Study Report.⁶⁸

Baseline disease characteristics of patients with AChR-Ab+ who enrolled in ADAPT+, stratified by the treatments received in the ADAPT trial and ADAPT+, are shown in Table 24. Patient characteristics were generally similar across the reported groups. As in the ADAPT trial, patients in the AChR-Ab+ population who rolled over to ADAPT+ had a mean time since diagnosis of 9.7 years (SD = 7.9 years) with mean scores of 9.5 points (SD = 3.1 points) for MG-ADL total score and 15.3 points (SD = 5.7 points) for QMG total score.

Table 24: Baseline Disease Characteristics in ADAPT+ (Safety Analysis Set)

Ab = antibody; AChR-Ab+ = acetylcholine receptor antibody positive; gMG = generalized myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; MuSK = muscle-specific kinase; NSIST = nonsteroidal immunosuppressive therapy; QMG = Quantitative Myasthenia Gravis; SD = standard deviation.

Source: ADAPT+ Clinical Study Report.⁶⁸

Patient Disposition

No disposition information was reported for the seropositive population.

Exposure to Study Treatments

Study Treatments

Data regarding extent of exposure for patients who were AChR-Ab+ were not reported in the sponsor’s evidence summary.

Concomitant Medications and Cointerventions

Patients were to remain on their stable dose and regimen of concomitant gMG treatment (without replacing, adding, removing, or adjusting the dose and/or frequency of treatments) during part A of ADAPT+. During part B, changes in the type, dose, or regimen of the concomitant gMG treatment were permitted, as was the additional use of other treatments (e.g., PE, IVIg, immunoadsorption, any new corticosteroid type, or an increase in the dose of current corticosteroid).

As shown in Table 24, concomitant gMG treatments at baseline were classified as NSISTs (60.4% of patients in the AChR-Ab+ population) or non-NSISTs (39.6% of patients in the AChR-Ab+ population).

Subsequent Treatment

Results

Efficacy

Efficacy results were descriptively reported as exploratory outcomes based on validated clinical scales: the patient-reported MG-ADL (part A and part B) and the physician-reported QMG (only part A). Clinically meaningful mean improvements in symptoms were observed during each cycle based on both scales, with improvements that were repeatable across multiple cycles and occurred regardless of previous efgartigimod alfa treatment (i.e., in the both the group that received efgartigimod alfa in both ADAPT and ADAPT+ and the group that received placebo in ADAPT and efgartigimod alfa in ADAPT+). These clinically meaningful improvements (e.g., 2 or more points on MG-ADL and 3 or more points on QMG) were also observed in 88.7% of patients with 2 or more points on MG-ADL and 64.4% of patients with 3 or more points on QMG of the AChR-Ab+ subgroups. Furthermore, in AChR-Ab+ subgroups, the mean improvement in MG-ADL score was greater than or equal to 5 points across all cycles, which is greater than the suggested MID of 2 points.

Exploratory: MG-ADL Scores (Part A and Part B)

Data at all assessment points up to cycle 14 (due to limited patient data at later cycles) are shown graphically for the AChR-Ab+ population (Figure 8) These results show that, across all the groups, the largest mean MG-ADL response (reduction from baseline) appeared to be reached around week 3 of each cycle (data were not recorded at weeks 4 or 5 of ADAPT+, which is where the maximum response was observed in the ADAPT trial; there were also no data for week 6) and a return to approximately cycle baseline sometime during week 7 to week 11.

Within the AChR-Ab+ population, the mean change from the baseline of the first cycle to week 3 in the MG-ADL score (5.0 points; SE = 0.33 points) was greater than the suggested MID of 2 points. During the first 10 cycles of ADAPT+, an improvement in MG-ADL score of 2 points or more was observed for 88.7% of the patients who received efgartigimod alfa (EFG-EFG or PBO-EFG) and an improvement of 5 points or more was observed for 62.0%. At the final ADAPT+ analysis (data cut-off: June 30, 2022), mean change in MG-ADL score was greater than or equal to 5 points for all cycles except cycle 16 (SE = 4.9 points) in the total efgartigimod alfa group of the AChR-Ab+ population.

Figure 11: MG-ADL Total Score Mean Change From Cycle Baseline in the AChR-Ab+ Population (Safety Analysis Set) [Redacted]

Source: ADAPT+ final Clinical Study Report (Figure 3, data cut-off: June 30, 2022).⁶⁸

Exploratory: QMG Scores (Part A Only)

As with the MG-ADL results, QMG data are presented for the cycle baseline and the week 3 time point within each cycle because the ADAPT+ visit schedule was less intensive than the ADAPT trial schedule. Data at all assessment points up to cycle 7 (only collected during the first 1 year of ADAPT+) are shown graphically for the AChR-Ab+ population (Figure 9).

Consistent with the MG-ADL results, in the AChR-Ab+ population, the mean change from cycle baseline in the QMG score was greater than the MID of 3 points across all time points.

During the first 7 cycles of ADAPT+, an improvement of greater than or equal to 3 points in QMG score was observed for 64.4% of patients with AChR-Ab+ gMG who received efgartigimod alfa (EFG-EFG or PBO-EFG) and an improvement of greater than or equal to 6 points was observed for 36.0%.

Figure 12: QMG Total Score Mean Change From Cycle Baseline in the AChR-Ab+ Population (Safety Analysis Set) [Redacted]

||| ||||| ||||| |||| |||||| |||| ||||| |||||||| || ||| |||||||| |||||||||| ||||||| |||||||| |||| |||| ||||| |||||| | ||||| ||| ||| ||||| ||||| |||| |||||| |||| ||||| |||||||| |||| ||||| | || ||||| | || ||| ||||| || |||||||||||| ||||||||||| ||| ||||||| |||| |||||| ||||||||||| |||| ||||| |||||||| |||||||| || |||| | || |||| ||||| || |||| |||||||||||| |||| | |||||||||||| |||| ||| ||||||| | |||||||||||| |||| ||||||| ||| ||||||||| || ||| |||||| |||| |||||||| |||||||| ||||||| || ||| ||| || |||| |||||| ||| ||||||| ||||||||| |||||||| ||||||| |||| ||||| | || ||||| || |||||||||| | |||||||||||||||||| |||||||| ||||||||| ||| | |||||||||||| |||||||||| ||||||| || | |||||||| |||||| ||||||| |||||| ||||| |||||||| ||||| |||||| | |||| |||||||| |||| ||| ||||||68

Harms

Some safety outcomes were reported for the AChR-Ab+ population in the sponsor’s Clinical Study Report. A summary of harms data collected in ADAPT+ up to the data cut-off (December 15, 2022) is shown in Table 25.

Adverse Events

The most commonly reported TEAEs were headache (|| ||||||||| |||||||) nasopharyngitis (|| ||||||||| |||||||), COVID-19 (|| ||||||||| |||||), and diarrhea (|| ||||||||| |||||||). These results are consistent with the safety profile of efgartigimod alfa in ADAPT.

Serious Adverse Events

SAEs occurred in || |||||||| |||||). The most common SAE was MG (| ||||||||| ||||||).

Adverse Events of Special Interest

TEAEs in the system organ class of infections and infestations were classified as AESIs and were reported by 35 patients in the group that received efgartigimod alfa in both ADAPT and ADAPT+ (57.4%) and 26 patients in the group that received placebo in ADAPT and efgartigimod alfa in ADAPT+ (52.0%).

Critical Appraisal

Internal Validity

The ADAPT+ study was limited by its open-label and noncomparative design. Because there was no comparator, it cannot be confirmed whether the observed results are attributable to the effects of the drug or to natural history. Furthermore, the missing outcome data and small sample size toward the end of ADAPT+ made it difficult to draw any firm conclusions on the efficacy and safety of efgartigimod alfa. Because the study was open label and nonblinded, the subjective outcomes (e.g., rates of self-reported AEs and MG-ADL and QMG scores) are at risk of bias, most likely in favour of the intervention (i.e., efgartigimod alfa), at least for efficacy outcomes. The population enrolled in ADAPT+ had demographic and clinical characteristics that were consistent with their characteristics at entry into the ADAPT study. The design, eligibility criteria, and efgartigimod alfa dosing regimen were consistent in both studies, although it is noteworthy that ADAPT+ had fewer scheduled visits for outcome assessments (no visits were scheduled at weeks 4 to 6 of a cycle) compared with the ADAPT trial (every week for the first 8 weeks of a cycle); ADAPT+ only collected MG-ADL data at week 3 of each study. Furthermore, the longer-term safety and tolerability profile of efgartigimod alfa treatment was hard to determine because the rates of AEs and SAEs may be underestimated, since ADAPT+ had fewer scheduled visits for outcome assessments and MG-ADL data were collected only at week 3 of each study cycle, instead of 3 times (week 3, 4, and 5 of each cycle) in ADAPT trial.

In ADAPT+, efficacy was assessed as an exploratory outcome using the patient-reported MG-ADL and physician-reported QMG scores. Patients were to remain on their stable dose and regimen of concomitant gMG treatment during part A of ADAPT+. Given the presence of rollover effects from part A, efficacy results related to part B may be difficult to interpret because changes were permitted in part B in the type, dose, or regimen of the concomitant gMG treatment and in the use of other treatments. Therefore, the effects of other therapies cannot be eliminated as contributors to treatment response in part B. Although no detailed safety outcomes for the AChR-Ab+ population were reported in sponsor’s evidence summary and Clinical Study Report (i.e., ADAPT+), some safety outcomes were reported for the AChR-Ab+ population in the Clinical Study Report.⁶⁸ These appear to be consistent with the safety profile of efgartigimod alfa observed in the ADAPT trial.

External Validity

Because the patients who took part in the open-label long-term safety extension phase were originally from the pivotal ADAPT trial, it is reasonable to expect that the same limitations to generalizability are relevant to the open-label long-term safety extension phase. Given the nature of noncomparative study design, it is not possible to compare the effectiveness and tolerability of efgartigimod alfa as add-on treatment of gMG against others add-on treatment (e.g., ravulizumab, eculizumab, and IVIg). In terms of outcome measures, some important long-term outcomes reported by patients and clinicians (e.g., HRQoL, MG exacerbations) were not measured in the ADAPT+ trial.

Indirect Evidence

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

Objectives for the Summary of Indirect Evidence

The objective of this section is to summarize and critically appraise available indirect evidence comparing efgartigimod alfa to other therapies for the treatment of patients with gMG in Canadian settings.

Description of the Sponsor-Submitted ITC

The efficacy and safety of efgartigimod alfa as add-on therapy for adult patients with AChR-Ab+ gMG whose symptoms persist despite adequate doses of conventional therapy (AChEIs, corticosteroids, and/or NSISTs) have been previously assessed in the ADAPT trial.²⁸ However, no head-to-head evidence of efgartigimod alfa compared against other treatments for gMG was available for this review. Due to this gap in evidence, the sponsor submitted an ITC, including an SLR³⁸ and an NMA,¹ that provides comparative evidence of the efficacy of efgartigimod alfa relative to ravulizumab and IVIg. Data from this ITC were used to inform the pharmacoeconomic model.

Indirect Treatment Comparison Design

Objectives of the Sponsor’s ITC

The objective of the sponsor’s ITC was to assess the comparative effectiveness of efgartigimod alfa compared with other therapies used in Canada for the treatment of patients with gMG whose symptoms are not well-controlled by the conventional treatment options.

Study Selection Methods

The sponsor conducted a systematic literature search to identify studies assessing any chronic or acute pharmaceutical interventions for adult patients with gMG, and to evaluate the comparative efficacy and safety of efgartigimod alfa and alternative treatments in this patient setting. Methods and results were described as outlined by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement and the corresponding extension statement for the NMA.⁶⁹ Table 26 shows the study selection criteria and key aspects of the methods for the systematic review.

An electronic search of the literature was conducted on February 14, 2020, and updated on April 7, 2022. The search was conducted using the following databases: MEDLINE, 1946 to present (OVID); MEDLINE In-Process and Other Non-Indexed Citations (OVID); MEDLINE Epub Ahead of Print (OVID); Embase, 1980 to present (OVID); Cochrane Central Register of Controlled Trials (CENTRAL) (Wiley); Cochrane Database of Systematic Reviews (CDSR) (Wiley); PubMed (NLM) — e-publications only; Database of Abstracts of Reviews of Effects (DARE); Health Technology Assessment (HTA) database; International HTA (INAHTA) database; Conference Proceedings Citation Index-Science (CPCI-S), 1990 to present (Web of Science, Clarivate Analytics); International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Presentations Database. Supplementary searches were conducted using ClinicalTrials.gov, EU Clinical Trials Register, and the WHO International Clinical Trials Registry Platform (WHO ICTRP), and the proceedings of recent conferences since March 1, 2020: European Academy of Neurology Congress, American Academy of Neurology annual meeting, and Peripheral Nerve Society annual meeting. A search for other published NMAs was not explicitly carried out for the SLR. The study screening, selection, data extraction, and risk-of-bias appraisal processes were conducted by 2 independent reviewers. Disagreements between researchers were resolved through discussion or, if necessary, through consultation with a third researcher.

Table 26: Study Selection Criteria and Methods for Systematic Review

AChR-Ab+ = acetylcholine receptor antibody positive; AE = adverse event; CRD = Centre for Reviews and Dissemination; gMG = generalized myasthenia gravis; HRQoL = health-related quality of life; MG-ADL = Myasthenia Gravis Activities of Daily Living; MGC = Myasthenia Gravis Composite; MG-QoL15 = Myasthenia Gravis Quality of Life 15-item; QMG = Quantitative Myasthenia Gravis; RCT = randomized controlled trial; SAE = serious adverse event; SLR = systematic literature review; TEAE = treatment-emergent adverse event.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

^aPublications in languages other than English were tagged during screening for record-keeping purposes but were not included.

^bData were not extracted; reference lists were cross-checked for additional relevant records.

Sources: Sponsor-submitted Indirect Treatment Comparison and Systematic Literature Review reports.^1,38

From the evidence base created from the SLR, the sponsor further identified studies that were potentially eligible for an ITC based on a set of additional criteria listed in Table 27. Studies were required to be placebo-controlled randomized trials of treatments for adults with AChR-Ab+ gMG, reporting on at least 1 of the following outcomes: MG-ADL responders, or change from baseline in MG-ADL score; and QMG responders, or change from baseline in QMG score. The eligible interventions for the ITC were limited to those used in Canada for the treatment of gMG to ensure that the comparators were relevant to Canadian settings. Eligible interventions to be included in the NMA were efgartigimod alfa, ravulizumab, maintenance IVIg or subcutaneous immunoglobulin, and PE. Ravulizumab has been approved by Health Canada for the treatment of adult patients with AChR-Ab+ gMG and is currently under review by CADTH. Eculizumab was not considered a relevant comparator for this ITC by the sponsor because as it has only been studied in patients with AChR-Ab+ and refractory gMG. Treatment guidelines⁷¹ indicate that eculizumab is expected to continue to be considered as a last resort for patients who have not experienced treatment success on other immunotherapies, which is different from the expected place in therapy for efgartigimod alfa. In addition, eculizumab is not currently publicly funded in Canada. Rituximab was not considered by the sponsor to be a key comparator for efgartigimod alfa in patients with AChR-Ab+ gMG and thus was not included in the primary ITC analysis.

Table 27: Feasibility Assessment Study Selection Criteria

AChR-Ab+ = acetylcholine receptor antibody positive; gMG = generalized myasthenia gravis; IVIg = intravenous immunoglobulin; MG = myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living scale; MGC = Myasthenia Gravis Composite; QMG = Quantitative Myasthenia Gravis; SCIg = subcutaneous immunoglobulin.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

^aTreatments commonly included in the conventional therapy for gMG were considered as a part of placebo treatment and thus not eligible for inclusion as independent treatment arms.

Sources: Sponsor-submitted Indirect Treatment Comparison report and Systematic Literature Review report.^1,38

Risk-of-Bias Assessment

Risk-of-bias assessments were performed by 2 independent reviewers using the CRD Guidance for Undertaking Reviews in Health Care.⁷⁰ Included studies were evaluated using a fixed set of domains of bias focused on different aspects of trial design, conduct, and reporting. Eight specific domains were examined: random sequence generation, allocation concealment, blinding of participants, blinding of investigators, blinding of outcome assessors, incomplete outcome data, selective reporting, and other potential sources of bias that may affect internal or external validity and generalizability of the study findings to the general population. Disagreements between researchers were resolved through discussion or, if necessary, through consultation with a third researcher.

Feasibility Assessment

Prior to conducting analyses, groups of studies informing distinct potential NMAs were compared in detail to validate the exchangeability assumption underlying the methodology. The following elements of study design were compared for similarity in methods across studies: blinding, study phase, geographical distribution of study centres, and the time-periods of enrolment and follow-up. Decisions to conduct distinct NMAs were made based on the degree and potential impact of cross-trial heterogeneity among the trials that would be indirectly compared.

ITC Analysis Methods

Analytical Framework

All NMAs were performed using a Bayesian framework.^72-74 Placebo was chosen as the reference treatment for all analyses, given its presence as an anchor treatment in all studies and the outcomes assessed in the network. All analyses only included patients with AChR-Ab+ gMG as all comparator studies were performed exclusively in the AChR-Ab+ subpopulation. Primary analyses were conducted based on efficacy estimates observed at each trial’s primary assessment time point or, if unreported, at the end of the randomized controlled period. A sensitivity analysis was conducted at or plus or minus 2 weeks of week 4, which was the primary time point of assessment in ADAPT.

Evidence Network

Evidence network diagrams were created to visualize the evidence base for each analysis, where similar interventions identified among the studies (based on dosage and frequency of administration) were grouped as treatment nodes. Studies included in the network were meant to inform the NMAs. In the network diagram, treatment nodes were sized to reflect the proportionate numbers of patients randomized to each treatment, with larger nodes representing more patients.

Model Effects, Iterations, and Convergence

Given the paucity of links between multiple studies in the evidence network, a random effects NMA was not considered feasible, and fixed effects NMAs were performed for all outcomes. All models were built using 4 unique sets of initial values and were based on burn-in and sampling durations of 20,000 iterations or more. Convergence was monitored quantitatively using the latest implementation of the Gelman-Rubin diagnostic (Rhat) based on 4 chains.⁷⁵ This new implementation captures nonconvergence from stationary but nonoverlapping chains, overlapping nonstationary chains, chains with heavy tails, and chains with different variance. Samples were considered to have converged if Rhat was equal to or less than 1.05. After convergence was reached, there was concern about whether there were sufficient independent samples for stable estimates. The newest version of effective sample size and Monte Carlo standard error (MCSE) estimation were used to ensure that there were enough samples after convergence to support the inference.⁷⁵ If the rank-normalized effective sample size was greater than 400 (i.e., 100 per chain), then samples were taken to ensure that the MCSE was small enough to allow for stable estimates to at least 1 decimal place.⁷⁵ All assessments of effective sample size and MCSE were produced for each reported parameter.^72,76

Model Priors

By default, vague prior distributions that assume no pre-existing information, according to the National Institute for Health and Clinical Excellence (NICE),⁷⁷ were assigned for the treatment effects and trial baselines (Table 28).

Table 28: Vague Prior Distributions Used in Network Meta-Analyses

MG-ADL = Myasthenia Gravis Activities of Daily Living; QMG = Quantitative Myasthenia Gravis.

Note: Details included in the table are from the Sponsor’s Summary of Clinical Evidence.

Source: Sponsor-submitted Indirect Treatment Comparison report.¹

Assessment of Consistency

While the use of an unrelated mean effects model (i.e., an inconsistency model) was planned to test for inconsistency, there were no independent closed loops in the evidence network; therefore, no analyses evaluating consistency of direct and indirect evidence were performed.⁷⁴ Decisions to conduct distinct NMAs were made based on the degree and potential impact of cross-trial heterogeneity of studies that would be compared indirectly.

Outcome Measures

The efficacy outcomes included change from baseline in MG-ADL and QMG at the primary assessment time points, ranging from 4 to 26 weeks. The primary analysis was conducted in patients with AChR-Ab+ gMG based on efficacy estimates observed at each trial’s primary assessment time point or, if unreported, at the end of the randomized controlled period. For ADAPT, the primary assessment time point was week 4. An NMA for continuous, arm-based outcomes was used to compare MG-ADL and QMG. Data inputs from the individual trials were the mean change from baseline and standard errors. Pairwise comparisons of interventions estimated from NMAs were expressed using mean differences with 95% CrIs.

Table 29: ITC Analysis Methods

AChR-Ab+ = acetylcholine receptor antibody positive; ITC = indirect treatment comparison; gMG = generalized myasthenia gravis; MG-ADL = Myasthenia Gravis Activities of Daily Living; NICE = National Institute for Health and Clinical Excellence; NMA = network meta-analysis; QMG = Quantitative Myasthenia Gravis; RCT = randomized controlled trial.

Note: Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Source: Sponsor-submitted Indirect Treatment Comparison report.¹

Results of ITC

Summary of Included Studies

The electronic search for clinical evidence identified 2,171 clinical and 1,503 economic citations across the databases searched. An additional 1,575 citations were collected from supplementary searching, including recent conference abstracts, registers, websites, and hand-searches. Out of these 5,429 citations, 2,113 were identified as duplicates and were removed. Of the 3,136 unique citations, a total of 2,811 were excluded during title and abstract screening because they did not meet the prespecified inclusion criteria. Out of the remaining 325 citations, 239 were excluded at the full-text screening phase. Overall, 80 records were eligible for inclusion in the SLR, representing 40 clinical studies. After feasibility assessment, the following 5 studies conducted with patients with AChR-Ab+ gMG were included in the NMA: 2 trials comparing efgartigimod alfa with placebo;^28,39 2 trials comparing IVIg with placebo;^40,41 and 1 trial comparing ravulizumab with placebo⁴² (Table 30). The authors performed a risk-of-bias assessment that showed a low or unclear risk of bias for most of the included studies.

Eligibility Criteria

Inclusion and exclusion criteria were generally similar across studies. All trials other than ADAPT included only the AChR-Ab+ subpopulation. Patients with a recent thymectomy were generally excluded, although the exclusion period ranged from 3 to 12 months before across studies. All trials excluded patients who had recently received either IVIg, subcutaneous immunoglobulin, or PE treatment and patients with very mild disease (e.g., MGFA Class I). All studies excluded patients under the age of 18, except for CHAMPION MG (ravulizumab versus placebo) which was restricted to patients aged 15 or older, although the mean age for patients in that study was not meaningfully different from the age of patients in the other trials. Most of the studies excluded patients with a history of malignancy (3 of 5) or treatment with rituximab within the past 6 to 12 months (4 of 5). Less common exclusion criteria included pregnancy (2 of 5), MG crisis (2 of 5), and renal impairment (2 of 5).

Treatment Definitions and Concomitant Therapies

Treatment definitions were generally similar between studies. All studies allowed for the use of concomitant standard-of-care treatments (i.e., steroids or NSISTs), but detailed information on the breakdown of the actual concomitant medications used was not available. The sponsor assumed that these treatments did not interact with the therapies under consideration and that any effect was additive. As such, the sponsor considered the placebo arm of the evidence network to be inclusive of standard-of-care therapies available in Canada.

Efficacy Outcomes

In all eligible studies, results for MG-ADL and QMG scores were reported in different formats (e.g., proportion of responders, or mean change from baseline). The mean change from baseline in MG-ADL and QMG scores were the most consistently reported outcomes; thus, trials that did not report these outcomes were excluded after a feasibility assessment. Change from baseline in MG-ADL and QMG scores were neither primary nor secondary end points for the ADAPT trial but given that those outcomes were reported in all other eligible studies and that these end points were required for the cost-effectiveness model, they were calculated for use in the NMA.

Table 30: Summary of Included Trials in the Sponsor-Submitted ITC