CADTH Reimbursement Review

Inebilizumab (Uplizna)

Sponsor: Horizon Therapeutics Canada

Therapeutic area: Neuromyelitis optica spectrum disorders

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

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

AQP4-IgG = anti-aquaporin-4 immunoglobulin G; NMOSD = neuromyelitis optica spectrum disorder; NOC = Notice of Compliance.

Introduction

Neuromyelitis optica spectrum disorder (NMOSD) is a rare, chronic disorder of the central nervous system (CNS) that is characterized by acute attacks or relapses that cause inflammation in the optic nerve (optic neuritis) and spinal cord (myelitis).¹ A defining feature of NMOSD is the presence of pathogenic serum autoantibodies against aquaporin-4 (AQP4), which led to its differentiation from multiple sclerosis (MS).^2-4 This disease has a relapsing nature in which patients experience acute attacks and relapses characterized by new or worsening signs and symptoms. These attacks are unpredictable and can lead to accruing disabilities and, often, to permanent impairment.^5-9 The clinical presentation of an NMSOD attack typically involves optic neuritis that causes ocular pain and vision loss. Myelitis causes sensory loss, weakness, or paralysis in the legs or arms, painful spasms,^10-13 and bladder and bowel dysfunction.^3,5 At its worst, severe high-cervical myelitis and brainstem lesions can lead to fatal respiratory failure.^3,11,13 The consequences of NMOSD extend beyond clinical settings and include physical, functional, and psychological effects that alter every aspect of patients’ and caregivers’ lives and impact their health-related quality of life (HRQoL).^9,14-16

NMOSD disproportionately affects females.^5,17 Systematic reviews based on data from several countries have estimated there are 0.053 to 0.4 incident cases per 100,000 people and 0.51 to 4.4 prevalent cases per 100,000 people.^18,19 No Canadian-specific estimates were identified in these studies.

Inebilizumab is a humanized afucosylated monoclonal antibody that binds to CD19 for the treatment of NMOSD.^20-22 Inebilizumab has a Health Canada indication as monotherapy for the treatment of adult patients with NMOSD who are seropositive for anti-AQP4 immunoglobulin G (AQP4-IgG).²³ The reimbursement request is as per the approved indication. The recommended dosage for inebilizumab is an initial 300 mg dose via IV infusion followed 2 weeks later by a second 300 mg dose via IV infusion; subsequent doses (starting 6 months from the first infusion) are administered as single 300 mg doses via IV infusion every 6 months.²³

Stakeholder Perspectives

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

Patient Input

Two patient groups, MS Canada and the Sumaira Foundation (TSF) responded to CADTH’s call for patient input on this topic.

MS Canada gathered information for this submission through an online survey in 2023 that included 13 respondents. TSF gathered information in 2023 through 2 online surveys and videoconferencing interviews with patients and caregivers, and through TSF’s experience working in the NMOSD communities. The TSF survey included data from 51 patients and 9 caregivers.

The 2 patient groups indicated that NMOSD follows a relapsing-remitting disease course and is initiated with a severe attack and continues with subsequent devastating attacks that affect the patient’s vision and lead to mobility issues and chronic pain. The disease has a tremendous impact on all aspects of patients’ and caregivers’ lives, including a negative effect on their quality of life, their independence and employment, and on their social, family, and school life.

The patient inputs stated that treatment for NMOSD involved IV steroids, IV immunoglobulin (IVIG), or plasmapheresis or plasma exchange, as well as the use of off-label drugs with varying levels of therapeutic benefit due to worsening symptoms and/or challenging side effects experienced by the patient while cycling through different therapies.

According to the 2 patient inputs, access is very limited for therapies such as eculizumab, satralizumab, and rituximab, and the administration schedule for eculizumab can be too arduous for some patients. According to these inputs, patients need to have access to therapy options that can reduce the risk of future attacks, maintain the current level of physical ability, and slow disease progression.

Clinician Input

Input From the Clinical Experts Consulted by CADTH

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

Unmet Needs

The clinical experts indicated that the treatment of NMOSD includes a series of treatment goals relating to 3 broad areas: prevention of relapses (disease modification), treatment of relapses, and treatment of residual symptoms. Although ideal, it is unlikely that any single treatment would cover all 3 areas. Of these 3 areas, preventive treatment was described to be of special interest because relapses are the major source of disability accumulation for people with NMOSD. As a result, preventive treatment is expected to result in downstream desirable effects, including better HRQoL, an increased ability to maintain independence and employment, and a reduced reliance on caregivers. The clinical experts highlighted that it is important to control progression as early as possible because damage leading to neurologic disability, including paralysis and blindness, may be irreversible after an attack.

The clinical experts agreed that most patients with NMOSD still have relapses despite their current treatment regimens, and current first-line therapies (e.g., azathioprine and mycophenolate mofetil) are not considered particularly effective in preventing NMOSD attacks. Additionally, these therapies are associated with significant adverse effects, especially if used in conjunction with corticosteroids. The greatest unmet need in the treatment of NMOSD is for therapies that prevent relapses more effectively without intolerable side effects.

Place in Therapy

The clinical experts agreed that the current clinical practice for treatment and attack prevention in NMOSD is suboptimal and inconsistent due to the low efficacy of the off-label treatments used in the first-line setting and barriers to treatment access, which vary by province. The newer therapies with indications for NMOSD, such as satralizumab and eculizumab, present especially significant barriers to access for most patients due to inconsistent coverage or the lack of public reimbursement, and the onerous dosing schedule of eculizumab. Due to the impact of NMOSD on a person’s ability to maintain employment, patients with NMOSD are more likely to lack private insurance.

As the vast majority of patients with NMOSD will relapse and relapses may lead to permanent disability or death, all people with NMOSD should be on relapse-preventive treatment.

The experts agreed that because the impact of relapses on people with NMOSD is devastating and preventing as many relapses as possible is critical to the prevention of significant disability, inebilizumab should be available as first- and later-line treatment for patients diagnosed with AQP4-IgG–seropositive NMOSD. The experts indicated that inebilizumab would be used as a monotherapy.

Patient Population

NMOSD is a rare disease. The clinical experts indicated that patients diagnosed with NMOSD who are seropositive for AQP4-IgG should be candidates for treatment with inebilizumab. It is the standard of care in Canada to assess patients with NMOSD for AQP4-IgG; the experts noted there are no major challenges in this regard in Canada, other than that the test results may be delayed in some locations. The specificity of AQP4-IgG is very high, so the risk of misdiagnosis is very low.

The patient eligibility criteria in the pivotal study, N-MOmentum, were considered by the experts to be broadly representative of patients with NMOSD in Canada, with the exception that the study excluded patients with recent steroidal treatment. Although appropriate from the perspective of clinical trial design, in real-world practice there are many comorbidities that may require steroid treatment, and these patients should not necessarily be excluded from treatment with inebilizumab; such a decision would need to be considered by the expert clinician managing a patient’s particular case. Patients who have received IVIG or have concomitant diseases should not necessarily be excluded from receiving inebilizumab in the real world.

Within the population of patients who have NMOSD and are AQP4-IgG seropositive, it is unknown which patients are more likely to benefit from inebilizumab.

It is possible that patients who are seronegative for AQP4-IgG may also benefit from inebilizumab. Fulfillment of seronegative NMOSD criteria would be necessary to establish the diagnosis to allow appropriate access.

Assessing the Response to Treatment

The experts indicated that a clinically meaningful response to treatment relates to the reduction of the relapse rate and prolongation of times to relapse. Although the absence of relapse is indicative of a clinically meaningful response, this may not be realistic, as the number and severity of relapses that patients experience differ on an individual level (e.g., some patients may have several relapses per year); thus, a reduction in the number of relapses is still a reasonable goal. The determination of relapses is fairly objective; however, it is not the only factor, and assessment of treatment response is based on a combination of patient-reported symptoms, a clinical exam, clinical tools, and patient history. Other important outcome measures include evaluation of attack severity and degree of recovery from attack, as well as accumulation of disability.

The experts noted there is a lack of formal guidance on how to assess treatment response, but it would be reasonable to assess initial treatment response 3 months after the initial infusion then every 6 months until stability is achieved, and then every year for patients with stable NMOSD. However, it was noted that within the first 6 months of treatment, the attack rate may still be higher than when stability is achieved.

MRIs are not routinely conducted for patients with NMOSD outside of initial diagnosis and so would not be used in assessing the response to treatment.

Discontinuing Treatment

Discontinuation of treatment should occur if the patient is completely dependent and unable to leave their bed (Expanded Disability Status Scale [EDSS] score of 9.0 and above).

Discontinuation of treatment should be considered on a case-by-case basis in the event of a severe relapse (e.g., requiring intubation and support on a ventilator) or if the patient has experienced 2 or more relapses within 2 years, has severe or unacceptable adverse events (AEs), or has contraindications for therapy.

Prescribing Considerations

The clinical experts agreed that treatment should be supervised by a neurologist with expertise in this area. Although NMOSD and MS are not the same disease, the populations and medications are similar and patients with NMOSD are often cared for in MS clinics. The clinical experts stated that neurologists with experience or expertise in related subspecialties should prescribe inebilizumab, including clinicians with experience in MS neurology, and/or neurologists working in an MS clinic or in neuroimmunology, autoimmune neurology, and/or neuro-ophthalmology. However, patients in remote areas may have issues with access to subspecialists. For patients living in remote areas, local neurologists without subspecialty expertise could work by distance in conjunction with neurologists who are experts in a relevant subspecialty.

Inebilizumab is expected to be used as a monotherapy and not combined with other monoclonal antibodies indicated for the treatment of NMOSD. However, there may be situations in which it is combined with classical immunosuppressants. There is a lack of data regarding combination therapies.

Clinician Group Input

One clinician group, the Canadian Network of Multiple Sclerosis Clinics (CNMSC), responded to CADTH’s call for clinician group input (input authored by 1 clinician).

According to the CNMSC, a variety of off-label therapies are used for the treatment of NMOSD in Canada, including corticosteroids, azathioprine, mycophenolate mofetil, and rituximab. However, breakthrough NMOSD attacks are reported on all of these drugs, and government drug program funding varies by province and territory. Recently, 2 monoclonal antibodies, eculizumab and satralizumab, were approved by Health Canada. However, access to these therapies is extremely limited due to their high cost and stringent funding criteria. All of the therapies in use for NMOSD work by suppressing the immune system to prevent attacks, with variable efficacy. Failure of treatment, resulting in even just 1 relapse, can lead to a profound, permanent disability, including blindness and paralysis.

The clinician group input noted there is a large unmet need in Canada for high-efficacy, well-tolerated therapies for NMOSD that have a significant impact on preventing and/or reducing attacks. Use of some off-label therapies is limited by many side effects, and many patients continue to have attacks despite treatment with drugs such as azathioprine and mycophenolate mofetil and, to a lesser extent, rituximab. Also, eculizumab is given by IV infusion every 2 weeks, which is too onerous for some patients to tolerate.

According to the clinician group, the main treatment goals include the use of an efficacious, safe, and tolerable therapy administered immediately after the first attack to ideally avoid all future relapses, reduce the severity of attacks and the cumulative disability associated with them, and minimize AEs related to therapies. In particular, there is a major unmet need among patients who have a breakthrough attack on their first therapy, as it can be challenging to identify a subsequent therapy that will be effective at preventing attacks and will be tolerated by the patient. The best approach for patients is to use a product after an attack that is as highly efficacious as possible so as to avoid potentially catastrophic subsequent attacks and, thus, optimize patient outcomes. The clinician group input noted that inebilizumab could be used as first-line treatment and as subsequent treatments for patients who have had breakthrough attacks on other therapies or who were intolerant of other therapies. Inebilizumab would be expected to be used as a monotherapy based on the available clinical evidence and to avoid cumulative immunosuppressive effects. The clinician group also noted that, although rituximab and inebilizumab both suppress B cells, there is some evidence that patients with polymorphisms in the FCGR3A gene may have an incomplete response to rituximab but not to inebilizumab. There is a lack of head-to-head data to compare inebilizumab with rituximab. The CNMSC also noted there is no clear preferred drug among the novel monoclonal antibodies for the treatment of NMOSD (e.g., eculizumab, inebilizumab, ravulizumab, and satralizumab) and that the best mechanism of action may vary by patient; however, there is generally limited access to these therapies in Canada at this time.

According to the CNMSC, the key outcome measure is a new attack, which is marked by new neurologic symptoms such as vision loss, weakness, sensory impairment, or dysfunction of the bladder or bowel. Although usually marked by a new enhancing lesion on MRI, this is not necessary to diagnose an attack. The clinician group indicated that the drug renewal process should consider the occurrence of any relapse in the previous year and the number of relapses, EDSS score or results of other disability measures, and any change from baseline (note that the EDSS is not validated in NMOSD). The CNMSC recommended discontinuing the drug if the patient has a new attack, a serious AE related to the therapy, or an EDSS score of 8 or higher.

The CNMSC stated that the treatment of patients with NMOSD should be assessed and managed by neurologists specialized in demyelinating diseases through an MS or demyelinating disease centre, and inebilizumab can be administered in a hospital or private clinic. Patients eligible for treatment with inebilizumab should have a confirmed diagnosis of NMOSD and a positive serum test for AQP4-IgG.

Drug Program Input

Input was obtained from the drug programs that participate in the CADTH Reimbursement Review process. The following were identified as key factors that could potentially impact the implementation of a CADTH recommendation for inebilizumab:

• relevant comparators

• considerations for:

  • the initiation of therapy

  • continuation or renewal of therapy

  • discontinuation of therapy

  • prescribing of therapy

• generalizability.

The clinical experts consulted by CADTH provided advice on the potential implementation issues raised by the drug program. Refer to Table 4 for more details.

Clinical Evidence

Systematic Review

Description of Studies

One double-blind, randomized, placebo-controlled phase II and III study (N-MOmentum) was included in this review. The N-MOmentum trial consisted of a 197-day randomized controlled period (RCP) and a single-arm open-label period (OLP) that had a minimum duration of 2 years. N-MOmentum randomized 231 patients who were adults with NMOSD, had a documented history of either 1 or more acute NMOSD attacks in the prior year or 2 or more attacks in the prior 2 years that required rescue therapy, and had an EDSS score of 7.5 or less or 8.0 in special circumstances (i.e., if the investigator and medical monitor assessed that the patient was reasonably able to complete the study). The EDSS is a measure of disability validated for use in patients with MS that has been applied in NMOSD due to the similarities in disability caused by these distinct conditions and due to the lack of an NMOSD-specific tool for assessing disease-related disability. The EDSS ranges from 0 to 10, where 0 represents no disability, 9 represents a complete lack of independent mobility, and 10 represents death. The majority of patients were seropositive for AQP4-IgG, and subgroup data were available for the seropositive population. The primary end point was the time in days from day 1 to the onset of an NMOSD attack (determined by an adjudication committee [AC]) on or before day 197. Key secondary end points included the proportion of patients with a worsening in EDSS score from baseline to last visit during the RCP, change in binocular low-contrast visual acuity score from baseline to last visit during the RCP, and the number of NMOSD-related inpatient hospitalizations during the RCP. Other secondary or exploratory outcomes included the NMOSD attack rate in inebilizumab-treated patients, safety outcomes, and HRQoL, which was measured using the Short Form (36) Health Survey (SF-36). The binocular low-contrast visual acuity score was measured using a low-contrast Landolt C broken ring chart; the scoring of this assessment is based on the number of characters on the chart that the patient is able to identify, from 0 to 70 inclusive, where 70 indicates the patient was able to correctly identify all characters on the chart (i.e., best visual acuity score), and 0 indicates they were not able to identify any characters correctly (i.e., poorest visual acuity score). The SF-36 is a generic HRQoL questionnaire that yields a physical component score and a mental component score, in which higher scores represent better HRQoL.

At baseline, the patients included in the N-MOmentum trial were mostly female (> 90%), had received prior acute or maintenance therapies for NMOSD (> 98%), had disease that was seropositive for AQP4-IgG (> 93%), and had a mean age of approximately 43 years. The median EDSS score at baseline was 4 in the placebo group and 3.5 in the inebilizumab group (range, 0.0 to 8.0); 29% and 24% had an EDSS score of greater than 5 points at baseline, respectively.

Efficacy Results

NMOSD Attacks

Among the patients with AQP4-IgG–seropositive NMOSD in the randomized period, treatment with inebilizumab (versus placebo) was associated with a 77.3% reduction in the risk of an AC-determined NMOSD attack (Kaplan-Meier hazard ratio [HR] = 0.227; 95% confidence interval [CI], 0.1214 to 0.4232; P < 0.0001). During the RCP, a larger proportion of patients were attack-free in the inebilizumab group (87.6%) than in the placebo group (56.6%). Treatment with inebilizumab likely results in a clinically important increase in the probability of having no attack at day 197 compared with placebo.

The results were similar in the overall intention-to-treat (ITT) population and were also consistent across prespecified subgroups. Additionally, the results were similar based on investigator-determined NMOSD attacks. During the RCP among patients with seropositive NMOSD, major attacks occurred in 6 of 18 attacks (33.3%) in patients treated with inebilizumab and in 10 of 22 attacks (45.5%) among patients treated with placebo. Recovery from attacks was graded by the AC based on improvements in the attack criteria. As a proportion of patients with attacks, “no attack recovery” was reported for 27.8% of patients in the inebilizumab group and 40.9% of patients in the placebo group. Most attacks were myelitis (11 of 18 in the inebilizumab arm and 14 of 22 in the placebo arm), followed by optic neuritis (8 and 10, respectively), and few occurred in the brainstem (0 and 1, respectively). When calculated across the RCP and OLP, the annualized rate of AC-determined NMOSD attacks in any patient treated with inebilizumab was 0.086 attacks per year in the total population and 0.09 attacks per year in the population with AQP4-IgG–seropositive NMOSD.

Proportion of Patients with a Worsening in EDSS Score

During the RCP among patients with AQP4-IgG–seropositive NMOSD, treatment with inebilizumab likely resulted in a clinically important reduction in the proportion of patients who experienced a worsening from baseline in EDSS score compared with placebo at 197 days (odds ratio = 0.355; 95% CI, 0.1704 to 0.7252; P = 0.0047) ||| ||||||||||||| ||||||||||| |||||| ||||||| ||| ||||||| ||||||||. Results were similar in the overall ITT population.

Change From Baseline in Low-Contrast Visual Acuity Score

The change in binocular low-contrast visual acuity score from baseline to the last RCP visit did not appear to differ by treatment group within the population with AQP4-IgG–seropositive NMOSD ||| |||| ||||||||||| ||||||| ||| ||| ||||||| || |||||||||||||||. Results were similar in the overall ITT population. Based on these results, inebilizumab likely does not result in a clinically important difference in low-contrast visual acuity compared with placebo at 197 days.

Number of NMOSD-Related Inpatient Hospitalizations

|||||| ||| ||| ||||| ||| |||||||| |||||||||||| ||||||||||| ||| |||| ||||| || ||||||||||||| ||||||||| |||||||||||||||| ||| ||||| |||| ||| |||||| || ||||||||||||| ||| ||| ||||||||||||| |||||||| |||||||||| ||| ||||| |||||| ||| |||||| |||||| ||||||| |||| ||||||| || ||| ||||||| ||| ||||||||||| || ||| ||||||||| || ||| |||||||| |||||||| |||||||||| ||| |||||||||| |||||||||| ||||||| ||| ||||| ||| ||||||| ||||| ||| ||| |||||||| || ||| ||||| ||| || |||||||||||| || |||||| ||||||||||| || ||||||||||||| ||||||||| ||||||||||||||||| ||| ||| || || ||| |||||||| |||||||||| |||||| |||| ||||||| || ||||||||| ||| ||| ||||||| |||||. The clinical experts consulted by CADTH indicated that any benefit would be clinically meaningful. In summary, inebilizumab may be associated with a benefit in this outcome but it is uncertain at the time frame assessed.

Short Form (36) Health Survey

|||||| ||| ||| ||||| |||||||| |||||||||||| |||||||| ||||||| |||| |||||||||||| ||| |||||||| ||| |||| |||||| |||| |||||||| || ||| |||||| ||||||||| ||||| ||| ||||| ||||||||| ||||||||| ||||| |||||| ||| ||||| |||| |||||| || |||| ||| ||||||||||||| ||| |||| |||||| |||| |||||||| || ||| |||||||| ||||||||| ||||| ||| ||||| |||| |||||| ||| ||||| |||| |||||| || |||| ||| ||||||||||||. No statistical test results were reported, but it was reported that there were no numerical differences between treatment arms.

Pain Numerical Rating Scale

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

Harms Results

Nearly all patients with AQP4-IgG–seropositive NMOSD experienced at least 1 AE during the study (87.2% of patients treated with placebo and 86.4% of patients treated with inebilizumab, respectively) and the results were similar in the ITT subpopulation. Among patients with AQP4-IgG–seropositive NMOSD, the rate of serious AEs (SAEs) was 4.3% in the inebilizumab group and 11.5% in the placebo group during the RCP. Over the entire duration of the study, 20.4% of patients who received any dose of inebilizumab experienced an SAE. Withdrawals due to AEs were uncommon: including both the RCP and OLP, withdrawals due to AE occurred in 1 patient (who had AQP4-IgG–seropositive disease) treated with placebo only, and in 4 patients who received inebilizumab (of which 3 had NMOSD that was AQP4-IgG seropositive, and 1 that was seronegative). There were no deaths during the RCP. During the entire study, among patients treated with any dose of inebilizumab, 3 patients died due to NMOSD, pneumonia, and COVID-19 pneumonia (1 case each).

In the RCP among patients with AQP4-IgG–seropositive NMOSD, 50.0% in the placebo group and 49.1% in the inebilizumab group experienced at least 1 treatment-emergent AE (TEAE) of special interest, most commonly infections (44.2% and 40.4%, respectively) followed by infusion-related reactions (9.6% and 9.3%, respectively), hepatic function abnormality (3.8% and 5.0%), and cytopenia (0% and 5.0%). Results were similar in the overall ITT population.

During the OLP, among patients with AQP4-IgG–seropositive NMOSD, most patients experienced at least 1 TEAE of special interest (85.1% of patients in the placebo-to-inebilizumab group and 71.4% in the inebilizumab-to-inebilizumab group). Similar to the RCP, the most common TEAE of special interest was infection, followed by infusion-related reaction, hepatic function abnormality, and then cytopenia. In addition, a few patients experienced hypersensitivity (0% and 1.3%, respectively) |||||||||| || ||||||||| ||||||||||| |||||||||| |||||||||||||||||||| || ||| |||||||||||||||||||||||||||| |||||| ||||||||||||||||| ||||||||| ||||| ||| || ||||||||||||. Again, results were similar in the overall ITT population. The infections that occurred were generally mild and did not lead to treatment discontinuation in the OLP or RCP. However, higher rates of infection were observed during the OLP (versus during the RCP), which may be related to the prolonged duration of treatment and follow-up. Cytopenias were more common in inebilizumab-treated patients, which is consistent with inebilizumab’s mechanism of action and the class effects of B-cell depletion.

Critical Appraisal

The N-MOmentum phase II and III trial was the only study included in this review. The N-MOmentum study included an initial period that was randomized, placebo-controlled, and double-blind, with a duration of up to 197 days (the RCP) in which patients received inebilizumab 300 mg IV or placebo IV on day 1 and day 15. Patients could proceed to an open-label, single-arm period (the OLP) with a minimum duration of 2 years, during which patients received inebilizumab 300 mg IV every 6 months starting 6 months after the first infusion. In the RCP, there were no major concerns with regard to internal validity related to the study design in terms of, for example, the method of randomization, concealment of allocation, maintenance of blinding, or balance of patient characteristics between treatment arms. As the trial was stopped early based on the recommendation of an independent data-monitoring committee, which found that the efficacy of inebilizumab had been established and, as there was no justification to keep exposing patients to placebo, there may be a risk of overestimating the true effect due to a slightly low information fraction (40 of 57 planned NMOSD attack events in the seropositive subpopulation). The end points in the trial were appropriately defined and were considered important to patients and clinicians, according to group inputs and clinical expert consultation. There was a high number of censored patients in the primary outcome of time to first NMOSD attack, especially in the inebilizumab treatment arm. However, this was considered unlikely to introduce bias because of the low number of early withdrawals and was considered likely to be due to the early cessation of the trial and the high proportion of attack-free patients at day 197. The key secondary outcome of change in EDSS score was an appropriate and important outcome, but the EDSS measure has some weaknesses, including over-reliance on ambulation as a metric of disability and a lower sensitivity to change in other types of disability at some ranges in the scale. EDSS is validated in MS but has not been validated in NMOSD; however, there are no superior scales for measuring disability in this population.

The eligibility criteria and baseline patient characteristics of the N-MOmentum trial were considered by the consulted clinical experts to be a reasonable approximation of patients with NMOSD in Canadian clinical practice, with the minor exception that some patients excluded for concomitant immunosuppressive or steroid therapy or prior IVIG would potentially be candidates for inebilizumab in real-world practice. All aspects of treatment management, including the steroid taper, rescue therapy, and preinfusion medications adequately reflected clinical practice, according to the clinical experts. The N-MOmentum trial had a high proportion of screen failures (236 screen failures of 467 patients screened) and only 5 of these were the result of the study’s early cessation. The clinical experts consulted by CADTH indicated that approximately one-third of the screen failures were due to tuberculosis testing because of the global nature of the study, which would be expected to be lower in Canadian clinical practice. As such, the CADTH team considered this not to be a major concern for generalizability.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For the pivotal studies and randomized controlled trials (RCTs) identified in the sponsor’s systematic review, the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) instrument was used to assess the certainty of the evidence for outcomes considered most relevant to inform the deliberations of the CADTH expert committee, and a final certainty rating was determined as outlined by the GRADE Working Group.^24,25

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null.

The selection of outcomes for the GRADE assessment was based on the sponsor’s summary of clinical evidence, consultation with clinical experts, and the input received from the patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with members of the expert committee:

• time to first NMOSD attack (assessed in GRADE as probability of no attack at day 168 and 197)

• disability (proportion with a worsening in EDSS score)

• low-contrast visual acuity (change from baseline to last visit)

• number of NMOSD-related inpatient hospitalizations

• HRQoL (SF-36 mean change from baseline)

• pain numerical rating scale (NRS) (mean change from baseline)

• proportion of patients with SAEs.

Long-Term Extension Studies

No long-term extension studies were submitted to CADTH.

Indirect Comparisons

Description of Studies

The sponsor-submitted indirect treatment comparisons (ITCs) included comparisons against satralizumab and eculizumab using published study data and matching-adjusted indirect comparison (MAIC) methodology to adjust for between-trial differences, and a comparison against rituximab using individual patient data (IPD). Other therapies (azathioprine and mycophenolate mofetil) were also of interest and were included in the study selection criteria, but ITCs against these therapies were ultimately not considered feasible. Additionally, there was a published network meta-analysis (NMA) comparing eculizumab with satralizumab and inebilizumab, which was summarized briefly for comparison but not formally assessed.²⁸

Efficacy Results

In the sponsor-submitted anchored MAICs, the result for time to NMOSD attack was assessed by comparing inebilizumab with each of satralizumab monotherapy and eculizumab. The results of the indirect comparison between inebilizumab and satralizumab were inconclusive due to wide 95% CIs that cross the null value and suggest imprecision (HR = 0.666; 95% CI, 0.182 to 2.435). The risk of an NMOSD attack trended toward being higher with inebilizumab than with eculizumab (HR = 3.947; 95% CI, 0.917 to 17.0), which agreed with the published NMA in terms of the direction of effect.

The sponsor also submitted unanchored IPD analyses comparing inebilizumab with rituximab for the outcomes of annualized attack rate (AAR) and EDSS. For AAR, the results lacked 95% CIs and could not be interpreted due to missing critical information in the reporting of methodology and results. For EDSS, no relative effect estimates were reported, so the results could not be interpreted.

Table 2: Summary of Findings for Inebilizumab Versus Placebo for Patients With AQ4P-IgG+ NMOSD in the N-MOmentum Study

AC = adjudication committee; AQP4-IgG+ = anti-aquaporin-4 immunoglobulin G seropositive; CI = confidence interval; CS = component score; EDSS = Expanded Disability Status Scale; HR = hazard ratio; HRQoL = health-related quality of life; KM = Kaplan-Meier; LS = least squares; MID = minimal important difference; NA = not applicable; NMOSD = neuromyelitis optica spectrum disorder; NR = not reported; NRS = numerical rating scale; OR = odds ratio; RCT = randomized controlled trial; RR = rate ratio; SAE = serious adverse event; SD = standard deviation; SE = standard error; SF-36 = Short Form (36) Health Survey.

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

^aRated down 1 level due to early cessation of the trial. Relatedly, there was a high degree of censoring that was imbalanced between treatment arms. However, the reasons for censoring did not appear concerning, so the CADTH review team judged that it was not likely to present an additional serious concern for bias; ergo, it was not rated down a second time. No threshold of importance could be determined, and the experts consulted by CADTH indicated that any benefit is clinically important, so potential benefits and harms were determined relative to the null value.

^bNo minimally important between-arm difference could be established, so the optimal information size approach was used to rate down 1 level in imprecision.

^cRated down 2 levels for very serious imprecision because the CI included both potential benefit and potential harm. No MID was identified so potential benefits and harms were determined relative to the null value.

^dRated down 1 level for serious imprecision because the CI included both potential benefit and potential harm. No threshold for clinically important differences was identified and the experts consulted by CADTH indicated that any benefit is clinically important, so potential benefits and harms were determined relative to the null value. The time frame of the study may be inadequate to determine a clinically meaningful difference in hospitalizations over a patient’s life, so the certainty was rated down 1 additional level for indirectness.

^eNo 95% CIs or between-group differences were reported for SF-36.

^fRated down 1 level for serious imprecision because the CI included both potential benefit and potential harm. No minimally important difference was established so potential benefits and harms were determined relative to the null value. Note that this outcome was not controlled for multiple comparisons.

^gRated down 1 level for serious concerns regarding imprecision; no 95% CI of the difference was available, so the optimal information size approach was used to judge imprecision. No minimally important threshold of difference was established, but the CADTH review team judged that the effect estimate might include an important between-group difference.

Sources: Details included in the table are from the sponsor’s summary of clinical evidence and from the sponsor’s responses to additional information requested by CADTH.^26,27

Harms Results

No harms outcomes were assessed in the ITCs.

Critical Appraisal

In the MAICs comparing inebilizumab with satralizumab monotherapy and eculizumab, there was unresolved between-trial heterogeneity with respect to patient populations and outcome definitions that were not mitigated by the MAIC methodology. Additionally, the factors selected for adjustment were not informed by clinical expert opinion or the literature regarding important treatment-effect modifiers in NMOSD. The MAICs may have been over-adjusted for clinically unimportant factors that were selected in an inappropriate manner, without consulting the literature or a clinical expert, and were based only on a statistical analysis of the N-MOmentum trial. The results of both comparisons had wide 95% CIs, suggesting substantial imprecision in the effect estimates, as well as small effective sample sizes, which results in the effects being overly influenced by small subgroups of patients and highlights poor overlap.

The submitted methodology for the MAIC analyses of inebilizumab versus rituximab is insufficient for critical appraisal. The 4 studies informing the rituximab data were small observational studies (N ≤ 32) in geographic regions likely to differ from Canadian demographics and clinical practice. The sponsor noted it was not feasible to conduct ITCs comparing inebilizumab with other off-label treatments such as azathioprine and mycophenolate mofetil due to small sample sizes and the use of observational study designs. However, the same limitations exist for the available rituximab data, and the use of the MAIC methodology cannot correct for these limitations. Despite having access to the IPD for the rituximab studies, the sponsor-submitted MAICs only adjusted the N-MOmentum trial data to reflect the rituximab-treated populations, which are less similar to the population of patients in Canada. Additionally, no relative effect estimates were reported for EDSS, and no 95% CIs were reported for AAR. There was no justification provided for why the sponsor selected single-arm observational studies to inform the ITCs instead of using an available published, placebo-controlled, double-blind, randomized trial comparing rituximab with placebo in seropositive NMOSD.²⁹ No conclusions can be drawn from the indirect comparisons with rituximab.

In all of the submitted MAICs, the submitted technical reports were missing critical details of the methods and results, which limited our ability to appraise the evidence and raised concerns about the validity of the analyses.

No safety-related outcomes were assessed.

There are no direct or indirect data available for the efficacy and safety of inebilizumab compared with azathioprine, mycophenolate mofetil, or ravulizumab.

Studies Addressing Gaps in the Evidence From the Systematic Review

No additional studies were submitted to CADTH.

Conclusions

NMOSD is a rare, disabling, and life-threatening inflammatory disorder of the CNS characterized by acute attacks with a relapsing pattern that cause potentially irreversible damage to the optic nerve and spinal cord. There is a large unmet need for high-efficacy, well-tolerated therapies for NMOSD that have a significant impact on preventing and/or reducing attacks. Patients and clinicians highlighted that the main treatment goals include the use of an efficacious, safe, and tolerable therapy administered as soon as possible after the first attack to avoid all relapses, reduce the severity of attacks and the cumulative disability associated with them, and minimize AEs related to therapies. One phase II and III, double-blind, randomized, placebo-controlled study (N-MOmentum) was included in this review, which had a duration of 197 days followed by an open-label, single-arm period of at least 2 years.

The data that were submitted to CADTH and the end points assessed in the study were considered clinically relevant for the treatment of patients with NMOSD. The results from the randomized period of the N-MOmentum trial demonstrated that treatment with inebilizumab likely results in a clinically meaningful benefit in time to first relapse and the proportion of patients with a worsening in EDSS score compared with placebo at 197 days. Other secondary or exploratory outcomes were less conclusive and did not show a clearly meaningful benefit of inebilizumab over placebo, including change in low-contrast visual acuity score, NMOSD-related inpatient hospitalizations, HRQoL, and pain. The reasons for this incongruence are uncertain, but it may be that the duration of the randomized study was insufficient to detect clinically meaningful differences in these outcomes.

The ITCs included MAICs comparing inebilizumab with satralizumab and eculizumab, and IPD analyses comparing inebilizumab with rituximab. Results were inconclusive for the comparison with satralizumab. The ITCs suggest the risk of an NMOSD attack is higher with inebilizumab treatment than with eculizumab treatment, but the magnitude of benefit is uncertain due to wide 95% CIs, imprecision, and unresolved between-trial heterogeneity. The comparison with rituximab could not be interpreted due to limitations in the analysis inherent to the data available, as well as sparsely reported methodology and results and inappropriate methodological decisions. There are no direct or indirect data available for the efficacy and safety of inebilizumab compared with azathioprine, mycophenolate mofetil, or ravulizumab.

Relative to previous studies in the clinical development program for inebilizumab, no new safety signals or concerns with the drug were identified in either study period, and long-term inebilizumab treatment was well tolerated after a median exposure of more than 3.2 years during the study. Almost all patients experienced at least 1 AE during the study regardless of treatment assignment, but the results suggested that patients treated with inebilizumab may have fewer SAEs than patients treated with placebo at 197 days. Cytopenias were more common in patients treated with inebilizumab compared with placebo, which is consistent with inebilizumab’s mechanism of action and the class effects of B-cell depletion.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of inebilizumab 300 mg IV in the treatment of adults with NMOSD in who are seropositive for autoantibodies against AQP4-IgG.

Disease Background

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

NMOSD is a rare, life-threatening, and chronic inflammatory disorder of the CNS that is characterized by acute attacks or relapses that cause potentially irreversible damage to the optic nerve and spinal cord. A defining feature of NMOSD is the presence of pathogenic serum autoantibodies against AQP4, which is detected in about 80% to 90% of NMOSD patients and which led to its differentiation from MS.^2-4 The clinical presentation typically involves inflammation of the optic nerve (optic neuritis) and/or spinal cord (transverse myelitis), but some cases may involve diencephalic, brainstem, and cerebral hemisphere attacks.¹ The consequences of NMOSD extend beyond clinical settings and include physical, functional, and psychological effects, causing loss of independence, reduced ability to work or perform activities of daily living, reduced sexual function, and reduced emotional well-being due to impaired HRQoL.^9,14-16

NMOSD attacks are unpredictable, lasting from days to weeks, and recovery from attacks is often incomplete, leading to an accumulation of morbidity, including permanent visual and motor disabilities such as blindness and/or paralysis.^5-9 Even a single attack can be so severe that, despite rescue treatment, patients may lose their ability to walk without assistance or become functionally blind in at least 1 eye. For example, almost 80% of patients will experience legal blindness (visual acuity ≤ 0.1) during their first NMOSD attack.^6,7 Unilateral or bilateral optic neuritis, including central vision loss with ocular pain, is often the initial event of relapsing NMOSD. Clinical manifestations of myelitis may include limb weakness, quadriplegia or paraplegia, sensory loss, bladder dysfunction, spasms, and pain.¹³ Brainstem involvement may manifest with nausea, vomiting, hiccups, vertigo, hearing loss, facial weakness, trigeminal neuralgia, diplopia, ptosis, or nystagmus; myelitis that extends into the brainstem may cause respiratory failure and death.^3,11,13 Neurologic deficits in patients with NMOSD accumulate after each inflammatory attack;^30,31 76% of patients have incomplete recovery after their first relapse, and 100% of patients have incomplete recovery after their sixth relapse.^7,32 At least 90% of NMOSD patients suffer from recurrent attacks,³³ and 94% of patients experienced 3 to 4 attacks over a median period of 5 years, according to 1 study.^34,35 As a result, prevention of relapse is the key goal of therapy and paramount in the overall management of patients with NMOSD to avoid increasing neurodegeneration and functional disability.^3-5

Estimated mortality rates due to NMOSD have varied considerably, from 7% to 32%, depending on age, relapse rate, and recovery from attacks.^31,36,37 A retrospective US-based study estimated an overall mortality rate of 7% and an annual mortality rate of 0.68 deaths per 100 patient-years for patients with NMOSD, with a mean disease duration of 6.9 years at the time of death.³⁶ In a study of 63 patients with late-onset NMOSD (≥ 50 years of age) from France, Germany, Turkey, and the UK, with a mean follow-up period of 4.6 years, the annual mortality rate was 2.8% and the 5-year survival rate was 86.7%; older age at onset and the AAR have independently predicted death.³⁷

NMOSD disproportionately affects females, with a reported female-to-male ratio of 9:1 to 12:1 in patients with AQP4-IgG–seropositive NMOSD.^5,17 No Canadian epidemiological studies specific to NMOSD are currently available, although systematic reviews based on data from several countries have estimated 0.053 to 0.4 incident cases per 100,000 people and 0.51 to 4.4 prevalent cases per 100,000 people.^18,19

In Canada, NMOSD is diagnosed using the criteria developed by the International Panel for Neuromyelitis Optica Diagnosis⁵ in patients with AQP4-IgG–seropositive disease and 1 or more of 6 core clinical characteristics (i.e., optic neuritis, acute myelitis, area postrema syndrome, acute brainstem syndrome, symptomatic narcolepsy or acute diencephalic clinical syndrome with NMOSD-typical MRI lesions, or symptomatic cerebral syndrome with NMOSD-typical brain lesions).⁵

Standards of Therapy

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

Currently, there is no cure for NMOSD. The goals of treatment for NMOSD relate to 3 broad areas: the prevention of relapses, the treatment of relapses, and the treatment of residual symptoms following an episode. The focus of this review is on the prevention of relapses based on the indication and place in therapy for inebilizumab. According to the clinician group input submitted to CADTH, relapse prevention is an important goal due to the permanent, life-altering disability that often arises from an attack despite acute management of attacks. Secondary goals of preventive therapy include reducing the severity of attacks, thereby reducing cumulative disability associated with attacks, and minimizing the AEs associated with therapy.

Current preventive treatment of NMOSD in Canada involves off-label therapies, including corticosteroids, azathioprine, mycophenolate mofetil, and rituximab, although public funding of these therapies varies by province and the clinical data for these treatments in NMOSD are absent or limited. Rituximab may exert its therapeutic effect on patients with NMOSD through B-cell–mediated humoral immunity,^38,39 and was shown to be superior to azathioprine for NMOSD in 1 open-label RCT³⁸ and superior to placebo in 1 Japanese double-blind RCT.²⁹ Azathioprine is a purine analogue that interferes with the DNA synthesis of rapidly proliferating cells and has been widely used as a first-line immunosuppressant medication for autoimmune diseases.⁴⁰ Azathioprine was first studied in 1998 in patients with NMOSD, where it was found to have a benefit in reducing disability,⁴⁰ but gastrointestinal and hematological side effects are associated with its use. Mycophenolate mofetil was developed to be a specific immunosuppressant drug with limited side effects by targeting guanosine more than adenosine and is used widely for the treatment of autoimmune diseases and NMOSD, with more adverse effects than azathioprine.^40,41

Two biologics, eculizumab and satralizumab, were recently approved in Canada for the treatment of patients with NMOSD who are AQP4-IgG seropositive. Eculizumab was the first drug approved in Canada for the prevention of NMOSD relapses based on data from a phase III RCT in adults with AQP4-IgG–positive NMOSD.⁴² In August 2020, the CADTH Canadian Drug Expert Committee (CDEC) recommended that eculizumab be reimbursed for the treatment of adult patients with AQP4-IgG–positive NMOSD if specific clinical and pricing criteria were met.⁴³ However, negotiations with the pan-Canadian Pharmaceutical Alliance concluded without agreement, so it is assumed that eculizumab is not a benefit on any public plans across Canada. In April 2021, CDEC recommended that satralizumab be reimbursed for the treatment of AQP4-IgG–seropositive NMOSD in patients aged 12 years and older who have had an inadequate response to at least 1 other preventive treatment, provided specific clinical and pricing criteria have been met, but public coverage of satralizumab varies by province.⁴⁴ The input received from patient groups indicated that financial barriers to access were a major concern among patients with NMOSD. Additionally, the clinical experts consulted by CADTH highlighted that reliance on private insurance is difficult in a population with high levels of unpredictable and accumulating disability that may prevent them from working as a result of their NMOSD. In addition to financial barriers to access, the patient and clinician group input and the clinical experts consulted by CADTH noted that the administration schedule of eculizumab requires infusions every 2 weeks, which is a practical barrier that may disproportionately affect patients living in rural areas, patients with disabilities as a result of NMOSD-related motor or visual difficulties who therefore cannot drive or use transit independently, and patients who otherwise have limited access to specialist treatment facilities.

Another monoclonal antibody, ravulizumab (Ultomiris), is also under review by Health Canada and CADTH for preventive treatment in adult patients with AQP4-IgG–seropositive NMOSD.

Drug Under Review

Key characteristics of inebilizumab and other treatments available for NMOSD are summarized in Table 3.

Inebilizumab is a humanized afucosylated monoclonal antibody treatment for NMOSD.^20-22 The precise mechanism by which inebilizumab exerts its therapeutic effects in NMOSD is unknown but is presumed to involve binding to CD19 on pre-B and mature B lymphocytes, which leads to antibody-dependent cellular cytolysis.²³ By depleting circulating CD19-positive B cells as well as AQP4-IgG–producing plasmablasts and plasma cells, inebilizumab acts upstream in the disease process by downregulating the production of AQP4-IgG, which represents a new mechanism to target the pathways that contribute to the pathogenesis of NMOSD.²

Inebilizumab has a Health Canada indication as monotherapy for the treatment of adult patients with NMOSD who are AQP4-IgG seropositive.²³ The reimbursement request is as per the approved indication. Treatment should be administered under the supervision of a qualified health care professional. The recommended dosage for inebilizumab is an initial 300 mg dose via IV infusion followed 2 weeks later by a second 300 mg dose via IV infusion; subsequent doses (starting 6 months from the first infusion) are administered as single 300 mg doses via IV infusion every 6 months.²³ The prepared infusion solution should be at room temperature before starting the infusion. Inebilizumab should be administered under the close supervision of an experienced health care professional with access to appropriate medical support to manage potential severe reactions, such as serious infusion reactions.²³

Although there are several comparators listed in Table 3, there are important caveats for each. Ravulizumab is not yet approved by Health Canada. The clinical trials of satralizumab and eculizumab were concurrent with the clinical trial for inebilizumab, which precluded using either therapy as an active control. Rituximab, azathioprine, and mycophenolate mofetil are the only listed comparators that were marketed in Canada at the time of the initiation of the inebilizumab phase III clinical trials, but each is used off-label, without a Health Canada–approved indication in NMOSD and with little to no robust clinical data supporting their use in NMOSD; additionally, based on the clinical expert input, azathioprine and mycophenolate mofetil are not considered highly effective therapies in NMOSD.

Table 3: Key Characteristics of Inebilizumab, Ravulizumab, Satralizumab, Eculizumab, Rituximab, Azathioprine, and MMF

AQP4 = aquaporin-4; AQP4-IgG = anti-aquaporin-4 immunoglobulin G; IST = immunosuppressive therapy; MMF = mycophenolate mofetil; NMOSD = neuromyelitis optica spectrum disorder; RA = rheumatoid arthritis; SC = subcutaneous.

^aHealth Canada–approved indication.

Sources: Product monographs for inebilizumab (Uplizna),²³ ravulizumab (Ultomiris),⁴⁵ satralizumab (Enspryng),⁴⁶ Eculizumab (Soliris),⁴⁷ rituximab (Rituxan),⁴⁸ azathioprine (Imuran),⁴⁹ and MMF (Myfortic⁵⁰ and CellCept⁵¹).

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 inputs received by CADTH have been included in the stakeholder section of this report.

Two patient groups, MS Canada and TSF, responded to CADTH’s call for patient input for the current review of inebilizumab as a monotherapy for the treatment of adult patients with NMOSD who are AQP4-IgG seropositive.

MS Canada gathered information for this submission through a survey that was conducted from August 4 to 14, 2023, that targeted Canadians living with NMOSD and their caregivers. The MS survey included data from 13 respondents; most of them were female (83%) and ranged in age from 25 years to older than 65 years.

TSF gathered information through 2 online surveys, 1 for patients and 1 for caregivers, between July 13 and 31, 2023. Additionally, TSF gathered information through videoconference interviews with 4 patients and 1 caregiver on August 8, 2023; 1 interviewee had direct experience with inebilizumab. The TSF survey included data from 51 patients and 9 caregivers from around the world, including Canada. Around 54% of the respondents were between 35 and 54 years of age, approximately 32% were between 55 and 74 years of age, and 25% of the patients were between the ages of 25 and 34 years.

The 2 patient groups indicated that NMOSD is a rare autoimmune syndrome of the CNS characterized by a relapsing-remitting disease course. NMOSD attacks are also called relapses, flare-ups, or exacerbations, and may include optic neuritis (affecting eye function), transverse myelitis (affecting limb function), and/or area postrema syndrome (episodes of otherwise unexplained hiccups or nausea and vomiting). Attacks may result in permanent neurologic damage and disability. Symptoms may include loss of vision, numbness, mobility issues, chronic pain, muscle cramps or painful spasms, fatigue, bowel and bladder dysfunction, and difficulty falling and staying asleep. The disease has a tremendous impact on all aspects of patients’ and caregivers’ lives, including a negative effect on their quality of life, their independence and employment, and on their social, family, and school life. Patients with NMOSD have a substantially elevated risk of mortality compared with the general population, and NMOSD-related deaths are typically attributed to respiratory failure following an attack. The patient inputs estimated that within the first 5 years of the disease, approximately 30% of NMOSD patients die and approximately 50% become either blind and/or require the use of a wheelchair.

The patient inputs stated that the standard treatment for NMOSD may involve IV steroids, IVIG, or plasmapheresis or plasma exchange. Off-label immunosuppressants may be used to help prevent further attacks, with varying levels of therapeutic benefit. Recently, 2 medications were approved by Health Canada with indications for the treatment of adults with AQP4-IgG–seropositive NMOSD (eculizumab and satralizumab), but access to these medications is limited. Symptoms such as neuropathy, pain, stiffness, muscle spasms, and bladder and bowel control problems can be managed with various medications and therapies. Many patients suffer significant additional attacks and additional disability while cycling through different therapies, and may experience challenging side effects of treatment.

Some respondents reported treatment experience with eculizumab, satralizumab, or rituximab, and felt these medications were effective or somewhat effective in managing their NMOSD. There were numerous barriers to access associated with these therapies, including out-of-pocket costs, lack of insurance coverage and, for eculizumab, scheduling time-consuming infusions too often and planning life around these appointments. According to the patient input groups, there is an unmet need for accessible therapy options that are able to reduce the risk of future attacks, maintain the current level of physical ability, and slow disease progression.

One respondent from the MS survey and the 5 patients and 2 caregivers who shared their experience using inebilizumab in the TSF input indicated that using inebilizumab has effectively reduced attacks, improved their quality of life and, because of the drug’s easy-to-manage schedule and manageable side effects, they could be involved with their family, friends, and workplace.

Clinician Input

Input From the Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise regarding the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, 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 NMOSD.

Unmet Needs

The clinical experts indicated that treatment of NMOSD includes a series of treatment goals relating to 3 broad areas: prevention of relapses (disease modification), treatment of relapses, and treatment of residual symptoms. Although ideal, it is unlikely that any single treatment would cover all 3 areas. Of these 3 areas, preventive treatment was described as being of special interest because relapses are the major source of disability accumulation for people with NMOSD. As a result, preventive treatment is expected to result in downstream desirable effects, including better HRQoL, increased ability to maintain independence and employment, and a reduced reliance on caregivers. The clinical experts highlighted that it is important to control progression as early as possible because damage leading to neurologic disability, including paralysis and blindness, may be irreversible after an attack.

The clinical experts agreed that most patients with NMOSD still have relapses despite their current treatment regimens, and current first-line therapies (e.g., azathioprine and mycophenolate mofetil) are not considered particularly effective in preventing NMOSD attacks. Additionally, these therapies are also associated with significant adverse effects, especially if used in conjunction with corticosteroids. The greatest unmet need in the treatment of NMOSD is for therapies that prevent relapses more effectively without intolerable side effects.

Place in Therapy

The clinical experts agreed that current clinical practice for treatment and attack prevention in NMOSD is suboptimal and inconsistent due to the low efficacy of the off-label treatments used in the first-line setting and barriers to treatment access, which vary by province. The newer therapies with indications for NMOSD, such as satralizumab and eculizumab, present especially significant barriers to access for most patients due to inconsistent coverage or the lack of public reimbursement, and the onerous dosing schedule of eculizumab. Due to the impact of NMOSD on a person’s ability to maintain employment, patients with NMOSD are more likely to lack private insurance.

As the vast majority of patients with NMOSD will relapse and relapses may lead to permanent disability or death, all people with NMOSD should be on relapse-preventive treatment.

The experts agreed that because the impact of relapses on people with NMOSD is devastating and preventing as many relapses as possible is critical to the prevention of significant disability, inebilizumab should be available as first- and later-line treatment for patients diagnosed with NMOSD who are AQP4-IgG seropositive. The experts indicated that inebilizumab would be used as a monotherapy.

Patient Population

NMOSD is a rare disease. The clinical experts indicated that patients diagnosed with NMOSD who are seropositive for AQP4-IgG should be candidates for treatment with inebilizumab. It is the standard of care in Canada to assess patients with NMOSD for AQP4-IgG; the experts noted there are no major challenges in this regard in Canada, other than that the test results may be delayed in some locations. The specificity of AQP4-IgG is very high, so the risk of misdiagnosis is very low.

The patient eligibility criteria in the pivotal study, N-MOmentum, were considered by the experts to be broadly representative of patients with NMOSD in Canada, with the exception that the study excluded patients with recent steroidal treatment. Although appropriate from the perspective of a clinical trial design, in real-world practice there are many comorbidities that may require steroid treatment, and these patients should not necessarily be excluded from treatment with inebilizumab; such a decision would need to be considered by the expert clinician managing a patient’s particular case. Patients who have received IVIG or have concomitant diseases should also not necessarily be excluded from receiving inebilizumab in the real world.

Within the population of patients who have NMOSD and are AQP4-IgG seropositive, it is unknown which patients are more likely to benefit from inebilizumab.

It is possible that patients who are seronegative for AQP4-IgG may also benefit from inebilizumab. Fulfillment of seronegative NMOSD criteria would be necessary to establish the diagnosis to allow appropriate access.

Assessing the Response to Treatment

The experts indicated that a clinically meaningful response to treatment relates to the reduction of the relapse rate and prolongation of times to relapse. Although the absence of relapse is indicative of a clinically meaningful response, this may not be realistic, as the number and severity of relapses that patients experience differ on an individual level (e.g., some patients may have several relapses per year); thus, a reduction in the number of relapses is still a reasonable goal. The determination of relapses is fairly objective; however, it is not the only factor, and assessment of treatment response is based on a combination of patient-reported symptoms, a clinical exam, clinical tools, and patient history. Other important outcome measures include evaluation of attack severity and degree of recovery from attack, as well as accumulation of disability.

The experts noted there is a lack of formal guidance on how to assess treatment response, but it would be reasonable to assess initial treatment response 3 months after the initial infusion then every 6 months until stability is achieved, and then every year for patients with stable NMOSD. However, it was noted that within the first 6 months of treatment, the attack rate may still be higher than when stability is achieved.

MRIs are not routinely conducted for patients with NMOSD outside of initial diagnosis and so would not be used in assessing the response to treatment.

Discontinuing Treatment

Discontinuation of treatment should occur if the patient is completely dependent and unable to leave their bed (EDSS score of 9.0 or above).

Discontinuation of treatment should be considered on a case-by-case basis in the event of a severe relapse (e.g., requiring intubation and support on a ventilator) or if the patient has experienced 2 or more relapses within 2 years, has severe or unacceptable AEs, or has contraindications for therapy.

Prescribing Considerations

The clinical experts agreed that treatment should be supervised by a neurologist with expertise in this area. Although NMOSD and MS are not the same disease, the populations and medications are similar and patients with NMOSD are often cared for in MS clinics. The clinical experts stated that neurologists with experience or expertise in related subspecialties should prescribe inebilizumab, including clinicians with experience in MS neurology, and/or neurologists working in an MS clinic or in neuroimmunology, autoimmune neurology, and/or neuro-ophthalmology. However, patients in remote areas may have issues with access to subspecialists. For patients living in remote areas, local neurologists without subspecialty expertise could work by distance in conjunction with neurologists who are experts in a relevant subspecialty.

Inebilizumab is expected to be used as a monotherapy and not combined with other monoclonal antibodies indicated for the treatment of NMOSD. However, there may be situations in which it is combined with classical immunosuppressants. There is a lack of data regarding combination therapies.

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 received by CADTH has been included in the stakeholder section of this report.

One clinician group, the CNMSC (input authored by 1 clinician), responded to CADTH’s call for clinician group input.

Clinician perspectives from the CNMSC were obtained through clinical experience, knowledge of the medical literature, and from clinicians across the country who specialize in this therapeutic area.

According to the clinician group, a variety of off-label therapies are used for the treatment of NMOSD in Canada, including corticosteroids, azathioprine, mycophenolate mofetil, and rituximab. However, breakthrough NMOSD attacks are reported for all of these drugs, and government drug program funding varies by province and territory. Recently, 2 monoclonal antibodies, eculizumab and satralizumab, were approved by Health Canada. However, access to these therapies is extremely limited due to their high cost and stringent funding criteria. All of the therapies in use for NMOSD work by suppressing the immune system to prevent attacks, with variable efficacy. Failure of treatment, resulting in even just 1 relapse, can lead to a profound, permanent disability, including blindness and paralysis.

As per the CNMSC, there is a large unmet need in Canada for high-efficacy, well-tolerated therapies for NMOSD that have a significant impact on preventing and/or reducing attacks. Use of some off-label therapies is limited by many side effects, and many patients continue to have attacks despite treatment with drugs such as azathioprine and mycophenolate mofetil and, to a lesser extent, rituximab. Also, eculizumab is given by IV infusion every 2 weeks, which is too onerous for some patients to tolerate.

According to the clinician group, the main treatment goals include the use of an efficacious, safe, and tolerable therapy administered immediately after the first attack to ideally avoid all future relapses, reduce the severity of attacks and the cumulative disability associated with them, and minimize AEs related to therapies. In particular, there is a major unmet need among patients who have a breakthrough attack on their first therapy, as it can be challenging to identify a subsequent therapy that will be effective at preventing attacks and will be tolerated by the patient. The best approach for patients is to use a product after an attack that is as highly efficacious as possible so as to avoid potentially catastrophic subsequent attacks and, thus, optimize patient outcomes. The clinician group input noted that inebilizumab could be used as first-line treatment and as subsequent treatments for patients who have had breakthrough attacks on other therapies or who were intolerant of other therapies. Inebilizumab would be expected to be used as a monotherapy based on the available clinical evidence and to avoid cumulative immunosuppressive effects. The clinician group also noted that, although rituximab and inebilizumab both suppress B cells, there is some evidence that patients with polymorphisms in the FCGR3A gene may have an incomplete response to rituximab but not to inebilizumab.⁵² There is a lack of head-to-head data comparing inebilizumab and rituximab. The CNMSC also noted there is no clear preferred drug among the novel monoclonal antibodies for the treatment of NMOSD (e.g., eculizumab, inebilizumab, ravulizumab, and satralizumab) and that the best mechanism of action may vary by patient; however, there is generally limited access to these therapies in Canada at this time.

According to the CNMSC, the key outcome measure is a new attack, which is marked by new neurologic symptoms such as vision loss, weakness, sensory impairment, or dysfunction of the bladder or bowel. Although usually marked by a new enhancing lesion on MRI, this is not necessary to diagnose an attack. The clinician group indicated that the drug renewal process should consider the occurrence of any relapse in the previous year and the number of relapses, EDSS score or results of other disability measures, and any change from baseline (note that the EDSS is not validated in NMOSD). The CNMSC recommended discontinuing the drug if the patient has a new attack, a serious AE related to the therapy, or an EDSS score of 8 or higher.

The CNMSC stated that the treatment of patients with NMOSD should be assessed and managed by neurologists specialized in demyelinating diseases through an MS or demyelinating disease centre, and inebilizumab can be administered in a hospital or private clinic. Patients eligible for treatment with inebilizumab should have a confirmed diagnosis of NMOSD and a positive serum test for AQP4-IgG.

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

CDEC = Canadian Drug Expert Committee; EDSS = Expanded Disability Status Scale; MS = multiple sclerosis; NMOSD = neuromyelitis optica spectrum disorder.

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 inebilizumab IV infusion in the treatment of adult patients with NMOSD who are AQP4-IgG seropositive. The focus will be placed on comparing inebilizumab with relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of inebilizumab is presented in 4 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 indirect evidence from the sponsor.

Included Studies

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

• 1 RCT with an open-label single-arm period identified in a systematic review

• 1 ITC.

Systematic Review

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

Description of Studies

Characteristics of the included studies are summarized in Table 5.

Table 5: Details of the Included Study (N-MOmentum)

AC = adjudication committee; AQP4-IgG = anti-aquaporin-4 immunoglobulin G; EDSS = Expanded Disability Status Scale; Gd = gadolinium; IDMC = independent data-monitoring committee; IVIG = IV immunoglobulin; LLN = lower level of normal; MCS = mental component score; MS = multiple sclerosis; NMO = neuromyelitis optica; NMSOD = neuromyelitis optica spectrum disorder; OLP = open-label period; PCS = physical component score; RCP = randomized controlled period; SF-36 = Short Form (36) Health Survey.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.

N-MOmentum was a multicentre, international, double-blind, placebo-controlled study with a 197-day RCP to compare the efficacy and safety of inebilizumab versus placebo, followed by a minimum 2-year single-arm OLP. Patients who experienced an NMOSD attack, as identified by a blinded AC, or who completed the day 197 visit without an attack, exited the RCP and had the option to enrol into an OLP that lasted a minimum of 2 years for each patient. During the OLP, patients initiated or continued treatment with inebilizumab in a manner that did not unblind their RCP treatment.

The primary objective of the study was to compare the efficacy of inebilizumab with placebo in reducing the risk of an NMOSD attack in adult patients diagnosed with NMOSD. A total of 467 patients entered the 28-day screening period, and 231 eligible patients were randomized 3:1 thereafter to receive either inebilizumab or placebo. Of the randomized patients, 213 had AQP4-IgG– seropositive NMOSD and 18 had seronegative NMOSD. Patients were stratified according to AQP4-IgG status (seropositive versus seronegative) and region (Japan versus non-Japan). The study included 99 investigational sites in 25 countries, including 1 site in Canada that enrolled 2 patients.

The purpose of the screening period was to establish patients’ eligibility to participate in the study based on the inclusion and exclusion criteria. AQP4-IgG serostatus was verified by a central laboratory. For patients with disease that was found in the screening period to be AQP4-IgG negative, relevant data documenting their neuromyelitis optica (NMO) disease were assessed by an independent eligibility committee to ensure the data were consistent with the diagnosis of NMO. Of the 467 patients who entered the study, 236 experienced screening failure. ||| ||||||| ||||| ||||||||| |||||||| ||||| ||||| |||||| |||||||||. The most common reason for screening failure (n = 204) was not meeting the inclusion or exclusion criteria. |||| |||||| ||||||| ||| ||| ||||||| ||| ||||||||||||||||||| |||||||| |||||||| |||| || ||||||| |||||||||||| ||||||| || ||||| ||| |||||||| ||| |||||||| ||||||| || |||||||||| |||||||||||| || ||| |||||| |||| |||||| ||||||||| |||||| ||||||||||||| |||||||| || ||||||||||||| |||||||||||| |||| |||||||| || |||||||||||| |||||| || |||||| ||||||||.

Patients were not allowed to receive other immunosuppressive therapies concomitantly during the N-MOmentum trial, which differed from the phase III study of eculizumab (PREVENT) and 1 of the 2 satralizumab phase III studies (the SAkuraSky trial evaluated combination therapy with immunosuppressants, while the SAkuraStar trial evaluated satralizumab monotherapy). A direct comparison of inebilizumab versus eculizumab or satralizumab was not possible due to their concurrent clinical development programs, and direct comparison with off-label treatments (rituximab, azathioprine, or mycophenolate mofetil) was considered by the sponsor to be infeasible due to uncertainty regarding their efficacy and safety.

The trial was designed with the intention of reducing risks to placebo-treated patients, including by establishing a relatively short duration for the placebo-controlled period (6.5 months), randomizing patients in a 3:1 ratio (allows for an enriched safety database while minimizing the number of required events and at-risk patients in the placebo arm), allowing patients to enter the OLP and transition to receive treatment after a single attack, providing close monitoring by an independent unblinded data-monitoring committee, and allowing liberal use of rescue therapies for patients who experienced an attack.

The N-MOmentum trial was stopped early on the recommendation of an independent data-monitoring committee, which found that the efficacy of inebilizumab had been established and that there was no justification to continue exposing placebo-treated patients to an increased risk of NMOSD attack. While still blinded to the study results, the sponsor offered all patients in the RCP the option to enter the OLP or exit to a safety follow-up period, per the recommendation of the independent data-monitoring committee. ||| ||||| ||||||| ||||||||| ||| ||||| ||| ||||| || |||||||| || |||||.

Figure 1: N-MOmentum Study Flow Diagram

MEDI-551 = inebilizumab; OLP = open-label period; RCP = randomized controlled period.

Source: Clinical Study Report for N-MOmentum.⁵³

Populations

Inclusion and Exclusion Criteria

Eligible patients were adults with NMOSD (AQP4-IgG seropositive or seronegative) who had a documented history of acute NMOSD attacks (1 or more attacks in the prior year or 2 or more attacks in the prior 2 years), required rescue therapy, and had an EDSS score at randomization of 7.5 or less, or 8.0 if the investigator and medical monitor agreed the patient was reasonably able to participate. Patients were excluded from the study if they had AQP4-IgG–seronegative NMOSD and a brain MRI abnormality that met the diagnostic criteria for MS, had an active infection, or had a known history of primary immunodeficiency, malignancy, or any concomitant disease other than NMOSD that required steroids at doses of greater than 20 mg/day for more than 21 days and that had been administered 6 months or less before screening. Patients were also excluded if they had received certain monoclonal antibodies or immunosuppressive drugs within 3 months before randomization; rituximab within 6 months before screening; IVIG within 1 month before randomization; or any of alemtuzumab, total lymphoid irradiation, bone marrow transplant, or T-cell vaccination therapy at any time.

Interventions

On days 1 and 15 at the start of the RCP, patients received either a 300 mg dose of inebilizumab or matched-administration placebo via a 90-minute IV infusion at the study site. Vials of inebilizumab and placebo were identically labelled and indistinguishable in appearance. After the first dose of inebilizumab or placebo during the RCP, all patients received a 2-week course of corticosteroids (prednisone 20 mg/day or equivalent) plus a 1-week taper to provide a measure of protection against attacks until the full pharmacodynamic effect of inebilizumab was reached, as all patients had discontinued prior immunosuppressive therapy before day 1.

During the OLP, all patients received inebilizumab. However, to maintain the blinding of the study, patients who were randomized to inebilizumab in the RCP received 300 mg IV inebilizumab on day 1 of the OLP, blinded IV placebo on day 15 of the OLP, then 300 mg IV inebilizumab every 26 weeks thereafter. Patients randomized to placebo in the RCP received 300 mg IV inebilizumab on day 1 of the OLP, blinded 300 mg IV inebilizumab on day 15 of the OLP, then 300 mg IV inebilizumab every 26 weeks thereafter.

Rescue therapy (e.g., IV corticosteroids, IVIG, and/or plasma exchange) was initiated as needed for NMOSD attacks at the discretion of the site investigator. The treatment of an attack was preferably initiated after completion of the attack assessments and the determination as to whether the protocol attack criteria were met. However, the investigator could initiate rescue therapy at any time before the full assessment was completed. Rescue therapy was given as directed by the investigator and could include IV corticosteroids, IVIG, and/or plasma exchange.

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. The summarized end points are based on the 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 the 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 the deliberations of the CADTH expert committee 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 the deliberations of CADTH’s expert committee were also assessed using GRADE.

Table 6: Outcomes Summarized From the N-MOmentum Trial

AC = adjudication committee; EDSS = Expanded Disability Status Scale; NMSOD = neuromyelitis optica spectrum disorder; NRS = numerical rating scale; SF-36 = Short Form (36) Health Survey.

^aStatistical testing of these end points was designed to control the overall 2-sided type I error at an alpha of 0.05 based on the Bonferroni chain procedure. The primary end point was first tested in the seropositive cohort and, if significant, further tested in the intention-to-treat population (both alpha = 0.05). If the difference was statistically significant in the intention-to-treat population, each secondary end point was tested via the same sequential strategy (first seropositive group then intention-to-treat population) at an alpha of 0.05 divided by 4 (0.0125).

Sources: Clinical Study Report for N-MOmentum.⁵³ Details included in the table are from the sponsor’s summary of clinical evidence.

Time to AC-Determined NMOSD Attack

The primary efficacy outcome in the N-MOmentum trial was time (days) from day 1 to the onset of an AC-determined NMOSD attack on or before day 197. This end point was chosen to reflect the most significant and expected clinical impact of treatment, which is reducing the risk of NMOSD attacks. The primary analysis included only AC-determined attacks with an assessment visit scheduled within 120 hours (5 days) of reporting symptoms ||| ||| ||||||||||| ||||||||| |||||| || |||| || || |||||||||| |||||. The AC consisted of 3 NMOSD experts (2 neurologists and 1 neuro-ophthalmologist) who performed blinded assessments and then determined by majority vote whether an event qualified as an NMOSD attack. Supportive analyses were also conducted using other AC-determined attacks that were not included in the primary analysis.

At the time of the study design, there was no widely accepted set of criteria for the diagnosis of an NMOSD attack. Therefore, an objective set of attack criteria was needed to ensure the uniform and consistent diagnosis of attacks across heterogeneous participating sites and diverse investigator diagnosis practices. The trial sponsor, working closely with a panel of NMOSD disease experts and with input from the FDA, developed a set of NMOSD attack criteria that recognizes attacks in all domains affected by NMOSD (optic neuritis, myelitis, brain, and brainstem). These criteria include substantial clinical manifestations alone, as well as MRI-augmented findings in cases with more modest clinical manifestations. The definition of an NMO or NMOSD attack was the presence of a new symptom(s) or worsening of an existing symptom(s) related to NMO or NMOSD that met at least 1 of the protocol-defined criteria for an NMO or NMOSD attack (refer to Table 7 for criteria).

As supportive analyses, secondary and exploratory outcomes related to attacks and attack rates are discussed in this report, but not formally assessed using GRADE. Attack severity and recovery were also considered as exploratory analyses in the N-MOmentum trial. Attack severity was graded using 2 scales: the Opticospinal Impairment Scale (OSIS) and a scale based on the 18 protocol-defined attack criteria. Higher scores on the OSIS indicate poorer functioning across domains of visual acuity (0 to 8 points), motor function (0 to 5 points), sensory function (0 to 5 points), and sphincter function (0 to 5 points). The second scale assigned severity categories (mild, moderate, or severe) based on the degree of change for the applicable criterion or criteria. Attack recovery was also graded based on the domain-specific change in OSIS scores and graded as major or minor.

Additionally, a descriptive summary of the rate of AC-determined NMOSD attacks over various lengths of exposure to inebilizumab was chosen as an end point in the N-MOmentum trial to provide an approximate rate of attacks on an annualized basis. The AAR was defined as the total number of AC-determined attacks divided by total person-years of inebilizumab exposure. |||||||| || |||||||||| |||||| |||| ||| ||| ||||||| ||||||||| |||||| ||||| ||| || |||||||||| ||||||| |||||||| |||| ||||||| |||| ||| |||||||||||||||||| ||||||| ||||| || |||||||||||||| ||||||| ||||| ||||| ||||||||| |||| ||| ||||||||||| |||||||||||| ||| |||||| ||||.

Table 7: Protocol-Defined Criteria for an NMOSD Attack in the N-MOmentum Trial

CF = counting fingers; EDSS = Expanded Disability Status Scale; FSS = functional system score; Gd = gadolinium; HM = hand motion; LP = light perception; NLP = no light perception; NMOSD = neuromyelitis optica spectrum disorder; ON = optic neuritis; RAPD = relative afferent pupillary defect.

^aThe symptoms listed are examples and are not inclusive of all NMOSD symptoms.

^bFour major areas of the body may be affected by an attack: the optic nerve, resulting in ON; the spinal cord, resulting in myelitis; the brainstem, resulting in a number of outcomes; and the brain.

^cAt least a 2-step drop (worsening) on the Landolt C broken ring chart (e.g., a drop from CF to LP or NLP, or from HM to NLP).

^dAt least a 1-step drop (worsening) on the Landolt C broken ring chart (e.g., a drop from CF to HM, LP, or NLP; from HM to LP or NLP; or from LP to NLP).

^eLesions seen in the optic chiasm also count toward these criteria.

^fA 1-point change in a single FSS without a change in the EDSS, with or without a new Gd-enhancing or new or enlarging T2 MRI lesion in the spinal cord, is not considered a clinically significant change and did not count as an attack per this protocol.

^gOther neurologic signs could include double vision, dysarthria, dysphagia, vertigo, oculomotor palsy, weakness, nystagmus, or other cranial nerve abnormality.

Sources: Clinical Study Report for N-MOmentum.⁵³ Details included in the table are from the sponsor’s summary of clinical evidence.

Proportion of Patients With Worsening in EDSS Score

One of the key secondary end points in the N-MOmentum trial was worsening in baseline EDSS score at last visit during the RCP. The EDSS is a validated tool for MS-related disability that ranges from 0 (normal neurologic exam) to 10 (death from MS), with half steps from 1.0 to 9.5; higher values reflect greater impairment and disability. Although MS and NMOSD are not the same condition, they share features in clinical presentation. No minimal important difference (MID) was identified that was specific to NMOSD, but estimates from patients with MS suggest that a single 1.5-point change in EDSS score could be considered a meaningful deterioration from the patient’s perspective,⁵⁴ and that important changes were considered, i.e., a change of 1.5 points or more from a baseline of 0 points, a change of 1.0 point or more from a baseline of 1 to 5.5 points, or a change of 0.5 points or more from a baseline of 6 points or greater.⁵⁵ The clinical experts consulted by CADTH agreed that patients with a higher baseline EDSS score would have a lower threshold for EDSS changes that would be considered clinically meaningful compared with patients who had a lower baseline EDSS score. In the N-MOmentum trial, trained and certified neurologists (EDSS raters), who were independent of study investigators and patients, conducted the assessments. A worsening in the overall EDSS score during the N-MOmentum study was defined based on the following criteria:

• a worsening in EDSS score of 2 or more points in patients with a baseline EDSS score of 0 points

• a worsening in EDSS score of 1 or more points in patients with a baseline EDSS score of 1 to 5 points

• a worsening in EDSS score of 0.5 points or more in patients with a baseline EDSS score of 5.5 points or greater.

Change From Baseline in Low-Contrast Visual Acuity Score

Another key secondary end point during the RCP was the change from baseline to the last visit in the binocular low-contrast visual acuity score measured using the Landolt C broken ring chart. This chart uses a standardized format to be equally detectable by observers with no sight impairment and comprises a series of incomplete rings resembling the letter C. The break in the ring can be oriented in any direction and the patient must indicate where the break is located for each ring when the chart is held at a distance of 3 m, with the result based on the smallest rings that can be correctly read on the chart. Trained, experienced, and blinded independent ophthalmologists or ophthalmology technicians conducted all examinations using standardized retro-illuminated light cabinets, which eliminated any inter-site variability in height and room lighting. An MID has not been identified for the change in the low-contrast visual acuity score in NMOSD. The low-contrast visual acuity score is commonly used in MS and the clinical experts consulted by CADTH indicated this is an appropriate outcome for patients with NMOSD, especially as a complement to the EDSS, which may be less sensitive to changes in visual acuity. The scoring of this assessment is based on the number of characters on the chart that the patient is able to identify, from 0 to 70 inclusive, where 70 indicates the patient was able to correctly identify all characters on the chart (i.e., best visual acuity score), and 0 indicates they were not able to identify any characters correctly (i.e., poorest visual acuity score).

Number of NMOSD-Related Inpatient Hospitalizations

The number of NMOSD-related inpatient hospitalizations was a key secondary end point. An NMOSD-related inpatient hospitalization was defined as a hospital stay |||| |||| |||||| |||||||| || ||| ||||| ||| || ||||||||| || lasted for more than 1 day ||||||||||| || ||| |||||||||| ||||||| ||||||||| |||| ||| ||||||||| |||||| |||| |||||||| |||| ||||| || || ||||||||| |||||||||| |||| |||||||||| ||||||||| |||||||||| ||||| ||| |||||||| |||||||||||||||| |||| |||| |||| ||| ||| |||||||||||||| || ||||||||||| ||||||||||| || ||||||||||.

Short Form (36) Health Survey

An exploratory end point was the change from baseline to the last RCP visit in the 4-week recall in the SF-36 physical component score and mental component score. The SF-36 measures 8 dimensions of a patient’s functional health and well-being and also provides 2 summary scores that characterize a patient’s mental and physical health status (the mental component score and physical component score, respectively), with higher global scores reflecting better HRQoL. The sponsor is not aware of a reported MID for the SF-36 specific to NMOSD. The SF-36 has been validated in patients with MS and neurologic disabilities.^56,57 Indirect evidence from patients with MS suggests that the MID ranges for the SF-36 domains are 4 to 9 points for physical functioning, 6 to 8 points for role physical, 6 to 7 points for social functioning, and 9 points for the physical component score.⁵⁷

Pain NRS

An exploratory end point was the change from baseline to the last RCP visit in the pain NRS in 5 anatomic locations (eyes, upper back, lower back, arms, and legs). Patients were asked to rate pain at each location |||| ||| ||||||||, with scores ranging from 0 (no pain) to 10 (worst pain imaginable). No MID for the NRS specific to NMOSD was identified.

Safety Outcomes

Safety data were collected and recorded from the time of enrolment, through the treatment periods, and through the end of the safety follow-up period. TEAEs were reported separately for each of the treatment periods and according to patient disposition. However, because they were already reported as the primary outcome, NMOSD attacks were not recorded as TEAEs during the RCP and OLP, |||||||| ||||| ||||||| |||| |||||||| || ||||| |||||| ||| ||||||||| ||||||| |||||| ||||||||| ||||||| || || ||| ||||||||| |||||||| ||||||| ||| ||||| ||| |||||||||| ||| ||| |||||||| ||||||||||). All TEAEs were coded according to system organ class and preferred term using the Medical Dictionary for Regulatory Activities (version 21.0). The severity of TEAEs was assessed by the investigator and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events. AEs of special interest were defined as:

• infusion-related reactions, anaphylactic reactions, and hypersensitivities

• infections

• hepatic function abnormalities

• cytopenias

• progressive multifocal leukoencephalopathy.

Table 8: Summary of Outcome Measures and Their Measurement Properties

HTA = health technology assessment; HRQoL = health-related quality of life; MCS = mental component score; MID = minimal important difference; mRS = modified Rankin Scale; MS = multiple sclerosis; NMOSD = neuromyelitis optica spectrum disorder; NRS = numerical rating scale; PCS = physical component score; SF-36 = Short Form (36) Health Survey.

Statistical Analysis

End Points

The primary end point (time to the onset of an NMOSD attack) was assessed using a Cox proportional hazards model in both the population with AQP4-IgG–seropositive NMOSD and the ITT population. The sponsor did not report any assessment of whether the proportional hazards assumption was appropriate. || |||||||||| ||| ||||||||| ||| ||||||| |||| ||| ||| ||||||| ||||||||| |||||||| |||| |||||||| ||||||||| || ||||| ||||||| ||||||||||||| |||||||||| || |||||| ||||||||| ||||||||.

Worsening in EDSS score was assessed using a logistic regression model, with treatment and baseline EDSS as explanatory variables. The percentage of patients meeting the end point and the 95% CIs and P values for the odds ratios are presented. Missing values due to dropout or missing data were conservatively considered as “worsening” according to the nonresponder imputation rule.

Change from baseline in low-contrast visual acuity was assessed using an analysis of covariance model that used treatment and baseline Landolt C broken ring chart binocular score as explanatory variables. The last nonmissing value was used for the analysis in instances of missing values due to dropout or missing data.

Sample Size and Power Calculation

The N-MOmentum trial was planned to detect a reduction in relative risk (RR) of 60% in the primary outcome (time from day 1 to the onset of an AC-determined NMOSD attack on or before day 197) with 90% power and a 2-sided alpha of 0.05. A total of 67 AC-determined NMOSD attacks were assumed for the ITT population. Patients were randomized in a 3:1 ratio to either inebilizumab or placebo within AQP4-IgG seropositive and seronegative strata, assuming approximately 80% of patients had seropositive NMOSD and 20% had seronegative NMOSD. ||||| || |||| ||||||||||||| || ||| |||||||||||| |||||| ||| ||| || ||| |||||||| ||| ||||| ||||| |||| ||||||||||||| ||| ||||| || |||||||| |||||||| |||| |||||||||.

The plan was to randomize and treat 252 patients, which was based on a blinded review of the attack rate for the first 78 patients to complete the RCP and a simulation based on these patients that indicated a 90% probability of achieving the required 67 AC-determined attacks with 252 patients. However, the data-monitoring committee recommended halting enrolment based on a clear demonstration of efficacy and condition power of more than 99%,²⁰ despite the targets of 252 patients and 67 adjudicated attacks not being met.

Statistical Testing

Unless otherwise stated, all efficacy analyses were performed with a 2-sided test at a 5% significance level. This summary contains the final analyses from the RCP as well as the descriptive data throughout the OLP |||||| ||||||||| ||||| |||||| |||||||| || ||||||.

The study was designed to strongly control the overall 2-sided type I error rate at an alpha of 0.05 based on the Bonferroni-based chain procedure. The primary null hypothesis was to be hierarchically tested first at an alpha of 0.05 in the AQP4-IgG–seropositive cohort and, if significant, would be further tested in the ITT population at an alpha of 0.05. If and only if the treatment-group comparison was statistically significant within the ITT population, then the secondary hypotheses would be tested.

The null hypotheses for the 4 key secondary end points followed the same sequential testing strategy as the primary analysis (testing first within patients with seropositive disease followed by the ITT population if the comparison within the first group of patients was statistically significant). Each secondary hypothesis was initially tested based on the Bonferroni method at an alpha of 0.05 divided by 4 (0.0125). If the null hypothesis for a particular secondary end point was rejected across both the seropositive and ITT populations, the type I error saved was to be propagated equally to other non-rejected sets of secondary null hypotheses. The testing procedure was to be repeated until all null hypotheses were rejected or no further null hypothesis could be rejected.

Subgroup Analyses

Prespecified subgroup analyses were performed for the efficacy end points in the ITT population:

• sex (male versus female)

• baseline EDSS score (< 5 points versus ≥ 5 points)

• number of prior NMOSD relapses (< 2 versus ≥ 2)

• disease duration category (< 5 years versus ≥ 5 years)

• AQP4-IgG serostatus at screening (positive versus negative).

Additional analyses were conducted post hoc according to previous rituximab treatment (yes versus no) and previous conventional immunosuppressant use as maintenance therapy (azathioprine or mycophenolate mofetil, yes versus no, distinct from rescue therapy).

Safety results were also reported separately according to AQP4-IgG serostatus (positive versus negative) and according to the sex (male versus female) of the patients participating in the RCP. |||||||||| |||| ||| || |||| |||||||| |||||||| |||| ||||||||| || ||||| |||| ||||||| ||| |||||||| ||||||||| ||| ||| |||||||||| || ||| ||||||| || ||| ||||||| |||||||| ||| ||||||| ||||||| || ||||| |||||| ||| |||.

Table 9: Statistical Analysis of Efficacy End Points in the N-MOmentum Study

AC = adjudication committee; ANCOVA = analysis of covariance; EDSS = Expanded Disability Status Scale; IDMC = independent data-monitoring committee; ITT = intention to treat; MCS = mental component score; mRS = modified Rankin Scale; NMOSD = neuromyelitis optica spectrum disorder; NR = not reported; NRS = numerical rating scale; PCS = physical component score; PD = pharmacodynamic; RCP = randomized controlled period; SF-36 = Short Form (36) Health Survey; SFP = safety follow-up period.

Sources: Clinical Study Report for N-MOmentum.⁵³ Details included in the table are from the sponsor’s summary of clinical evidence.

Analysis Populations

The analysis sets for each study are summarized in Table 10.

Table 10: Analysis Populations of the N-MOmentum Trial

IP = investigational product; ITT = intention to treat; OLP = open-label period; RCP = randomized controlled period.

Sources: Clinical Study Report for N-MOmentum.⁵³ Details included in the table are from the sponsor’s summary of clinical evidence.

Results

Patient Disposition

Patient disposition is summarized in Table 11. Although only the patients with AQP4-IgG–seropositive NMOSD are relevant for this review, this table reflects the total population, including patients with AQP4-IgG–seronegative NMOSD.

Of 467 screened patients, 175 were randomized to treatment with inebilizumab and 56 to placebo. Screening failures occurred primarily as a result of not meeting the inclusion or exclusion criteria (n = 204), while a minority was lost to follow-up, withdrew consent, or were excluded due to other unspecified reasons. ||||| ||||||| ||| ||||| ||| ||||| ||||| || |||||||||| |||||||||| |||||||| || ||| ||||||| || ||||||||| || |||| |||| |||| |||||||| || |||||| ||||||||| || || ||||||| |||| |||||| || |||||||| |||| ||||||| ||.

Of the patients randomized to treatment with inebilizumab, 1 did not receive treatment. The majority of patients in both arms completed the randomized period of the study, while fewer than 4% in each arm did not complete it due to AEs, withdrawal of consent, or other unspecified reasons. |||||| ||| |||||||||| ||||||| ||||||||||||| ||| || |||||||| ||| |||||||| ||| ||| || |||||||||||| ||| ||| || ||||||| ||| ||| |||||||| ||| |||| || ||| ||||| ||||||||| |||||||||| || |||||||||||| ||| ||| ||| |||||||| ||| |||| ||| |||| |||||| ||||||| |||| |||||||||| || |||||||| || ||||||||||| ||||| |||||||| ||||||||||| |||||||| ||||||||||| ||| || |||| || ||| ||||| ||||||||| |||||||||| || ||||||| ||| ||| ||| |||||||| ||| |||| ||| |||| |||||| |||||| ||| |||||||| |||||||||||| |||||||| ||| || || ||||||||.

Table 11: Summary of Patient Disposition in the N-MOmentum Trial

ITT = intention to treat; OLP = open-label period; RCP = randomized controlled period.

Sources: Clinical Study Report for N-MOmentum.⁵³ Details included in the table are from the sponsor’s summary of clinical evidence.

Baseline Characteristics

The baseline characteristics outlined in Table 12 are limited to those that are most relevant to this review or were felt to affect the outcomes or interpretation of the study results.

Although only the patients with AQP4-IgG–seropositive NMOSD are relevant for this review, baseline data for both patients with seropositive NMOSD and the total population (including patients with seropositive and seronegative NMOSD) are summarized in the table for comprehensiveness. This is also the case for the safety section of this review.

Among those with seropositive NMOSD, the patients were approximately 43 years old on average and the vast majority were female (approximately 94%). Approximately half of patients were white in both treatment arms, but the distribution of other races differed between the arms. In the placebo arm, the next most common race was “other” (approximately 19%), followed by Asian (15%), American Indian or Alaskan Native (10%), and Black or African American (10%). In the inebilizumab arm, the next most common race after white was Asian (approximately 23%) followed by Black or African American (9%), “other” (7.5%), American Indian or Alaskan Native (7%), and multiple categories (< 1%). || |||||||| |||| |||||| |||||||| || ||||| ||||||| |||||||| || |||||| ||||||||| |||| ||||||||||||| ||| || |||||||| |||| ||||||| || ||| || ||||| ||||||||||||| ||| || |||||||| |||| || ||||| ||||||||||.

||||| |||||||||||| ||||||||| ||| |||| |||||||||||| ||| ||||||||||||| |||||| |||||| ||||| || |||||||| |||| ||||||||||| |||||||| ||||| || ||| ||||||| ||| |||| ||| |||||||||||| |||| |||||||||| the mean number of gadolinium-enhancing lesions was slightly lower among patients randomized to placebo.

Nearly all patients (> 98%) had received prior acute or maintenance treatments for NMOSD. A higher proportion of patients assigned to placebo (50%) than inebilizumab (36%) had previously been treated with plasmapheresis as an acute therapy for NMOSD; a minority of patients had previously been treated with IVIG (6% and 5%, respectively). Approximately 60% of patients in both arms had a treatment history with maintenance therapies for NMOSD, including azathioprine and mycophenolate mofetil, and the proportions were similar between treatment arms. Approximately 14% of patients in both treatment arms had prior exposure to a biologic drug for NMOSD.

As very few patients included in the study had AQP4-IgG–seronegative NMOSD (18 out of 230 patients randomized), the baseline characteristics of the total population (including those with either seropositive or seronegative NMOSD) were very similar to the baseline characteristics of the seropositive population.

Table 12: Summary of Baseline Characteristics in the N-MOmentum Trial

AQP4-IgG = anti-aquaporin-4 immunoglobulin G; BMI = body mass index; EDSS = Expanded Disability Status Scale; Gd = gadolinium; IVIG = IV immunoglobulin; NMOSD = neuromyelitis optica spectrum disorder; SD = standard deviation; sero− = seronegative; sero+ = seropositive.

^aIncludes patients from South and Central America.

|||| ||| ||||||||||||| |||||| |||||| |||||| ||| ||| |||||| || |||||||||| |||| ||| || |||| |||||||||||||| |||||||||||| ||||| ||||| |||| |||||||| |||||||| ||||| ||| ||| ||||||||||||| |||||||||| ||||||||||||.

Sources: Clinical Study Report for N-MOmentum.⁵³ Details included in the table are from the sponsor’s summary of clinical evidence.

Exposure to Study Treatments

Exposure to study treatments is summarized in Table 13. No major differences were observed between the patients with AQP4-IgG–seropositive NMOSD and the patients in the overall population.

||||| |||||||||| ||||| |||||| ||| |||| ||||| |||| |||||||| || ||| |||| |||||||| || |||||||| || ||| |||||||||||| ||| ||||||| ||| ||| ||||||| ||| |||||||| ||| |||| |||| |||||||||||| |||| ||| ||||| |||||| ||| |||| |||||| ||||||||| || ||||||||| |||||||| |||| |||| |||| |||| |||| || ||| |||||||||||| ||| ||||||| ||||| ||||| ||| ||||| |||||||||||| || |||||||||||| |||||||| |||||| ||| ||||||||||| ||| |||| |||||| |||||||| || ||||||||| |||||||| ||| ||||||| |||| || ||| |||||||||||||||||||||||||||| ||||| ||| ||||||| |||| || ||| ||||||||||||||||||||||| |||||| ||| |||| |||| |||||| || |||||||||||| ||||| |||||| ||| ||| ||| ||||| || ||| |||||||||||||||||||||||||||| ||||| |||| ||||| |||||| |||| || ||| ||||||||||||||||||||||| ||||| |||| ||||| ||||||| || |||||||||||||||||||||||||||| ||||| |||||||||||||| |||| || ||| ||| || || |||||||| ||||||||.

Total inebilizumab exposure was 730.4 person-years among all patients who received inebilizumab across the RCP and OLP. The median duration of inebilizumab treatment was 3.2 years and the maximum duration of exposure was 5.4 years at the data cut-off (November 6, 2020).

During the RCP, all patients received a protocol-mandated 2-week course of oral corticosteroids plus a 1-week taper following the first administration of the study drug (active or placebo). Mean total corticosteroid use from day 1 to day 14 was 19.71 mg (range, 1.4 mg to 25.0 mg) in the inebilizumab arm and 19.78 mg (range, 11.4 mg to 25.0 mg) in the placebo arm, with appropriate tapering from day 15 to 21 for most patients in the placebo arm (78.6%) and in the inebilizumab arm (82.2%).

Rescue therapies are summarized in Table 14. More patients treated with placebo required rescue therapies during the trial compared with patients who received inebilizumab.

Table 13: Summary of ||||||| || ||| ||||| || |||||||||| [Redacted]

|||||||||||||||||||||| ||||||| |||||||||||| |||||||||||| ||||||| ||||||||||||||||||||||| ||||||| ||||||||| |||||||||||||||||| |||||||| ||||| |||||| ||| |||||||||||^||||||| |||||||| || ||| ||||| ||| |||| ||| ||||||||| ||||||| || |||||||| |||||||||

Table 14: Summary of ||||||| || ||| ||||| || |||||||||| [Redacted]

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

Efficacy

NMOSD Attacks

Time to NMOSD Attack During the RCP

Among patients with AQP4-IgG–seropositive NMOSD in the randomized period, treatment with inebilizumab (versus placebo) was associated with a 77.3% reduction in the risk of an AC-determined NMOSD attack (HR = 0.227; 95% CI, 0.1214 to 0.4232; P < 0.0001). During the RCP, a larger proportion of patients were attack-free in the inebilizumab group (87.6%) than in the placebo group (56.6%). Results were similar in the overall ITT population and were also consistent across prespecified subgroups. Additionally, the results were similar based on investigator-determined NMOSD attacks. Assessment of the proportional hazards assumption was not reported. The results were consistent across subgroups (Figure 3).

Time to NMOSD Attack During Open-Label Period

Descriptive results were reported for NMOSD attacks during the single-arm OLP. ||||| |||| || ||||||||| ||||||| || || |||||||| |||||||||| || ||| ||||||||||||| ||| |||||||||| ||||||||| |||| |||||||||||||||| |||||||| |||||||| |||| ||| || || ||||| ||||| ||| |||| ||| ||| || || ||||| |||||| ||||| ||| || ||||||||||||||||||||||| |||||||| ||| || |||||||||| ||||||||||||||||||| |||||| ||| |||||||| || || |||||| ||| || |||| |||||||||||||| |||| |||||||||||||||| |||||| |||| ||| ||||||||||||| ||| |||||| ||| ||| |||| ||| |||||| |||||||||

Figure 2: KM Plot of Time to AC-Determined NMOSD Attack in the N-MOmentum Trial, RCP (Patients With AQP4-IgG–Seropositive NMOSD)

AC = adjudication committee; anti-aquaporin-4 immunoglobulin G; CI = confidence interval; KM = Kaplan-Meier; MEDI-551 = inebilizumab; NA = not applicable; NMOSD = neuromyelitis optica spectrum disorder; RCP = randomized controlled period.

Sources: Clinical Study Report for N-MOmentum⁵³ and publication by Cree et al. (2019).²⁰

Figure 3: Redacted

Figure 4: Redacted

Severity of Attacks During Randomized Period

As an exploratory analysis, attack severity was graded by investigators using a predefined exploratory scale based on the degree of neurologic worsening since the prior assessment. During the RCP among patients with seropositive NMOSD, major attacks occurred in 6 of 18 attacks (33.3%) in patients treated with inebilizumab and in 10 of 22 attacks (45.5%) among patients treated with placebo.

Recovery From Attacks During Randomized Period

Recovery from attacks was graded by the AC based on improvements in the attack criteria. As a proportion of patients with attacks, “no attack recovery” was reported for 27.8% of patients in the inebilizumab group and 40.9% of patients in the placebo group.

Domain of Attacks During Randomized Period

In the cohort of patients with AQP4-IgG–seropositive NMOSD who were treated with inebilizumab, there were 18 AC-determined attacks in 161 patients (11.2%). Of these 18 attacks, none were located in the brainstem, 11 were myelitis, and 8 were optic neuritis. One of the attacks was both myelitis and optic neuritis.

In the cohort with AQP4-IgG–seropositive NMOSD who were treated with placebo, there were 22 AC-determined attacks in 54 patients (42.3%). Of these 22 attacks, 1 was located in the brainstem, 14 were myelitis, and 10 were optic neuritis. These counts are not mutually exclusive because some attacks occurred in 2 domains and were counted under each: 1 was both brainstem and optic neuritis, and 1 was both myelitis and optic neuritis.

Annualized Attack Rate

An analysis of the “any inebilizumab” population (i.e., patients who received inebilizumab at any point during the RCP and/or OLP) identified probabilities of being attack-free that ranged from 84.5% at 1 year to 77.1% at 4 years, and the AAR ranged from 0.185 events per year in year 1 to 0.019 events per year in year 4.

When calculated across the RCP and OLP, the annualized rate of AC-determined NMOSD attacks in any patient treated with inebilizumab was 0.086 attacks per year, and a similar rate was observed in the population with AQP4-IgG–seropositive NMOSD (0.09 attacks per year).

A post hoc analysis of 75 patients with AQP4-IgG–seropositive NMOSD who received inebilizumab for 4 years or more in the RCP and OLP demonstrated that most attacks during inebilizumab treatment occurred within the first 6 months, with few attacks occurring after 6 months of inebilizumab treatment.

Note that data were not available for placebo-treated patients because the study design allowed such patients to move into the OLP after the first NMOSD attack.

Proportion of Patients With Worsening in EDSS Score

During the RCP, worsening in the EDSS score was less common in the inebilizumab group than in the placebo group among patients with AQP4-IgG–seropositive NMOSD (odds ratio = 0.355; 95% CI, 0.1704 to 0.7252; P = 0.0047) || ||||||||||||| ||||||||||| |||||| ||||||| ||| |||||| || ||||||| ||| ||||||| |||| ||||||| || ||| ||||||| ||| |||||||||||.

During the OLP, the overall inebilizumab-to-inebilizumab group had low proportions of patients who experienced worsening from baseline in EDSS score, varying from 8.1% at week 13 of the OLP to 10.8% at week 104 of the OLP. In the overall placebo-to-inebilizumab group, a higher proportion of patients experienced worsening from baseline in EDSS score at week 13 of the OLP (22.4%), although this proportion decreased to 7.0% at week 104 of the OLP. Consistent trends were observed in the population of patients with AQP4-IgG–seropositive NMOSD.

Post hoc analyses of EDSS worsening during the RCP in patients with AQP4-IgG–seropositive NMOSD with or without prior azathioprine or mycophenolate mofetil use revealed numerical benefits with inebilizumab regardless of patients’ prior use of conventional immunosuppressants.

Change From Baseline in Low-Contrast Visual Acuity Score

The change in binocular low-contrast visual acuity score from baseline to the last RCP visit did not appear to differ by treatment group within the AQP4-IgG–seropositive population (least squares |||| |||| ||||||||||| ||||||| ||| ||| ||||||| || |||||||||||||||. The results were similar in the overall ITT population.

No major changes in the binocular low-contrast visual acuity score were observed during the OLP in the placebo-to-inebilizumab group or in the inebilizumab-to-inebilizumab group ||||| |||||| |||| ||||||||| |||| || |||| ||||||||.

Number of NMOSD-Related Inpatient Hospitalizations

During the RCP among the population of patients with AQP4-IgG–seropositive NMOSD, the rate ratio of NMOSD-related inpatient hospitalizations was 0.291 (95% CI, 0.1054 to 0.8017; P = 0.0170) ||| ||| ||||||||||||| |||||||| |||||||||| ||| ||||| |||||| ||| |||||| ||||||. The results were similar in the overall ITT population ||| ||| ||||||||| || ||| |||||||| |||||||||| ||| |||||||||| |||||||||| ||||||| ||| ||||| || ||||||| ||||| ||| ||| |||||||| || ||| ||||| ||| || |||||||||||| || |||||| ||||||||||| || ||||||||||||| ||||||||| ||||||||||||||||.

|||||| ||| |||| ||| |||| || ||||||||||||| ||||||||| |||||||||||||||| || ||| |||||||||||||||||||||||||||| ||||| ||| |||||| ||| || ||| ||||||||||||||||||||||| ||||| ||| |||||.

Short Form (36) Health Survey

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

Pain NRS

During the RCP, the mean changes from baseline to week 28 in the average pain score for all body locations were similar across treatment groups and between the population of patients with AQP4-IgG–seropositive NMOSD and the overall ITT populations. Similarly, the average pain scores for all body locations remained relatively constant during the OLP in the AQP4-IgG–seropositive and overall ITT populations, regardless of treatment assignment during the RCP.

Table 15: Summary of Key Efficacy Results From the N-Momentum Trial

AC = adjudication committee; anti-aquaporin-4 immunoglobulin G; CI = confidence interval; EDSS = Expanded Disability Status Scale; LS = least squares; ITT = intention to treat; N = total number; n = sample size; NA = not applicable; LS = least squares; NA = not applicable; NMOSD = neuromyelitis optica spectrum disorder; NRS = numerical rating scale; SD = standard deviation; SE = standard error; sero− = seronegative; sero+ = seropositive; SF-36 = Short Form (36) Health Survey.

^aBased on Cox regression, logistic regression, or negative binomial models, with placebo as the reference group.

^bStatistically significant at an alpha threshold of 0.05.

^cStatistically significant at an alpha threshold of 0.0125.

^dNot statistically significant at an alpha threshold of 0.05.

^eStatistically significant at an alpha threshold of 0.025 (nor an alpha of 0.05; data not shown).

||||||| |||| ||||| |||| ||| |||| ||||||||| ||| |||||||||| |||| |||| ||||||||| |||| ||||||||||| |||||||.

||||||| || |||| ||||||||||| ||| ||| ||| |||||||| ||| ||||||||| || ||||| || |||||||| || |||||||||| |||||| ||||| |||| ||||||||||| |||||| ||| |||| ||||||||||| ||||||||||||| ||||||||||| ||||||||| |||||| ||||||| ||| ||| |||||||||| ||| |||||||| ||||||||||||.

Sources: Clinical Study Report for N-MOmentum.⁵³ Details included in the table are from the sponsor’s summary of clinical evidence.

Harms

Refer to Table 16 and Table 17 for harms data during the RCP and OLP, respectively.

Adverse Events

Among the population of patients with AQP4-IgG–seropositive NMOSD, most patients experienced at least 1 AE during the RCP (71.2% of patients treated with placebo and 73.9% of patients treated with inebilizumab) and during the entire study duration (||||| ||| |||||| ||||||||||||). Results were similar for the overall ITT population. In the “any inebilizumab” population, 92.3% of patients with AQP4-IgG–seropositive NMOSD ||| ||||| || ||| |||||||| ||||||||||| || |||||||.

Relative to previous studies in the clinical development program, no new safety signals or concerns with the drug were identified in either study period, and long-term inebilizumab treatment was well tolerated after a median exposure of approximately 3.2 years during the study. Across both treatment periods, the highest severity TEAE experienced by most patients was a mild (grade 1) or moderate (grade 2) event.

During the RCP, most patients in the total as-treated population experienced 1 or more TEAEs (inebilizumab: ||||||; placebo: 73.2%) and similar TEAE incidence rates were observed in both treatment groups |||||||||||||| |||||| |||||||||| |||||||||||||| |||||||| |||||| |||||||||| |||||||||||||||.

No major differences were observed in the safety profile of inebilizumab in subgroups of patients with or without prior azathioprine or mycophenolate mofetil use || || |||||||| ||||||||||||||| |||||||.

Serious Adverse Events

Among patients with AQP4-IgG–seropositive NMOSD, the rate of SAEs was 4.3% in the inebilizumab group and 11.5% in the placebo group during the RCP.

|||||| ||| |||||| ||||| ||||||||| ||| |||| || |||| ||| ||||| || ||| |||||||||||||||||||||||||||| |||||| ||| ||||| || ||| ||||||||||||||||||||||| |||||| Among the “any inebilizumab” total ITT population, 20.4% experienced an SAE. Results were similar in the overall ITT population.

Withdrawals Due to Adverse Events

During the RCP, withdrawals due to AEs occurred in 0 patients treated with placebo (in either the AQP4-IgG–seropositive or overall ITT population), and in 2 patients treated with inebilizumab ||||| ||||||||||| ||| |||||||| ||| || |||||||||||||| |||||||| ||||||||| ||||||||||||| |||||||||| |||||||

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

Mortality

There were no deaths during the RCP. During the entire study, among the “any inebilizumab” total ITT population, 3 patients died due to NMOSD, pneumonia, and COVID-19 pneumonia (1 case each).

Notable Harms

In the RCP among patients with AQP4-IgG–seropositive NMOSD, 50.0% in the placebo group and 49.1% in the inebilizumab group experienced at least 1 TEAE of special interest, most commonly infections (44.2% and 40.4%, respectively) followed by infusion-related reactions (9.6% and 9.3%, respectively), hepatic function abnormality (3.8% and 5.0%), and cytopenia (0% and 5.0%). Results were similar in the overall ITT population.

During the entire study duration, among patients with AQP4-IgG–seropositive NMOSD, most patients experienced at least 1 TEAE of special interest (85.1% in the placebo-to-inebilizumab group and 71.4% in the inebilizumab-to-inebilizumab group). Similar to the RCP, the most common TEAE of special interest was infection, followed by infusion-related reaction, hepatic function abnormality, and then cytopenia. In addition, a few patients experienced hypersensitivity (0% and 1.3%, respectively) and ||||||| || ||||||||| ||||||||||| |||||||||| |||||||||||||||||||| || ||| |||||||||||||||||||||||||||| |||||| |||||||||||||||| ||||||||| ||||| ||| || |||||||||||. Again, results were similar in the overall ITT population. Cytopenias were more common in patients treated with inebilizumab compared with placebo, which is consistent with inebilizumab’s mechanism of action and the class effects of B-cell depletion.

Table 16: Key Harms During the RCP of the N-MOmentum Trial (As-Treated Population)

AQP4-IgG = anti-aquaporin-4 immunoglobulin G; RCP = randomized controlled period; SAE = serious adverse event; sero− = seronegative; sero+ = seropositive; TEAE = treatment-emergent adverse event.

Sources: Clinical Study Report for N-MOmentum.⁵³ Details included in the table are from the sponsor’s summary of clinical evidence.

Table 17: Overall Summary of ||||||| || ||| ||||| || |||||||||| ||||||||||| |||||| ||| |||||||||||||||| ||||||||||| [Redacted]

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

Critical Appraisal

Internal Validity

The N-MOmentum phase II and III trial was the only study included in this review. The N-MOmentum trial included an initial period that was randomized, placebo-controlled, and double-blind, with a duration of up to 197 days (the RCP) in which patients received an IV infusion of inebilizumab 300 mg or placebo on day 1 and day 15. Patients could proceed to an open-label, single-arm period (the OLP) with a minimum duration of 2 years, during which patients received inebilizumab 300 mg IV every 6 months starting 6 months after the first infusion. Blinding of initial treatment allocation was maintained during the OLP by mimicking the initial administration schedule (i.e., day 1 and day 15 IV) using matched-administration placebo. All key outcomes were assessed during the RCP, so this critical appraisal does not focus on the OLP; however, in general, an open-label and single-arm study design in the OLP has implications for the overall strength and interpretability of the results, and there is an increased risk of bias in the estimation of treatment effects due to potential confounding related to natural history and any other known or unknown prognostic factors; therefore, the data from the OLP are considered supportive. One strength of including an OLP in the study design is a longer duration of exposure for the assessment of safety outcomes.

In the RCP, there were no major concerns with regard to internal validity related to the study design in terms of, for example, the method of randomization, concealment of allocation, maintenance of blinding, or balance of patient characteristics between treatment arms. Although there were some minor between-arm differences in the distribution of races, median EDSS score, baseline visual acuity scores, and the rates of prior plasmapheresis, these differences were not expected to cause bias or confound interpretation of the results of the study. Based on consultations with the clinical experts, these factors were not expected to influence the relapse rate or time to first NMOSD attack, which are the most important drivers of all other outcomes (e.g., disability worsening, decline in visual acuity) in NMOSD. Regarding the maintenance of blinding, there were no particular AEs that were more frequent in the treatment arm that would be expected to unblind patients. The patient baseline characteristics and the results of the study end points were similar between the AQP4-IgG–seropositive subpopulation (n = 213) and the total population (N = 230), which was expected, given that few patients had seronegative NMOSD (n = 17). Randomization was stratified by serotype, followed by Japan versus non-Japan regions. The subgroup results of the seropositive population therefore retained the study design benefits of randomization.

The trial was designed to reduce potential risks to patients treated with placebo by limiting the RCP duration, implementing a 3:1 randomization ratio, and allowing patients to enter the OLP after a single attack. The N-MOmentum trial was stopped early on the recommendation of an independent data-monitoring committee, which found that the efficacy of inebilizumab had been established (conditional power > 99%) and that there was no justification to continue exposing placebo-treated patients to an increased risk of NMOSD attack. While still blinded to the study results, the sponsor offered all patients in the RCP the option to enter the OLP or exit to a safety follow-up period, per independent data-monitoring committee recommendation. No information fraction for the analysis relative to the preplanned analysis was reported. Because the study was originally powered for 67 NMOSD attack events in the seropositive subgroup and there were only 42 attacks during the RCP as of the end of the study, the information fraction would be somewhat low, which implies there is some risk of overestimating the true effect. However, statistical testing of the primary and key secondary outcomes was designed to control 2-sided type I error at an alpha of 0.05 based on the Bonferroni chain procedure.

The primary end point of time to AC-determined NMOSD attack was considered appropriate by the clinical experts consulted by CADTH, and this was supported by the patient and clinician group inputs, which stated that prevention of attacks was the most critical outcome due to all other symptoms and the potential for disability stemming from the damage that occurs and accumulates during attacks. The secondary end points of the N-MOmentum trial, including EDSS score, low-contrast visual acuity score, and NMOSD-related inpatient hospitalizations, also addressed outcomes considered valuable to patients and clinicians.

A small proportion of patients did not complete the RCP: 6 out of 174 patients assigned to inebilizumab, and 2 out of 56 assigned to placebo. There were no concerns regarding imbalanced withdrawals or the reasons for withdrawals. || ||| |||| || |||||||| |||||||| || |||||||||||| |||||||||| || ||||||| ||| ||| |||||||| ||| |||||| ||||||||| ||| || ||||||||||| || |||||||| ||| |||||||| ||| |||||| || ||||| |||||||| ||| ||| ||||||| ||||||||||.

Despite low withdrawals, a high number of patients were censored in the primary analysis of time to first NMOSD attack, especially in the inebilizumab arm. However, this was not considered concerning because the primary reason for censoring patients appeared to be the early termination of the study, given that withdrawals were generally low. As a result, the CADTH team concluded that censoring was unrelated to prognosis or other confounding factors. The sponsor did not report any evaluation of the proportional hazards assumption in the context of the Cox model used for time to first NMOSD attack, but it was considered unlikely to represent a major concern, based on visual appraisal of the Kaplan-Meier curves.

The proportion of patients who had a worsening in EDSS score was evaluated as a key secondary outcome. The usefulness and validity of the EDSS in NMOSD was discussed with the clinical experts consulted by CADTH. EDSS is a validated scale for disability in MS but has limitations, including an over-reliance on ambulation as a metric of disability, when other motor challenges in MS can contribute to loss of independence and impair the patient’s ability to do daily tasks of living and working. As a result, EDSS is known to become less sensitive to a change in disability at some ranges. This weakness is exacerbated when applied to NMOSD, for which the EDSS was neither designed nor validated. A strength of the EDSS is its ubiquity and familiarity among clinicians and its resulting ease of interpretation and, moreover, there is no more robust or valid scale to replace it with. Ultimately, despite its weaknesses in general and in its application to NMOSD instead of MS, the EDSS was an appropriate choice for the purpose of the clinical trial, and some of its insensitivities to change are overcome by the trial’s inclusion of low-contrast visual acuity as a key secondary outcome. The thresholds for “worsening” are the same as those commonly used in MS trials although, notably, there is some minor inter-trial variability and subjectivity in how worsening is defined across trials of MS. Nonetheless, the thresholds chosen were considered appropriate by the clinical experts consulted by CADTH.

External Validity

The eligibility criteria and baseline patient characteristics of the N-MOmentum trial were considered by the consulted clinical experts to be a reasonable approximation of patients with NMOSD in Canadian clinical practice. The exclusion criteria related to concurrent immunosuppressive or steroid therapy, prior IVIG therapy, and other prior therapies were appropriate for the context of the clinical trial design, but it was noted by the clinical experts that these patients would not necessarily be excluded from accessing inebilizumab for the treatment of NMOSD in clinical practice. For instance, it would not be appropriate in clinical practice to exclude patients who require concomitant steroid treatment for comorbid conditions such as asthma or inflammatory bowel disease, or to exclude patients who have received prior IVIG. Patients in the trial were also required to receive a 2-week course of corticosteroids followed by a mandatory taper, with the intention of protecting them from attack until the development of effect from the study drug; the clinical experts consulted by CADTH indicated that this adequately reflects real-world clinical practice. Premedication for infusions and practices related to rescue medication were also reflective of real-world practice in Canada.

The N-MOmentum trial had a high proportion of screen failures that was considered potentially concerning for external validity. Of 467 patients who entered the N-MOmentum study, 236 were screen failures, and only 5 of these were the result of the study’s early cessation. The most common reason was not meeting the inclusion or exclusion criteria (n = 204). Some common reasons, in no order, included unacceptable biomarker metrics for organ function, low B cells, recent receipt of prohibited therapeutics before screening (e.g., rituximab in the prior 6 months), positive or indeterminate tuberculosis test results, or insufficient number of recent NMOSD attacks.²⁷ A clinical expert consulted by CADTH indicated that this is a high rate of screen failures for clinical practice but less surprising for clinical trials, which are always more restrictive. |||||||||||||||| || ||| |||||| |||||||| |||| ||| || |||||||||||| |||||||| ||||| ||| |||||||| |||||| ||||| ||| ||||||||| |||| || ||| ||||||| || |||||||| |||||||| ||| ||||| || ||| || |||||| ||||| || |||||||||||| || ||||||| ||||| ||||||||| |||||||| || ||| |||||.

The assessment of NMOSD attacks may differ in clinical practice, as there are no standardized guidelines. The criteria for NMOSD attacks in the N-MOmentum trial were more selective and specific than assessment in clinical practice but were not considered generally inappropriate by the experts consulted by CADTH.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For the pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for the outcomes considered most relevant to inform the deliberations of CADTH’s expert committee, and a final certainty rating was determined as outlined by the GRADE Working Group:^24,25

• 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 the RCTs started 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.

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.

Results of GRADE Assessments

Table 2 presents the GRADE summary of findings for inebilizumab versus placebo.

Indirect Evidence

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

Objectives for the Summary of Indirect Evidence

There is a lack of head-to-head data comparing inebilizumab with other active therapies for adults with NMOSD. The objective of this section is to summarize and critically appraise the ITC submitted by the sponsor, which assessed the relative efficacy of inebilizumab versus each of satralizumab, eculizumab, and rituximab in patients with NMOSD. This summary also informs the pharmacoeconomic evaluation.

Description of Indirect Comparisons

The sponsor-submitted ITCs included comparisons of inebilizumab against satralizumab and eculizumab using published study data and MAIC methodology to adjust for between-trial differences, and a comparison against rituximab using IPD. Other therapies (azathioprine and mycophenolate mofetil) were also of interest and were included in the study selection criteria, but ITCs against these therapies were ultimately not considered feasible. The study selection criteria are presented in Table 18.

Table 18: Study Selection Criteria and Methods for ITCs Submitted by the Sponsor

AAR = annualized attack rate; AC = adjudication committee; ITC = indirect treatment comparison; NMOSD = neuromyelitis optica spectrum disorder; SLR = systematic literature review; vs. = versus.

Sources: Technical reports for SLR⁶² and for the indirect comparisons of inebilizumab with satralizumab and eculizumab⁶³ and rituximab.⁶⁴ Details included in the table are from the sponsor’s summary of clinical evidence.

ITC Design

Objectives

At the time of this submission, there were no direct comparisons of inebilizumab with relevant comparator treatments for adult patients with NMOSD.

The sponsor was aware of a published NMA that attempted to compare the efficacy of eculizumab, satralizumab, and inebilizumab in adult patients with AQP4-IgG–seropositive NMOSD,²⁸ but noted that the analysis had several weaknesses: small sample sizes in the monotherapy-based comparisons; significant differences in patient baseline data such as prior attack history, baseline disability, and disease duration; substantial variation in the definitions of attack criteria across the included trials; and no attack events were detected for patients who received eculizumab monotherapy, so an imputed HR was selected by the authors for use in the NMA. Additionally, the trials differed in the age range of eligible patients and whether patients could receive concurrent immunotherapy during the study.

The sponsor therefore conducted indirect comparisons with the stated objective of generating more robust evidence. Comparisons were conducted against satralizumab and eculizumab using published summary-level data from placebo-controlled trials, and against (off-label) rituximab using IPD from a collection of small observational studies.

Although azathioprine and mycophenolate mofetil were also off-label comparators of interest, the sponsor noted that the studies available for these therapies in NMOSD were predominantly retrospective or prospective observational studies; only 1 small unblinded randomized trial was identified, which compared azathioprine with rituximab in patients in Iran.³⁸ Considerable heterogeneity was noted across all studies that evaluated azathioprine or mycophenolate mofetil (e.g., different study designs, variable dosing, different definitions of NMOSD attack) and therefore it was unfeasible to conduct a robust indirect comparison of inebilizumab versus azathioprine or mycophenolate mofetil. Notably, this was also the case for rituximab (i.e., data selected by the sponsor were from small observational studies in narrow populations); however, the sponsor had access to IPD for these studies.

Study Selection Methods

To inform the comparisons, relevant publications were identified through a global systematic literature review that was conducted to identify clinical and economic studies published between 2010 and June 2021, with an updated targeted literature search for azathioprine or mycophenolate mofetil studies conducted in July 2023.

||| |||||||| ||| ||||||||| ||||| || ||| ||||||||| ||||||||| ||| |||||||||| ||||||| ||| ||||||||||||| |||||||| ||||||||| ||| |||||||||| |||||| || ||| |||||||| ||| |||||||| ||||||||||||| ||| |||||||||| ||||||||| ||| ||||||| |||||||||| |||||||| || |||||||||| |||||||| ||||||| ||| ||||||| ||| ||||||||||||| ||||||| |||||||| ||| ||||||||||| ||||||| || |||||| |||||| |||||||| |||||||| ||| |||||||||| ||||||| || |||||||||||| ||||||||||||||||||||| |||||||||| |||||||| ||| ||||| ||||||||||||| ||||||| ||||||||| |||||||| ||||||||||||| |||||| |||||||| |||||||||| |||| |||||||||||| ||||| ||| ||||| |||||||| || ||| ||||||||||||| ||| ||||||||| |||| || ||||||||| || ||| ||||||||||| ||||||||| |||| |||||||||| || ||||||| ||| || || |||||||||| ||| |||||||||| |||||||| |||||||| |||| ||| || || ||||||| || |||||||||||| |||||| ||||||| |||| |||||||| |||| |||||||| ||||||||||| |||||||||||| ||||. Table 18 || ||| ||||| |||||||| |||||||||||| || || |||||||| ||| ||||| |||||||| |||||||| || ||||||||| ||| ||||||| ||||||||||||| |||||||| || |||||||| |||| |||| || ||||| || || ||| || ||||| ||||||| ||| |||| ||||||| |||| |||||||| ||| ||||||||| || ||| ||||||||||| |||||||| ||||| || ||| ||||| ||| ||||||||| ||| |||||||||| ||||| || ||| |||| ||||| |||| ||||||||| ||||||| |||||| ||| ||||||||||||| || ||||||||| ||||||| ||| ||| |||||||| |||| |||||||| ||||||| |||||||||| |||||||||||| ||| |||||||| |||| |||||||||| ||| ||||||||| || ||| ||||||||||| ||||||||| || ||||||||||| |||||| ||||||||| ||| ||| ||| ||||||| || ||||||||| ||| ||||.

The studies from the systematic literature review were used to inform the indirect comparisons of inebilizumab versus rituximab, satralizumab, and eculizumab. To supplement missing data from the trial publications (especially regarding the AQP4-IgG–seropositive subpopulation), data to guide the satralizumab comparisons were also obtained from the FDA Center for Drug Evaluation and Research’s review of satralizumab in NMOSD⁶⁹ and CADTH’s Clinical Review Report for satralizumab in NMOSD.⁷⁰ Data to guide the eculizumab comparisons were also obtained from the CADTH Clinical Review Report for eculizumab in NMOSD.⁷¹|||| ||||||| || |||| |||||||| ||||| ||| |||||||| || ||| |||||||| ||| |||| ||||||| ||||||| ||||||||| ||||||| ||| |||||||| |||||||||||| |||| ||| |||||||||| ||||| |||| |||||||| ||| ||||||| || ||||| ||| ||||||| |||||||||| ||||||||| |||| ||||||| |||| ||| ||||||||||||| |||||||| |||| ||| || |||||||||| |||| |||||||| |||||||||||||| |||| ||| || ||||||| ||||||| ||||| |||| || ||||||||||| || |||||||||| ||| |||| || ||||||||| || ||| |||||||| ||||||| ||| |||| || ||||||| |||| ||| ||||| || |||| |||| ||||||||| || ||||| ||| |||||||| |||| || |||| || |||||||||||||| ||||||| || ||||||||||||| |||||||||| |||| |||||| |||||| ||| ||||||| || |||||||||||||| ||||||| |||| |||||||| |||||| ||||||| || ||| || |||| |||||||||||||| ||| |||||| ||||||| |||| |||||||| ||||| ||| |||||||| ||||||||||||||| |||||||||| ||||||| |||||| ||||||||| ||||||||| |||||||.

ITC Analysis Methods

Comparison Against Satralizumab and Eculizumab

Because only summary-level data were available for the comparators satralizumab and eculizumab, anchored indirect comparisons were conducted using a Bucher indirect comparison⁷⁵ (i.e., unadjusted) and anchored MAIC methodology⁷⁶ (i.e., adjusted for between-trial heterogeneity) using placebo as the common comparator. The MAIC reweighted the IPD of the N-MOmentum study to match the average baseline characteristics of the comparator trial based on summary-level data. With the weights from the MAIC, the results of the inebilizumab trial were re-estimated as if the patients had the same select baseline characteristics as in the comparator trial.

The MAIC approach approximates a propensity score–weighting approach, where the patients in the trial with IPD are weighted by their inverse odds of being in that trial compared with the comparator trial. The weights can be estimated using the generalized method of moments.⁷⁷ HRs of weighted Cox regressions were generated using the weights from the MAIC analyses. ||||||||| |||||| ||||| |||| |||||||| || ||| |||||| || ||| |||||| ||||||| ||||||| || ||| ||| || ||| ||||||| |||||||.

The baseline characteristics selected for adjustment were those identified by the sponsor as having statistically significant differences between the trials in the ITT populations. No literature or clinical expert opinion was sought to inform potentially important treatment-effect modifiers.

The only outcome assessed was the time to NMOSD attack.

Comparison Against Rituximab

||| ||||||| ||| |||||| || ||| |||| ||||||||||||||| |||||||||| ||||||||| |||||||. An unanchored MAIC was conducted using entropy balancing⁷⁸ to reweight the IPD data from the N-MOmentum trial on patients receiving inebilizumab to achieve covariate balance with IPD data for patients treated with rituximab. |||| |||||||||||||| |||||||| ||| |||| ||||| || ||||||| ||||||||| ||||||| |||||||| || ||||||| |||||||| |||| |||||||| ||||| ||| ||||||| ||||||||||||||| ||||||||||||||| || ||||| || ||||||| ||| ||||||||||||| ||||| |||||||| ||||||||||||| || ||||||||||| || ||| ||||| |||| |||||||| |||| ||| ||||||||||. The analyses were unanchored. The technical report states that AAR and EDSS were both assessed; however, only results for AAR were provided by the sponsor.

Table 19: ITC Analysis Methods

AAR = annualized attack rate; AC = adjudication committee; AQP4-IgG = anti-aquaporin-4 immunoglobulin G; EDSS = Expanded Disability Status Scale; ITC = indirect treatment comparison; MAIC = matching-adjusted indirect comparison; NMOSD = neuromyelitis optica spectrum disorder; NR = not reported; vs. = versus..

Sources: Technical reports for the systematic literature review⁶² and for the indirect comparisons of inebilizumab with satralizumab and eculizumab⁶³ and rituximab.⁶⁴ Details included in the table are from the sponsor’s summary of clinical evidence.

Results of ITC

Summary of Included Studies

Data for inebilizumab were based on IPD from the N-MOmentum trial, as previously described.

Data for the comparators were identified from the following sources:

• Inebilizumab versus eculizumab: Summary-level data from 1 double-blind, randomized, placebo-controlled, time-to-event trial (PREVENT) restricted to patients with AQP4-IgG–seropositive NMOSD with a baseline EDSS score of 7 points or less.

• Inebilizumab versus satralizumab: Summary-level data from 2 double-blind, randomized, placebo-controlled, time-to-event trials (SAkuraSky and SAkuraStar) that enrolled adolescent and adult patients with an EDSS score of 6.5 points or less. One trial, SAkuraStar, evaluated satralizumab monotherapy. The other trial, SAkuraSky, evaluated satralizumab added onto stable immunosuppressive therapy. An additional dataset was available to facilitate a comparison of inebilizumab versus satralizumab monotherapy in the subgroup of patients with AQP4-IgG–seropositive NMOSD. Results for the comparison with SAkuraSky are not discussed in this report due to the difference in study design (i.e., combination therapy).

• Inebilizumab versus rituximab: IPD from 4 small observational studies evaluating different induction and maintenance dosing regimens with variable follow-up times.

For the eculizumab and satralizumab comparisons, the included studies and their key design elements are summarized in Table 20, and the baseline characteristics of the study populations are summarized in Table 21. There were considerable differences in the study populations (e.g., the SAkuraSky trial uniquely included adolescent patients) and the outcome definition of NMO or NMOSD “attacks” or “relapses.” Additionally, the trials differed in the age range of eligible patients and whether patients could receive concurrent immunotherapy during the study. The N-MOmentum and SAkuraStar trials disallowed concurrent immunotherapy, while the SAkuraSky trial required it and the PREVENT trial allowed it. In the PREVENT trial, the majority of patients were taking concomitant immunotherapy during the trial. Patient baseline characteristics differed between the trials in terms of race, region, and prior maintenance therapy (N-MOmentum versus PREVENT trial only; prior maintenance therapy status was not reported by the satralizumab trials), and whether recent attacks were first attacks or relapses. The sponsor also identified that there were differences in disease duration between the trials, but these data were not reported.

The comparator studies informing the IPD analyses of inebilizumab compared with rituximab are summarized in Table 22 and the baseline characteristics are summarized in Table 23. All of the included rituximab studies were small, uncontrolled, regional observational studies with 18 to 32 patients. The published, placebo-controlled, double-blind RCT of rituximab conducted by Tahara et al. (2020)²⁹ in adult patients with AQO4-IgG–seropositive NMOSD was not discussed by the sponsor as a potential source of data for ITCs. It is unclear why this study was not selected, as justification was not reported.

Table 20: Summary of Studies Included in the MAICs Comparing Inebilizumab With Satralizumab and Eculizumab

AC = adjudication committee; AQP4 = aquaporin-4; AQP4-IgG = anti-aquaporin-4 immunoglobulin G; AZA = azathioprine; EDSS = Expanded Disability Status Scale; FSS = functional system score; HAI = Hauser Ambulation Index; KFS = Kurtzke Functional System; MAIC = matching-adjusted indirect comparison; MMF = mycophenolate mofetil; mRS = modified Rankin Scale; NCT = National Clinical Trial; NMO = neuromyelitis optica; NMOSD = neuromyelitis optica spectrum disorder; SC = subcutaneous.

Sources: Clinical Study Report for N-MOmentum⁵³ and publications of the SAkuraStar,⁷⁹ SAkuraSky,⁸⁰ and PREVENT⁴² trials. Technical report for indirect treatment comparisons comparing inebilizumab with satralizumab and eculizumab.⁶³ Details included in the table are from the sponsor’s summary of clinical evidence.

Table 21: Baseline Characteristics in Studies Included in the MAICs Comparing Inebilizumab With Satralizumab and Eculizumab (Anti-AQP4–Seropositive Population)

AAR = annualized attack rate; AQP4 = aquaporin-4; EDSS = Expanded Disability Status Scale; ITC = indirect treatment comparison; MAIC = matching-adjusted indirect comparison; PBO = placebo; NA = not applicable; NR = not reported; SD = standard deviation.

Sources: Clinical Study Report for N-MOmentum⁵³ and publications of the SAkuraStar,⁷⁹ SAkuraSky,⁸⁰ and PREVENT⁴² trials. Technical report for ITCs comparing inebilizumab with satralizumab and eculizumab.⁶³ Details included in the table are from the sponsor’s summary of clinical evidence.

Table 22: Summary of Comparator Studies Included in the IPD Analyses Comparing Inebilizumab With Rituximab

IPD = individual patient data.

Sources: Clinical Study Report for N-MOmentum⁵³ and technical report for indirect comparison with rituximab.⁶³

Table 23: Baseline Characteristics in Studies Included in the IPD Analyses Comparing Inebilizumab With Rituximab (Total Population)

AAR = annualized attack rate; EDSS = Expanded Disability Status Scale; IPD = individual patient data; SD = standard deviation.

Sources: Clinical Study Report for N-MOmentum⁵³ and technical report for indirect comparison with rituximab.⁶³

Table 24: Assessment of Homogeneity for ITC

AAR = annualized attack rate; AQP4 = aquaporin-4; EDSS = Expanded Disability Status Scale; IPD = individual patient data; ITC = indirect treatment comparison; MAIC = matching-adjusted indirect comparison; NA = not applicable; NMO = neuromyelitis optica; NMOSD = neuromyelitis optica spectrum disorder; NR = not reported; vs. = versus.

Sources: Clinical Study Report for N-MOmentum⁵³ and publications of the SAkuraStar,⁷⁹ SAkuraSky,⁸⁰ and PREVENT⁴² trials. Technical report for ITCs comparing inebilizumab with satralizumab and eculizumab⁶³ and technical report for indirect comparison with rituximab.⁶³ Details included in the table are from the sponsor’s summary of clinical evidence.

Results

Inebilizumab Versus Satralizumab Monotherapy

For the comparison of inebilizumab versus satralizumab, the AQP4-IgG–seropositive subpopulations were used based on data from an FDA Center for Drug Evaluation and Research clinical review⁶⁹ and the CADTH Clinical Review Report.⁷⁰ |||||||| |||| |||||||||| |||| ||||||| |||||||| |||| ||| ||| || ||||| ||| ||||||||||| |||||||| || ||| |||||||||||| ||||||| The sponsor identified that the comparison against SAkuraStar was more relevant due to its evaluation of monotherapy being more similar to the trial design of N-MOmentum, which also disallowed concomitant immunotherapy; in contrast, SAkuraSky evaluated satralizumab plus stable immunotherapy. The results of the comparison with SAkuraSky are not assessed in this review for the same reason.

In the comparison with satralizumab, the following characteristics were adjusted in the MAIC because the sponsor identified that they differed significantly at baseline:

• sex

• geographical region

• race.

Additional analyses were conducted in which different combinations of the previously mentioned characteristics were adjusted (e.g., sex and geography but not race, or sex only).

When all of the selected characteristics were adjusted (model 4), the result for time to first NMOSD attack (inebilizumab versus satralizumab) was an HR of 0.666 (95% CI, 0.182 to 2.435), suggesting a numerically lower risk of attack with inebilizumab. However, the result was not statistically significant because the 95% CI crossed 1. The effective sample sizes were very low: in the placebo arm (N = 52), the effective sample size was 26.8 (approximately 52%) and in the inebilizumab arm (N = 161), it was 38.7 (approximately 24%).

The results differed numerically by model but were not statistically significant in any case.

|||||||||| |||||||| ||| ||||||||| || ||||| ||| |||||||||| |||| ||||||||| |||||||||||||||||||||||| ||||||| ||| ||| |||||| ||| |||||||| |||| ||||| |||||| ||| ||| ||||| || |||||| ||||| |||| ||||||| ||||||||||| |||| || |||||| |||||||||| |||| |||||||||||| ||| |||| |||| ||||||||||| || ||| ||||||||| ||||||||||.

Table 25: Time to First NMOSD Attack, Inebilizumab Versus Satralizumab Monotherapy (N-MOmentum and SAkuraStar Trials, Anchored MAIC)

AQP4 = aquaporin-4; CI = confidence interval; HR = hazard ratio; ITC = indirect treatment comparison; MAIC = matching-adjusted indirect comparison; NMOSD = neuromyelitis optica spectrum disorder; SE = standard error; vs. = versus.

^aFor studies that included patients with seronegative NMOSD, only the seropositive subgroup was included in the analyses.

Sources: Clinical Study Report for N-MOmentum.⁵³ Technical report for ITCs comparing inebilizumab with satralizumab and eculizumab.⁶³ Details included in the table are from the sponsor’s summary of clinical evidence.

Inebilizumab Versus Eculizumab

The PREVENT trial only enrolled patients with AQP4-IgG–seropositive NMOSD and patients with a baseline EDSS of 7 points or less, so patients with seronegative NMOSD and those with a baseline EDSS of greater than 7 were excluded from the inebilizumab IPD to facilitate the MAIC. The eculizumab trial, PREVENT, allowed immunotherapy. The majority of patients did have concomitant immunotherapy during the PREVENT trial, and more attacks were required in the previous 1 to 2 years before enrolment than were required for the N-MOmentum trial; these differences could not be adjusted for in the MAICs.

In the comparison with eculizumab, the following characteristics were adjusted in the MAIC because the sponsor identified that they differed significantly at baseline:

• race

• geographical region

• prior maintenance therapy experience.

There are differences in the definition of NMOSD attack between the N-MOmentum and PREVENT trials that could not be accounted for by the MAIC methodology.

In model 4, which matched for all 4 characteristics (race, geographical region, experience with prior maintenance therapy, and AQP4 seropositive status), the risk of a committee-adjudicated NMOSD attack was higher with inebilizumab than with eculizumab (HR = 3.947; 95% CI, 0.917 to 17.0). The effective sample sizes were low; in the placebo arm (N = 52), the effective sample size was 24.5 (approximately 47%) and in the inebilizumab arm (N = 161), it was 97.0 (approximately 60%).

The process for the adjudication of Committee-adjudicated attacks in the PREVENT trial were less stringent than in the N-MOmentum trial and committee adjudication was conducted retrospectively.⁴² Additionally, the results of the PREVENT trial differed based on whether attacks were committee-adjudicated (HR = 0.06; 95% CI, 0.02 to 0.20; P < 0.001) or identified by physicians (HR = 0.18; 95% CI, 0.10 to 0.34; P < 0.001).⁴² Comparing the investigator-assessed NMOSD attacks from both trials results in an RR of 1.507 (95% CI, 0.700 to 3.245). Regardless of approach, there was a higher risk of attack associated with inebilizumab than eculizumab, although the results were only statistically significant using the committee-adjudicated values in some but not all models. The 95% CIs were very wide in all analyses, with the majority of the interval falling within a range that would imply a benefit of eculizumab.

Table 26: Risk of Attack With Inebilizumab Versus Eculizumab (N-MOmentum and PREVENT Trials, Anchored MAIC)

AQP4 = aquaporin-4; CI = confidence interval; HR = hazard ratio; MAIC = matching-adjusted indirect comparison; SE = standard error; vs. = versus.

Sources: Clinical Study Report for N-MOmentum.⁵³ Technical report for indirect treatment comparisons comparing inebilizumab with satralizumab and eculizumab.⁶³ Details included in the table are from the sponsor’s summary of clinical evidence.

Inebilizumab Versus Rituximab

The MAICs were based on the RR of lowering the AAR in patients with NMOSD with inebilizumab compared with rituximab. In all comparisons, including the combined rituximab IPD (RR = 1.32; standard error = 0.12), the RR was greater than 1 for all comparisons, which implies a higher risk of attack for patients with NMOSD who received rituximab compared with inebilizumab. However, no 95% CI was provided, which limits the ability to assess the uncertainty around these estimates. Although an effective sample size was reported for each comparison, it is unclear what these numbers represent, as they far exceed the populations of each of the rituximab trials alone or in combination, and the sponsor did not appear to adjust the rituximab trial data. Most of the reported effective sample sizes also exceeded the number of patients treated with inebilizumab in N-MOmentum. There was therefore no effective sample size reported for the N-MOmentum IPD after MAIC adjustment.

Additionally, no relative effect estimates were provided for the outcome of EDSS, so the results cannot be assessed.

Table 27: Inebilizumab Versus Rituximab for Lowering the AAR (N-MOmentum and Rituximab Studies With IPD, MAIC)

AAR = annualized attack rate; ESS = effective sample size; IPD = individual patient data; MAIC = matching-adjusted indirect comparison; RR = relative risk; SE = standard error.

Note: RR was estimated via ordinary least squares regression using the weights from the entropy balancing. A value > 1 favours inebilizumab.

^aIt is unclear from the submitted report what population this ESS refers to, as it greatly exceeds the population of the rituximab trials, and most values exceed the population of the inebilizumab arm in the N-MOmentum trial.

Sources: Technical report for indirect comparison with rituximab.⁶³ Details included in the table are from the sponsor’s summary of clinical evidence.

Published NMA

The methods and results of the published NMA²⁸ were not assessed in detail for this review. However, for the purpose of critically appraising the sponsor-submitted ITCs, the published NMA will be summarized briefly. The published NMA compared eculizumab with inebilizumab, satralizumab monotherapy, and satralizumab plus stable immunotherapy using a fixed-effects Bayesian NMA approach using the same evidence base as the sponsor (i.e., the N-MOmentum, SAkuraStar, SAkuraSky, and PREVENT trials) but with access to the PREVENT trial’s IPD, and only summary-level data from the N-MOmentum study. It was the opinion of the study authors that the studies included in the NMA were sufficiently similar in study design and patient population, although no detailed justification was provided. The authors highlighted that background immunotherapy was the most important difference between the included studies, and conducted separate NMAs for monotherapies and combination therapies, respectively, using IPD from the PREVENT trial to separate patients into those with or without background immunotherapy. The published NMA therefore reported on 3 comparisons that were conducted:

• Analysis 1: Combined monotherapy and combination therapy in the PREVENT, SAkuraStar, SAkuraSky, and N-MOmentum trials.

• Analysis 2: Monotherapy in the PREVENT trial monotherapy subgroup (n = 21 out of 96) and the SAkuraStar and N-MOmentum trials.

• Analysis 3: Combination therapy in the PREVENT trial combination therapy subgroup (n = 75 out of 96) and the SAkuraSky and N-MOmentum trials.

The results of the published NMA, briefly, were as follows:

• Analysis 1: Eculizumab (monotherapy or combination therapy) was superior to satralizumab (monotherapy or combination therapy) for time to first NMOSD attack; inebilizumab was not reported, as no data were present for combination therapy.

• Analysis 2: Eculizumab monotherapy was superior to both satralizumab monotherapy and inebilizumab monotherapy for time to first NMOSD attack.

• Analysis 3: Due to a wide credible interval, the results were too uncertain to draw conclusions in the comparison between eculizumab combination therapy and satralizumab combination therapy. The point estimate was considered by the authors to be numerically in favour of eculizumab combination therapy.

Critical Appraisal of the Sponsor-Submitted ITCs

Indirect Comparison With Satralizumab and Eculizumab

Both the sponsor-submitted ITCs and the published NMAs were anchored analyses. The sponsor-submitted ITCs and the published NMA disagreed on whether there was important between-trial heterogeneity in the evidence base. Neither provided robust justification for their conclusions, although both agreed that background immunotherapy was an important complicating factor, which affects any comparisons between the N-MOmentum (no background immunotherapy) and PREVENT (allowed immunotherapy, and most patients received it) trials.

The sponsor identified the following weaknesses of the published NMA: small sample sizes in the monotherapy-based comparisons; significant differences in patient baseline information, such as data on prior attack history, baseline disability, and disease duration; substantial variation in the definitions of attack criteria across the included trials; and no attack events were detected for patients who received eculizumab monotherapy, so an imputed HR was selected by the authors for use in the NMA. In addition to these factors, the trials differed in the age range of eligible patients and whether patients could receive concurrent immunotherapy during the study, although the published NMA was able to control for the latter by selecting subgroups from the PREVENT trial’s IPD. Although these are valid limitations to discuss, the majority of these concerns were not addressed by the methodology used in the sponsor-submitted ITCs, with the exception of the adjustment for seropositive status, baseline disability and, in 1 case (the comparison with eculizumab), treatment experience, which may be a rough proxy for other disease-related factors such as attack history.

Using MAIC methodology and the IPD from the N-MOmentum trial, the sponsor adjusted for differences in some eligibility and/or patient baseline characteristics such as race, geography, and treatment experience (eculizumab) or sex (satralizumab). Importantly, the sponsor selected these characteristics based only on whether they differed between the trials, without any external input from the published literature or clinical expert opinion as to whether they were important treatment-effect modifiers. The clinical experts consulted by CADTH considered that demographic factors such as sex, race, or geographic location were not likely to be important treatment-effect modifiers in NMOSD, which suggests that all of the submitted MAICs may be over-adjusting for these factors; accordingly, the effective sample sizes for the MAIC models became quite low, indicating poor overlap between the populations and reducing precision. Furthermore, identifying a factor based on statistical significance alone is not an appropriate approach since effect modifiers with small imbalances or even no imbalances could introduce bias if not included in the weighted adjustment of the MAIC.⁸¹ Thus, it is likely that other important effect-modifying factors were not accounted for using this approach.

The sponsor identified that the comparison with the SAkuraSky trial was not relevant because the patients enrolled in that trial were required to receive background immunotherapy, in contrast with the N-MOmentum trial, which disallowed concurrent immunotherapy due to the introduction of important between-trial heterogeneity. The results for this comparison are not included in this report for the same reason, as the clinical expert consulted by CADTH agreed this was an important confounding factor. However, the majority of patients in the PREVENT trial also received background immunotherapy, which could not be adjusted for in the sponsor-submitted ITCs. This therefore represents an important source of remaining between-trial heterogeneity despite the MAIC methodology. The result of the published NMA comparing eculizumab with inebilizumab may be marginally more accurate because the publication authors were able to exclude the patients in the PREVENT trial who were on concurrent immunotherapy, although there is still important uncertainty introduced because the monotherapy group in the PREVENT trial is a nonrandomized post hoc subgroup with a low sample size.

The ITCs only evaluated time to first NMOSD attack, which the clinical experts consulted by CADTH agreed is the most critical outcome in NMOSD. No safety outcomes were evaluated. The definitions of NMOSD attacks differed substantially between the trials: the N-MOmentum trial had the most strict and specific criteria, whereas the comparator trials had more general and relaxed criteria. As there are no well-defined, widely accepted criteria for defining an attack, no single approach is more or less appropriate than the other. However, it does introduce between-trial heterogeneity that was not resolved using the MAIC methodology reported by the sponsor.

The results of the comparison between inebilizumab and satralizumab were too imprecise to draw conclusions from due to wide 95% CIs that included the null value and the limitations of the methodology, as previously described. The sponsor-submitted MAIC comparing inebilizumab with eculizumab, although also imprecise, agreed with the published, unadjusted NMA that there was a higher risk of an NMOSD attack associated with inebilizumab compared with eculizumab. However, the magnitude of effect is uncertain due to imprecision and the limitations of both the submitted MAIC and published NMA, as previously described.

The sponsor-provided technical report lacked significant detail on the methodology and the results of the analyses and did not appear to adhere to best practices for conducting MAICs. The poor reporting substantially limited CADTH’s ability to appraise the evidence and raises concerns about the validity of the analyses.

Indirect Comparison With Rituximab

The sponsor-submitted MAICs comparing inebilizumab with rituximab were based on 4 small (N = 18 to 30) single-arm observational studies conducted in specific regions with populations that likely differ from Canadian clinical practice. Unanchored analyses were conducted due to the lack of a comparator arm in the rituximab studies, which raises the threshold of the assumptions required to interpret the ITC results as unbiased, as it must be assumed that all prognostic factors and treatment-effect modifiers have been accounted for. Importantly, the sponsor did not report any effort to identify important prognostic factors and treatment-effect modifiers from a clinical perspective. The sponsor noted it was not feasible to conduct ITCs comparing inebilizumab with other off-label treatments such as azathioprine and mycophenolate mofetil due to small sample sizes and observational study designs. However, the same limitations exist for the available rituximab data, and the use of IPD methodology cannot correct for these limitations. Additionally, adjusting the N-MOmentum data to reflect these studies may result in the target population becoming more dissimilar to Canadian clinical practice. Furthermore, such an adjustment was unnecessary, given the availability of IPD from the rituximab studies. Given that IPD were available for all studies in the comparison, a comparison that utilized all available data to target the treatment effect of interest, such as an appropriately specified propensity score–based comparison, would have been more appropriate.

The published, placebo-controlled, double-blind RCT of rituximab in adult patients with NMOSD conducted by Tahara et al. (2020)²⁹ was not discussed by the sponsor as a potential source of data for ITCs. It is unclear why this study was not selected, as justification was not reported. Due to the inherent weaknesses of unanchored analyses, it would have been important to assess the feasibility of using higher-quality evidence in the MAICs.

Only the outcome of AAR was assessed. There is an evidence gap for time to first NMOSD relapse. Some of the methodologies in the sponsor-submitted ITCs suggested that an analysis was also conducted for EDSS. However, no relative effect estimates were reported. No safety outcomes were evaluated.

The submitted methodology for the IPD analyses of inebilizumab versus rituximab is insufficient for critical appraisal because of poor reporting and omission of key details in both the methodology and results, in addition to the limitations noted previously. No conclusions can be drawn from the indirect comparisons with rituximab.

Discussion

Summary of Available Evidence

One completed, double-blind, randomized, placebo-controlled phase II and III study (N-MOmentum) was included in this review. The N-MOmentum trial consisted of a 197-day RCT phase (RCP) and a single-arm OLP that had a minimum duration of 2 years. The N-MOmentum trial randomized 231 patients who were adults with NMOSD, had a documented history of either 1 or more acute NMOSD attacks in the prior year or 2 or more attacks in the prior 2 years that required rescue therapy, and had an EDSS score of 7.5 or less, or 8.0 in special circumstances. The majority of patients had NMOSD that was seropositive for AQP4-IgG, and subgroup data were available for the seropositive population. The primary end point was the time in days from day 1 to the onset of an AC-determined NMOSD attack on or before day 197. The key secondary end points included the proportion of patients with a worsening in EDSS score from baseline to last visit during the RCP, change in binocular low-contrast visual acuity score from baseline to last visit during the RCP, and the number of NMOSD-related inpatient hospitalizations during the RCP. Other secondary or exploratory outcomes included the NMOSD attack rate in inebilizumab-treated patients, safety outcomes, and HRQoL (SF-36).

At baseline, the patients included in the N-MOmentum trial were mostly female (> 90%), had received prior acute or maintenance therapies for NMOSD (> 98%), were seropositive for AQP4-IgG (> 92%), and had a mean age of approximately 43 years. The median EDSS score at baseline was 4 in the placebo group (range, 1.0 to 8.0) and 3.5 in the inebilizumab group (range, 0.0 to 8.0); 29% and 24% had an EDSS score of greater than 5 points at baseline, respectively.

A sponsor-submitted ITC was summarized and critically appraised that consisted of anchored MAICs comparing the N-MOmentum trial results (inebilizumab versus placebo) indirectly with phase III RCTs of satralizumab and eculizumab, and unanchored IPD analyses comparing inebilizumab indirectly with rituximab in several small observational studies. A published NMA was briefly summarized to compare with the sponsor-submitted ITC but was not formally appraised.

Interpretation of Results

Efficacy

Evidence from the N-MOmentum trial demonstrated that treatment with inebilizumab likely resulted in a clinically important reduction in NMOSD attacks and worsening in EDSS score among patients with AQP4-IgG–seropositive NMOSD compared with placebo during the RCP of the study. Little to no differences were observed between treatment arms for the secondary outcomes of change from baseline in low-contrast visual acuity score, number of NMOSD-related inpatient hospitalizations, and change from baseline in pain NRS, nor were differences observed in the exploratory outcome of change from baseline in SF-36. There may also be a clinical benefit of inebilizumab in the exploratory outcomes of attack severity and attack recovery, but these outcomes were not formally assessed and are supportive only.

NMOSD is a rare, severe, debilitating, and life-threatening disease with no cure, and patients have limited access to targeted treatment options across Canada due to financial barriers and, for some therapies, arduous IV administration schedules. According to the patient and clinician group input and the clinical experts consulted by CADTH, the main goal of preventive treatment in NMOSD is the delay, reduction, or elimination of NMOSD-related attacks, and secondary goals include reducing the severity and improving recovery from attacks, and reducing disability accumulated from incomplete recovery after attacks. The outcomes included in the trial were therefore considered to be of clinical importance to patients and clinicians and inebilizumab demonstrated a clinically significant benefit in some of the most important outcomes.

There was incongruence between NMOSD attacks, for which there was an observed clinically meaningful benefit of inebilizumab, and the secondary outcomes related to visual acuity, pain, and NMOSD-related inpatient hospitalizations, for which there was no clear clinically important benefit observed and which are also important to patients and clinicians. The development of worsening visual acuity and pain occurs as a result of attacks, particularly those that are incompletely resolved, and so it was expected that a reduction in attacks, the occurrence of milder attacks, and better recovery from attacks as observed with inebilizumab compared with placebo may lead to improvements in these outcomes. The domain of attacks (i.e., optical neuritis, myelitis, or brainstem) may also affect these outcomes, but there was no difference between treatment arms. The clinical experts consulted by CADTH noted that in the setting of a clinical trial, there may be earlier rigorous intervention with rescue therapy in the acute treatment of attacks that may mask differences in residual symptoms that could occur in real-world treatment. Patients treated with placebo also more commonly received rescue therapy. It is also possible that the duration of the RCP was insufficient to detect a clinically meaningful difference in pain and visual acuity, as these symptoms are accumulated over time because of repeat attacks, as the clinical experts consulted by CADTH noted that the result may be due to the overall low incidence of optical neuritis over the trial duration in either treatment arm. The estimated AAR for any inebilizumab-treated patient with AQP4-IgG–positive NMOSD in the N-MOmentum trial (RCP and OLP) was 0.09 attacks per year. A large proportion of patients were attack-free at day 197 (87.6% in the inebilizumab group and 56.6% in the placebo group), which may similarly suggest that a longer study duration would be needed to observe the impact of repeated attacks on visual acuity and pain. Although the OLP had a minimum duration of 2 years, the lack of a control arm precludes making judgments on the benefit of inebilizumab over placebo. The clinical experts consulted by CADTH noted that other factors that could explain this may include a floor effect in the placebo group (15 of 56 participants [27%] read 0 of 70 characters on a binocular test using a low-contrast visual acuity chart at baseline), or preserved fellow-eye visual acuity, obscuring the effect of monocular optic neuritis on binocular visual activity.

The duration of the trial was also considered insufficient to observe meaningful differences in NMOSD-related inpatient hospitalizations. There was also no important difference identified for HRQoL-related outcomes, which was not statistically analyzed in the submission to CADTH.

In the absence of direct comparative evidence between inebilizumab and other preventive therapies for the treatment of NMOSD, the sponsor submitted indirect comparisons with eculizumab, satralizumab, and rituximab. The indirect evidence submitted to CADTH suggested that treatment with eculizumab may be associated with a clinically meaningful benefit in time to first NMOSD attack when compared indirectly with inebilizumab. No other outcomes were assessed and there were substantial unresolved sources of between-trial heterogeneity in the analysis, such as differences in the outcome definition and patient populations, small effective sample sizes, and imprecision. The results were consistent between the submitted MAIC and a published NMA in terms of the direction of effect (i.e., suggesting a benefit of eculizumab compared with inebilizumab), although the precise magnitude and clinical importance of the benefit of eculizumab over inebilizumab are uncertain due to significant limitations and imprecision in both analyses. Despite evidence of a benefit of eculizumab over inebilizumab, the clinical experts consulted by CADTH noted that inebilizumab would still have an important place in therapy due to patient preferences, differences in mechanism of action, and barriers to access.

Results were more uncertain for the indirect comparison of inebilizumab with satralizumab monotherapy for the outcome of time to first NMOSD attack because the CIs were wide, including the null value and including potentially important harms and benefits. Additionally, there was unresolved between-trial heterogeneity in the outcome definitions, and the selected treatment-effect modifiers were not selected or validated by clinical experts and may have been over-matched, causing small effective sample sizes and imprecision.

The indirect comparison of inebilizumab with rituximab was an unanchored analysis of AAR and EDSS scores, and the rituximab data were informed by 4 single-arm observational studies with small numbers of patients (≤ 30 per trial) that were conducted in specific regions that are likely to differ from the population in Canada. The indirect comparison appeared to use MAIC methodology in which the N-MOmentum trial data were adjusted to be more similar to the less relevant, international observational study data, despite access to IPD for these studies, which would have allowed a much more appropriate propensity-scoring approach. The sponsor noted that it was not feasible to conduct ITCs comparing with other off-label treatments such as azathioprine and mycophenolate mofetil due to small sample sizes and observational study designs. However, the same limitations exist for the observational studies selected by the sponsor to inform the rituximab ITCs, and the use of IPD and MAIC methodologies cannot correct for these limitations. There was no justification provided for why the sponsor did not assess the feasibility of using an available published, double-blind, placebo-controlled RCT of rituximab in the treatment of adult patients with NMOSD as a source for rituximab trial data in the ITCs. Additionally, as the comparisons were unanchored due to the selection of single-arm observational studies, it would have been imperative to adjust for all prognostic factors and treatment-effect modifiers, but the sponsor did not report an effort to identify these. No methodology or justification was provided to support the selected adjustment factors in the sponsor-submitted MAIC. In the results of the MAICs, the CIs for the relative effect estimates of AAR were not reported, and relative effect estimates for the EDSS were not reported in the submission to CADTH. Overall, the reporting of both the methodology and results was insufficient to allow for critical appraisal. As a result, no conclusions can be drawn about the comparative efficacy of inebilizumab versus rituximab.

No direct or indirect evidence exists to compare inebilizumab with azathioprine, mycophenolate mofetil, or ravulizumab. Based on clinical experience and the current trial data for inebilizumab, inebilizumab is expected to be more effective than azathioprine and mycophenolate mofetil, which are considered generally ineffective therapies in this population and are not considered disease-modifying therapies. The results of the ravulizumab trial are concurrently under review in a separate CADTH assessment at the time of this report. As no patients experienced an NMOSD attack in the single-arm ravulizumab study,⁸² it may be speculated that an indirect comparison between ravulizumab and inebilizumab could resemble the results of the eculizumab comparison previously described. Ravulizumab has a different mechanism of action than inebilizumab and has an administration schedule of IV infusions every 8 weeks. The clinical experts consulted by CADTH, in considering whether ravulizumab and inebilizumab were both hypothetically accessible, noted that patients may respond to, prefer, or tolerate 1 therapy better than the other, even if there is a difference in clinical efficacy for time to first NMOSD attack. Additionally, patients in clinical practice may be treated with ravulizumab and inebilizumab in sequence after the failure of 1 drug to prevent attacks, if both are accessible.

Harms

Over the duration of the study, nearly all patients experienced at least 1 AE, but the majority were mild to moderate. No new safety signals were identified. SAEs were more common in the placebo group than in the inebilizumab group during the study. Withdrawals due to AEs were uncommon. Most patients experienced an AE of special interest, most commonly infection, followed by infusion-related reaction, hepatic function abnormality, and cytopenia. Cytopenias were more common in patients treated with inebilizumab, which is consistent with inebilizumab’s mechanism of action and the class effects of B-cell depletion. In addition, a few patients experienced hypersensitivity ||| ||||| |||||||||| || ||||||||| ||||||||||| |||||||||| |||||||||||||||||||| || ||||||| ||||||| |||| ||||||||||||| ||||||||||||||| ||||||||| ||||| ||| || |||||||||||. The infections that occurred were generally mild and did not lead to treatment discontinuation in the OLP or RCP.

Conclusion

NMOSD is a rare, disabling, and life-threatening inflammatory disorder of the CNS characterized by acute attacks with a relapsing pattern that cause potentially irreversible damage to the optic nerve and spinal cord. There is a large unmet need for high-efficacy, well-tolerated therapies for NMOSD that have a significant impact on preventing and/or reducing attacks. Patients and clinicians highlighted that the main treatment goals include the use of an efficacious, safe, and tolerable therapy administered as soon as possible after the first attack to avoid all relapses, reduce the severity of attacks and the cumulative disability associated with them, and minimize AEs related to therapies. One phase II and III, double-blind, randomized, placebo-controlled study (N-MOmentum) was included in this review, which had a duration of 197 days followed by an open-label, single-arm period of at least 2 years.

The data submitted to CADTH and the end points assessed in the study were considered clinically relevant for the treatment of patients with NMOSD. Results from the N-MOmentum trial randomized period demonstrated that treatment with inebilizumab likely results in a clinically meaningful benefit in time to first relapse, and in the proportion of patients with a worsening in EDSS score compared with placebo at 197 days. Other secondary or exploratory outcomes were less conclusive and did not show a clearly meaningful benefit of inebilizumab over placebo, including change in low-contrast visual acuity score, NMOSD-related inpatient hospitalizations, HRQoL, and pain. The reasons for this incongruence are uncertain, but it may be that the duration of the randomized study was insufficient to detect clinically meaningful differences in these outcomes.

The ITCs included MAICs comparing inebilizumab with satralizumab and eculizumab, and IPD analyses comparing inebilizumab with rituximab. Results were inconclusive for the comparison with satralizumab. The ITCs suggest that the risk of NMOSD attack is higher with inebilizumab treatment than with eculizumab, but the magnitude of benefit is uncertain due to wide 95% CIs, imprecision, and unresolved between-trial heterogeneity. The comparison with rituximab could not be interpreted due to limitations in the analysis inherent to the data available, as well as sparsely reported methodology and results and inappropriate methodological decisions. There are no direct or indirect data available for the efficacy and safety of inebilizumab compared with azathioprine, mycophenolate mofetil, or ravulizumab.

Relative to previous studies in the clinical development program, no new safety signals or concerns with inebilizumab were identified in either study period and long-term inebilizumab treatment was well tolerated, with a median exposure of greater than 3.2 years during the study. Almost all patients experienced at least 1 AE during the study regardless of treatment assignment, but the results suggested that patients treated with inebilizumab may have fewer SAEs at 197 days than patients treated with placebo. Cytopenias were more common in patients treated with inebilizumab compared with placebo, which is consistent with inebilizumab’s mechanism of action and the class effects of B-cell depletion.

Reduction in Cumulative Total Active MRI Lesions

The reduction in cumulative total active MRI lesions (new gadolinium-enhancing or new or enlarging T2) during the RCP was a key secondary end point. To establish eligibility and a baseline value, all patients underwent a CNS (neuroaxis) MRI scan that included the optic nerve, spinal cord, and brain. ||||||||| ||| |||| ||| |||||||| || || ||||| ||| |||||||| ||| ||||||||| ||| ||| ||||||| || ||||| ||||||. The images were read by 2 independent neuroradiologists at the central reading site, with any discrepancies resolved via a consensus re-read. To reduce the risk of bias, the MRI findings were not to be reviewed by the site staff. Although an MID has not been established for changes in MRI findings, health care providers may use these findings to diagnose and manage NMOSD in practice.

The cumulative number of active MRI lesions in the AQP4-IgG–seropositive population was lower in patients treated with inebilizumab compared with placebo (1.7 versus 2.3 lesions; rate ratio = 0.568; 95% CI, 0.3851 to 0.8363).

|| ||| ||| || ||| |||| ||| ||||||||||||||||||||||| ||||| ||||||| || ||| ||||||| ||| ||| |||||||||||||||||||||||||||| ||||| |||||| || ||| |||||||.

References

1.Wingerchuk DM, Hogancamp WF, O’Brien PC, Weinshenker BG. The clinical course of neuromyelitis optica (Devic’s syndrome). Neurology. 1999;53(5):1107-1114. PubMed

2.Weinshenker BG, Wingerchuk DM. Neuromyelitis spectrum disorders. Mayo Clin Proc. 2017;92(4):663-679. PubMed

3.Wingerchuk DM, Lucchinetti CF. Neuromyelitis optica spectrum disorder. N Engl J Med. 2022;387(7):631-639. PubMed

4.Carnero Contentti E, Correale J. Neuromyelitis optica spectrum disorders: from pathophysiology to therapeutic strategies. J Neuroinflammation. 2021;18(1):208. PubMed

5.Wingerchuk DM, Banwell B, Bennett JL, et al. International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. Neurology. 2015;85(2):177-189. PubMed

6.Seok JM, Cho EB, Lee HL, et al. Clinical characteristics of disabling attacks at onset in patients with neuromyelitis optica spectrum disorder. J Neurol Sci. 2016;368:209-213. PubMed

7.Jarius S, Ruprecht K, Wildemann B, et al. Contrasting disease patterns in seropositive and seronegative neuromyelitis optica: a multicentre study of 175 patients. J Neuroinflammation. 2012;9:14. PubMed

8.Borisow N, Mori M, Kuwabara S, Scheel M, Paul F. Diagnosis and treatment of NMO spectrum disorder and MOG-encephalomyelitis. Front Neurol. 2018;9:888. PubMed

9.Eaneff S, Wang V, Hanger M, et al. Patient perspectives on neuromyelitis optica spectrum disorders: data from the PatientsLikeMe online community. Mult Scler Relat Disord. 2017;17:116-122. PubMed

10.Zarei S, Eggert J, Franqui-Dominguez L, et al. Comprehensive review of neuromyelitis optica and clinical characteristics of neuromyelitis optica patients in Puerto Rico. Surg Neurol Int. 2018;9:242. PubMed

11.Sellner J, Boggild M, Clanet M, et al. EFNS guidelines on diagnosis and management of neuromyelitis optica. Eur J Neurol. 2010;17(8):1019-1032. PubMed

12.Jarius S, Wildemann B, Paul F. Neuromyelitis optica: clinical features, immunopathogenesis and treatment. Clin Exp Immunol. 2014;176(2):149-164. PubMed

13.Ajmera MR, Boscoe A, Mauskopf J, Candrilli SD, Levy M. Evaluation of comorbidities and health care resource use among patients with highly active neuromyelitis optica. J Neurol Sci. 2018;384:96-103. PubMed

14.Mealy MA, Boscoe A, Caro J, Levy M. Assessment of patients with neuromyelitis optica spectrum disorder using the EQ-5D. Int J MS Care. 2019;21(3):129-134. PubMed

15.Beekman J, Keisler A, Pedraza O, et al. Neuromyelitis optica spectrum disorder: patient experience and quality of life. Neurol Neuroimmunol Neuroinflamm. 2019;6(4):e580. PubMed

16.Mutch K, Methley A, Hamid S, Moore P, Jacob A. If they are OK, we are OK: the experience of partners living with neuromyelitis optica. Disabil Rehabil. 2017;39(13):1279-1286. PubMed

17.Papp V, Illes Z, Magyari M, et al. Nationwide prevalence and incidence study of neuromyelitis optica spectrum disorder in Denmark. Neurology. 2018;91(24):e2265-e2275. PubMed

18.Marrie RA, Gryba C. The incidence and prevalence of neuromyelitis optica: a systematic review. Int J MS Care. 2013;15(3):113-118. PubMed

19.Etemadifar M, Nasr Z, Khalili B, Taherioun M, Vosoughi R. Epidemiology of neuromyelitis optica in the world: a systematic review and meta-analysis. Mult Scler Int. 2015;2015:174720. PubMed

20.Cree BAC, Bennett JL, Kim HJ, et al. Inebilizumab for the treatment of neuromyelitis optica spectrum disorder (N-MOmentum): a double-blind, randomised placebo-controlled phase 2/3 trial. Lancet. 2019;394(10206):1352-1363. PubMed

21.Schiopu E, Chatterjee S, Hsu V, et al. Safety and tolerability of an anti-CD19 monoclonal antibody, MEDI-551, in subjects with systemic sclerosis: a phase I, randomized, placebo-controlled, escalating single-dose study. Arthritis Res Ther. 2016;18(1):131. PubMed

22.Chen D, Gallagher S, Monson NL, Herbst R, Wang Y. Inebilizumab, a B cell-depleting Anti-CD19 antibody for the treatment of autoimmune neurological diseases: insights from preclinical studies. J Clin Med. 2016;5(12). PubMed

23.Uplizna (inebilizumab for injection): 100 mg / 10 mL (10 mg / mL) solution for intravenous infusion [product monograph]. Vaughan (ON): Horizon Therapeutics Canada; 2023 Dec 15.

24.Rugge B, Balshem H, Sehgal R, Relevo R, Gorman P, Helfand M. Appendix A, lipid conversion factors. Screening and treatment of subclinical hypothyroidism or hyperthyroidism Rockville (MD): Agency for Healthcare Research and Quality (US); 2011.

25.Santesso N, Glenton C, Dahm P, et al. GRADE guidelines 26: informative statements to communicate the findings of systematic reviews of interventions. J Clin Epidemiol. 2020;119:126-135. PubMed

26.Horizon Therapeutics Canada response to September 5, 2023 CADTH request for additional information regarding Uplizna review [internal sponsor’s report]. Vaughan (ON): Horizon Therapeutics Canada; 2023 Sep 15.

27.Horizon Therapeutics Canada response to September 29, 2023 CADTH request for additional information regarding Uplizna review [internal sponsor’s report]. Vaughan (ON): Horizon Therapeutics Canada; 2023 Oct 5.

28.Wingerchuk DM, Zhang I, Kielhorn A, et al. Network meta-analysis of Food and Drug Administration-approved treatment options for adults with aquaporin-4 immunoglobulin G-positive neuromyelitis optica spectrum disorder. Neurol Ther. 2021;11(1):123-135. PubMed

29.Tahara M, Oeda T, Okada K, et al. Safety and efficacy of rituximab in neuromyelitis optica spectrum disorders (RIN-1 study): a multicentre, randomised, double-blind, placebo-controlled trial. Lancet Neurol. 2020;19(4):298-306. PubMed

30.Kitley J, Leite MI, Nakashima I, et al. Prognostic factors and disease course in aquaporin-4 antibody-positive patients with neuromyelitis optica spectrum disorder from the United Kingdom and Japan. Brain. 2012;135(Pt 6):1834-1849. PubMed

31.Jasiak-Zatonska M, Kalinowska-Lyszczarz A, Michalak S, Kozubski W. The immunology of neuromyelitis optica-current knowledge, clinical implications, controversies and future perspectives. Int J Mol Sci. 2016;17(3):273. PubMed

32.Cabre P, Gonzalez-Quevedo A, Bonnan M, et al. Relapsing neuromyelitis optica: long term history and clinical predictors of death. J Neurol Neurosurg Psychiatry. 2009;80(10):1162-1164. PubMed

33.Sherman E, Han MH. Acute and chronic management of neuromyelitis optica spectrum disorder. Curr Treat Options Neurol. 2015;17(11):48. PubMed

34.Oh J, Levy M. Neuromyelitis optica: an antibody-mediated disorder of the central nervous system. Neurol Res Int. 2012;2012:460825. PubMed

35.Mealy MA, Wingerchuk DM, Greenberg BM, Levy M. Epidemiology of neuromyelitis optica in the United States: a multicenter analysis. Arch Neurol. 2012;69(9):1176-1180. PubMed

36.Mealy MA, Kessler RA, Rimler Z, et al. Mortality in neuromyelitis optica is strongly associated with African ancestry. Neurol Neuroimmunol Neuroinflamm. 2018;5(4):e468. PubMed

37.Collongues N, Marignier R, Jacob A, et al. Characterization of neuromyelitis optica and neuromyelitis optica spectrum disorder patients with a late onset. Mult Scler. 2014;20(8):1086-1094. PubMed

38.Nikoo Z, Badihian S, Shaygannejad V, Asgari N, Ashtari F. Comparison of the efficacy of azathioprine and rituximab in neuromyelitis optica spectrum disorder: a randomized clinical trial. J Neurol. 2017;264(9):2003-2009. PubMed

39.Etemadifar M, Salari M, Mirmosayyeb O, et al. Efficacy and safety of rituximab in neuromyelitis optica: review of evidence. J Res Med Sci. 2017;22(18). PubMed

40.Kessler RA, Mealy MA, Levy M. Treatment of neuromyelitis optica spectrum disorder: acute, preventative, and symptomatic. Curr Treat Options Neurol. 2016;85(1). PubMed

41.Kümpfel T, Giglhuber K, Aktas O, et al. Update on the diagnosis and treatment of neuromyelitis optica spectrum disorders (NMOSD) - revised recommendations of the Neuromyelitis Optica Study Group (NEMOS). Part II: attack therapy and long-term management. J Neurol. 2024;271(1):141-176. PubMed

42.Pittock SJ, Berthele A, Fujihara K, et al. Eculizumab in aquaporin-4–positive neuromyelitis optica spectrum disorder. New Engl J Med. 2019;381(7):614-625. PubMed

43.CADTH Canadian Drug Expert Committee recommendation: eculizumab (Soliris-Alexion Pharma Canada Corp). Ottawa (ON): CADTH; 2020 Aug 24: https://www.cadth.ca/sites/default/files/cdr/complete/SR0640%20Soliris%20-%20CDEC%20Final%20Recommendation%20August%2024%2C%202020%20for%20posting.pdf. Accessed 2023 Apr 18.

44.CADTH reimbursement recommendation: satralizumab (Enspryng) for treatment of neuromyelitis optica spectrum disorder. Ottawa (ON): CADTH; 2021 Apr: https://www.cadth.ca/sites/default/files/cdr/complete/SR0663%20Enspryng%20-%20CDEC%20Final%20Recommendation%20April%2023%2C%202021_for%20posting.pdf. Accessed 2023 Apr 18.

45.Soliris (eculizumab): 30 mL parenteral solution (10 mg/mL) for injection [product monograph]. Zurich (CH): Alexion Pharma GmbH; 2021 Mar 25.

46.Enspryng (satralizumab injection): solution, 120 mg/mL, subcutaneous [product monograph]. Mississauga (ON): Hoffmann-La Roche Ltd; 2021 Apr 21.

47.Ultomiris (ravulizumab for injection): 10 mg/mL & 100 mg/mL concentrate for solution for intravenous infusion [product monograph]. Mississauga (ON): Alexion Pharma Canada Corp; 2023 Oct 30.

48.Rituxan (rituximab): 10 mg/mL intravenous infusion [product monograph]. Mississauga (ON): Hoffmann-La Roche Ltd; 2023 Jun 2.

49.Imuran (azathioprine): azathioprine tablets (50 mg); azathioprine sodium for injection (50 mg azathioprine per vial) [product monograph]. Toronto (ON): Aspen Pharmacare Canada, Inc; 2018 Nov 8.

50.Myfortic (myciphenolic acid enteric-coated tablets): 180 mg, 360 mg (as mycophenolate sodium) [product monograph]. Dorval (QC): Novartis Pharmaceuticals Canada Inc; 2018 May 10.

51.CellCept (myciphenolic mofetil): capsules – 250 mg; film-coated tablets – 500 mg; powder for oral suspension – 200 mg/mL (when reconstituted) [product monograph]. Mississauga (ON): Hoffmann-La Roche Ltd; 2018 Nov 14.

52.Kim HJ, Aktas O, Patterson KR, et al. Inebilizumab reduces neuromyelitis optica spectrum disorder risk independent of FCGR3A polymorphism. Ann Clin Transl Neurol. 2023. PubMed

53.Clinical Study Report: CD-IA-MEDI-551-1155. Clinical Study Report for CD-IA-MEDI-551-1155, a double-masked, placebo-controlled study with open-label period to evlauate the efficacy and safety of MEDI-551 in adult subjects with neuromyelitis optia and neuromyelitis optica spectrum disorders [internal sponsor’s report]. Gaithersburg (MD): MedImmune, LLC; 2021 Jun 9.

54.de Groot V, Beckerman H, Uitdehaag BM, et al. The usefulness of evaluative outcome measures in patients with multiple sclerosis. Brain. 2006;129(Pt 10):2648-2659. PubMed

55.Goldman MD, LaRocca NG, Rudick RA, et al. Evaluation of multiple sclerosis disability outcome measures using pooled clinical trial data. Neurology. 2019;93(21):e1921-e1931. PubMed

56.Riemsma RP, Forbes CA, Glanville JM, Eastwood AJ, Kleijnen J. General health status measures for people with cognitive impairment: learning disability and acquired brain injury. Health Technol Assess. 2001;5(6):1-100. PubMed

57.Robinson D, Jr., Zhao N, Gathany T, Kim LL, Cella D, Revicki D. Health perceptions and clinical characteristics of relapsing-remitting multiple sclerosis patients: baseline data from an international clinical trial. Curr Med Res Opin. 2009;25(5):1121-1130. PubMed

58.Jensen MP, Karoly P, Braver S. The measurement of clinical pain intensity: a comparison of six methods. Pain. 1986;27(1):117-126. PubMed

59.Jensen MP, Turner JA, Romano JM, Fisher LD. Comparative reliability and validity of chronic pain intensity measures. Pain. 1999;83(2):157-162. PubMed

60.Sharrack B, Hughes RA, Soudain S, Dunn G. The psychometric properties of clinical rating scales used in multiple sclerosis. Brain. 1999;122 (Pt 1):141-159. PubMed

61.Shahrbanian S, Duquette P, N EM. Impairment, disability and fatigue in multiple sclerosis. Caspian J Intern Med. 2018;9(3):244-251. PubMed

62.Systematic literature review of clinical and economic evidence of current treatments for neuromyelitis optica spectrum disorder [internal sponsor’s report]. In: Drug Reimbursement Review sponsor submission: Uplizna (inebilizumab), 10 mg/mL solution for intravenous infusion. Vaughan (ON): Horizon Therapeutics Canada; 2021 Oct.

63.Technical note for comparative analyses of the efficacy of inebilizumab compared to satralizumab and eculizumab [internal sponsor’s report]. In: Drug Reimbursement Review sponsor submission: Uplizna (inebilizumab), 10 mg/mL solution for intravenous infusion. Vaughan (ON): Horizon Therapeutics Canada; 2022 Aug 19.

64.Efficacy of inebilizumab compared to rituximab for the indication neuromyelitis optica spectrum disorder - IPD analyses [internal sponsor’s report]. In: Drug Reimbursement Review sponsor submission: Uplizna (inebilizumab), 10 mg/mL solution for intravenous infusion. Vaughan (ON): Horizon Therapeutics Canada; 2022 Dec 30.

65.Shamseer L, Moher D, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ. 2015;350:g7647. PubMed

66.CRD’s guidance for undertaking reviews in health care York (GB): Centre for Reviews and Dissemination; 2009: https://www.york.ac.uk/media/crd/Systematic_Reviews.pdf. Accessed 2020 May 11.

67.Higgins J, Thomas J. Cochrane handbook for systematic reviews of interventions 2019: https://training.cochrane.org/handbook/current Accessed 2020 May 11.

68.Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7):e1000097. PubMed

69.Center for Drug Evaluation Research. Clinical review(s). Enspryng (satralizumab) injection: 120mg/mL in a single-dose prefilled syringe. Company: Genentech, Inc. Application No.:761149. Approval date: 8/14/2020 (FDA approval package). Silver Spring (MD): U.S. Food and Drug Administration (FDA); 2020: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2020/761149Orig1s000MedR.pdf. Accessed 2023 Aug 22.

70.CADTH common drug review. Clinical review report: satralizumab (Enspryng) for neuromyelitis optica spectrum disorder. Ottawa (ON): CADTH; 2021 Jun: https://www.cadth.ca/sites/default/files/cdr/clinical/sr0663-enspryng-clinical-review-report.pdf. Accessed 2023 Mar 16.

71.CADTH common drug review. Clinical review report: eculizumab (Soliris) for neuromyelitis optica spectrum disorder. Ottawa (ON): CADTH; 2020: https://www.cadth.ca/sites/default/files/cdr/clinical/sr0640-soliris-nmosd-clinical-review-report.pdf. Accessed 2023 Mar 16.

72.Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898. PubMed

73.Sterne JA, Hernán MA, Reeves BC, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ. 2016;355:i4919. PubMed

74.Scottish Intercollegiate Guidelines Network (SIGN) checklist for cohort studies. Edinburgh (GB): SIGN; 2012: https://www.sign.ac.uk/what-we-do/methodology/checklists/. Accessed 2020 May 11.

75.Bucher HC, Guyatt GH, Griffith LE, Walter SD. The results of direct and indirect treatment comparisons in meta-analysis of randomized controlled trials. J Clin Epidemiol. 1997;50(6):683-691. PubMed

76.Signorovitch JE, Wu EQ, Yu AP, et al. Comparative effectiveness without head-to-head trials. PharmacoEcon. 2010;28(10):935-945. PubMed

77.Signorovitch JE, Sikirica V, Erder MH, et al. New tool for timely comparative effectiveness research. Value Health. 2012;15(6):940-947. PubMed

78.Phillippo DM, Dias S, Ades AE, Welton NJ. Equivalence of entropy balancing and the method of moments for matching‐adjusted indirect comparison. Res Synth Methods. 2020;11(4):568-572. PubMed

79.Traboulsee A, Greenberg BM, Bennett JL, et al. Safety and efficacy of satralizumab monotherapy in neuromyelitis optica spectrum disorder: a randomised, double-blind, multicentre, placebo-controlled phase 3 trial. Lancet Neurol. 2020;19(5):402-412. PubMed

80.Yamamura T, Kleiter I, Fujihara K, et al. Trial of satralizumab in neuromyelitis optica spectrum disorder. New Engl J Med. 2019;381(22):2114-2124. PubMed

81.Phillippo D, Ades T, Dias S, Palmer S, Abrams KR, Welton N. NICE DSU technical support document 18: Methods for population-adjusted indirect comparisons in submissions to NICE Sheffield (GB): NICE Decision Support Unit; 2016: https://research-information.bris.ac.uk/ws/portalfiles/portal/94868463/Population_adjustment_TSD_FINAL.pdf. Accessed 2023 Nov 7.

82.Pittock SJ, Barnett M, Bennett JL, et al. Ravulizumab in aquaporin-4-positive neuromyelitis optica spectrum disorder. Ann Neurol. 2023;93(6):1053-1068. PubMed

Appendix 1: Detailed Outcome Data

Note that this appendix has not be copy-edited.

Table 28: Absolute ||||||| ||| |||||||| || |||||||||| [Redacted]