Reimbursement Review

Nemolizumab (Nemluvio)

Sponsor: Galderma Canada Inc.

Therapeutic area: Atopic dermatitis

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Clinical Review

Abbreviations

Executive Summary

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

Table 1: Background Information of Application Submitted for Review

AD = atopic dermatitis; NOC = Notice of Compliance.

Introduction

Atopic dermatitis (AD) is a chronic, relapsing, inflammatory, and noncontagious skin disease that is thought to involve a complex interplay of genetic, immune, and neuroimmune dysregulation, skin barrier dysfunction, microbial imbalance, and environmental factors.^1-4 It occurs most frequently in early childhood; it is estimated that the prevalence of eczema in Canada is 15% in children and adolescents (aged 0.5 years to 17 years), and 3.5% in adults.^5,6 Acute worsening of AD, commonly referred to as a flare, presents as dry, red, itchy skin that can lead to lesions that blister, ooze, and crust. AD can be mild, moderate, or severe, depending on the extent and the severity of symptoms.⁷ An intense and debilitating itch and chronically relapsing lesions are the key clinical hallmarks of moderate to severe disease and can result in sleep disturbances, psychosocial distress, and reduced quality of life for patients and caregivers.⁷

The management of AD involves a multipronged approach including patient education, skin hydration, restoration of the skin, elimination of exacerbating factors, and pharmacologic treatment of skin inflammation.⁸ The clinical experts consulted for this review noted that the goals of treatment are reduction of signs and symptoms (e.g., red scaly plaques, itch, scratching, lichenification, dyspigmentation, scarring), prevention of secondary infections, improvement of quality of life (e.g., sleep, daily activities, hobbies), maintenance of independence, reduction of caregiver burdens, and enhancement of work and school productivity. Patients with persistent, moderate to severe AD despite optimized topical therapy require systemic treatment to achieve disease control. Topical drugs may be used concurrently with phototherapy or systemic drugs for the maintenance of response, rescue treatment, or control of flares.⁹ The clinical experts noted that a typical progression would be for patients to trial conventional systemic therapies for a period of 3 months and if well controlled, they would remain on that therapy, but if they do not respond, then they may switch to another conventional systemic therapy or the advanced therapies. They noted a trial of 3 to 6 months is usually used to determine response. If the patient’s skin is not completely clear, they may add a topical therapy. Occasionally, the experts noted that a combination of systemic therapies may be used. Dupilumab, abrocitinib, and upadacitinib are recommended by Canada’s Drug Agency (CDA-AMC) for reimbursement for the treatment of moderate to severe AD. Lebrikizumab and tralokinumab are recommended not to be reimbursed for moderate to severe AD but are approved for use in Canada.

Nemolizumab is a humanized, monoclonal antibody of the IgG2 subclass that inhibits interleukin (IL)-31 signalling by selectively binding to IL-31RA to ameliorate pruritus, inhibit inflammatory responses, and restore barrier integrity in AD. Nemolizumab received a Health Canada indication for the treatment of patients aged 12 years and older with moderate to severe AD whose disease is not adequately controlled with topical prescription therapies or when those therapies are not advisable. The sponsor’s reimbursement request is for the treatment of moderate to severe AD in patients aged 12 years and older whose disease is not adequately controlled with topical prescription therapies or when those therapies are not advisable, and who are refractory to or ineligible for systemic immunosuppressant therapies.

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of nemolizumab subcutaneous (SC) (30 mg/0.49 mL in either a single-dose, prefilled dual-chamber pen or prefilled dual-chamber syringe with copackaged needle) for the treatment of moderate to severe AD in patients aged 12 years and older whose disease is not adequately controlled with topical prescription therapies or when those therapies are not advisable, and who are refractory to or ineligible for systemic immunosuppressant therapies.

Perspectives of Patients, Clinicians, and Drug Programs

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

Patient Input

One patient group submitted input for this review, the Eczema Society of Canada (ESC), a registered charity in Canada aiming to improve the lives of people living in Canada with eczema through support, education, awareness, and research. Information was gathered for this submission through questionnaires, patient interviews, review of published literature, and data from multiple surveys with a combined total of more than 3,000 people living in Canada.

Patients with AD described substantial physical and emotional burdens, including painful, itchy skin lesions that crack, bleed, and lead to scarring, pigment changes, and infection risk. Persistent itching disrupts sleep, school and work, and daily life, with some patients experiencing chronic pain and no relief. Emotionally, patients reported isolation, frustration, anxiety, depression, and difficulties with social and intimate relationships due to low self-esteem and treatment failures. Historically, systemic options were limited to phototherapy, off-label immunosuppressants, and steroids, which many fear using long term. While newer systemic therapies have helped some, access and response remain inconsistent. Patients expressed frustration with the trial-and-error nature of treatment and emphasized that due to the heterogeneous nature of AD, no single therapy works for all.

ESC highlighted the urgency for innovative therapies that can offer long-term relief and improve quality of life. Outcomes of treatment identified as most important to patients with AD include improved disease control, symptom and pain relief (e.g., itch, dryness, inflammation, blistering), convenience, and improved quality of life. ESC described general patient experiences with nemolizumab, although the source of these experiences (e.g., patient interviews, survey responses, general trial observations) was not specified. Patients treated with nemolizumab reported substantial reductions in itch, with noticeable improvements in skin clearance from rashes, lesion, sores, and open wounds, which provided further benefit in improving sleep, productivity, self-esteem, and mental health. ESC noted that many patients who were previously unsuccessful with approved and off-label medications reported success with nemolizumab.

Clinician Input

Input From Clinical Experts Consulted for This Review

The clinical experts noted that pruritus (itch) is a very debilitating and unpredictable aspect of AD, and better treatments are needed to target this symptom. In addition, approximately 30% to 40% of patients will either not respond to, or will become refractory to, the currently available nonbiologic treatments (1 expert noted this would be approximately 20% of patients if biologics were also considered); therefore, new treatments are needed. Moderate to severe AD often comes with a high burden of disease, which also impacts patient’s health-related quality of life (HRQoL) and their ability to carry out activities due to symptoms, or due to a feeling of self-consciousness. The currently available treatments (e.g., JAK inhibitors) are not always well tolerated and have contraindications or black box warnings, further reinforcing the need for new treatment options.

The clinical experts noted that nemolizumab is the only treatment with a mechanism of action targeting IL-31, which is involved in the sensation of pruritus. It could be used as monotherapy or in combination with topical therapies, phototherapy, and other conventional immunosuppressing systemic drugs. They noted that nemolizumab should be considered a first-line systemic therapy after topical therapies, and considered alongside the other biologics such as dupilumab, tralokinumab, or lebrikizumab. The clinical experts noted that the reimbursement criteria for dupilumab could also be considered appropriate for nemolizumab (e.g., having been unsuccessful with 1 conventional systemic therapy trial); however, the choice to use a conventional treatment or advanced systemic drug would be partially determined by the drug coverage.

The experts noted that nemolizumab would be suitable for all patients with moderate to severe AD, and 1 expert noted it would also be appropriate for patients who do not tolerate, respond to, or cannot access phototherapy, as well as those who have a relative contraindication to, or have been unsuccessful with, alternate systemic therapies. The experts further noted that patients who are substantially impacted by pruritus, who have substantial flares of AD, whose current therapy has stopped working, or whose AD has a substantial impact on their lives would be those most in need of additional interventions and most likely to benefit. There are no specific criteria or laboratory tests required for diagnosis, as AD is typically diagnosed clinically and is a common disease. The experts noted that there are no biomarkers or tests that would indicate a patient’s likelihood to respond to nemolizumab.

Measures used in the pivotal trials are also used in clinical practice to assess response, including patient global assessment of improvement, Eczema Area and Severity Index (EASI) reduction, numeric rating scale for itch, and Dermatology Life Quality Index (DLQI). The clinical experts also noted that other subjective changes reported by patients in clinical practice may include changes to sleep, daily activities, work productivity, improvement in mental health, and reduction of scratching or itching. They commented that the measures used in clinical practice vary and may be chosen according to the reimbursement criteria set by payers to obtain coverage for a treatment (e.g., change in EASI score at 6 months). Patients are often assessed every 3 months initially and when patients are improved and their condition is considered stable, they are typically assessed every 6 months.

Treatment should be discontinued if there has been no meaningful response after 6 months of therapy, if the patient experiences an allergy or hypersensitivity, or if they experience intolerable adverse events (AEs). The experts noted that definition of a meaningful response could vary and the decision to discontinue is often shared between the clinician and the patient. One expert shared an example that patients with minimal disease response according to a tool such as EASI may still wish to continue treatment if their pruritus is controlled. If there is a partial response, the experts noted that the addition of a new therapy should be considered first, rather than discontinuing the current treatment.

According to the experts, patients should receive a proper diagnosis of AD by a dermatologist before nemolizumab is prescribed. Patients can be managed and monitored by dermatologists in any outpatient clinic (community or hospital). After 6 months of treatment and confirmation of clinical efficacy, primary care providers could also manage the prescribing until reinvolvement of the dermatologist is necessary.

Clinician Group Input

One joint input was submitted for this review by the Dermatology Association of Ontario (DAO) and the Atlantic Dermatology Group. DAO provides broad representation for more than one-half of registered dermatologists in Canada. The Atlantic Dermatology Group is a group of hospital-based and academic-based dermatologists practising in the Atlantic provinces. Information for this submission was gathered through a review of published literature, clinical trial data, and experience with nemolizumab as clinical trialists.

The input from clinician groups was generally consistent with the clinical experts consulted for this review, emphasizing the ongoing unmet need for treatments that provide rapid and sustained relief from pruritus, as well as improvements in sleep and quality of life for patients with moderate to severe AD. Like the experts, the clinician groups highlighted the limitations of current therapies, including slow onset of action with biologics and safety concerns or monitoring requirements with JAK inhibitors. Aligning with expert input, the clinician groups considered nemolizumab a suitable first-line systemic option, to be positioned alongside existing biologics, particularly for patients with an inadequate response to topicals or systemic therapies, or who require more rapid symptom relief. Both also highlighted the novel mechanism and monthly dosing schedule as factors that could improve adherence and convenience. The clinician groups provided additional details on patient selection, noting that nemolizumab may be particularly appropriate for patients with contraindications to JAK inhibitors (e.g., cardiovascular disease, immunosuppression concerns, thromboembolism). They also noted that patients with mild AD that can be managed effectively by topicals alone are less appropriate candidates.

Consistent with expert input, the clinician groups noted a range of clinical and patient-reported tools (e.g., Investigator Global Assessment [IGA], EASI, Peak Pruritus Numeric Rating Scale [PP-NRS], sleep scores) that could be used to assess treatment response, and support reassessment intervals from 4 to 12 months. However, they provided more specific discontinuation criteria, such as the inability to achieve a 4-point or more reduction in PP-NRS or sleep numeric rating scale or less than 50% improvement in EASI by week 16, while the experts emphasized a flexible, shared decision-making approach. Both clinical experts and clinician groups agreed that initiation should be done by dermatologists in an outpatient setting, but the clinician groups added that allergists could also initiate therapy, and that self-injection could be possible after training. The clinician group noted that because misdiagnosis of AD can occur, particularly in adult-onset cases, clinical expertise in differentiating AD from other eczematous conditions (e.g., contact dermatitis, seborrheic dermatitis, cutaneous T-cell lymphoma) is essential to ensure appropriate patient selection.

Drug Program Input

Input was obtained from the drug programs that participate in the CDA-AMC reimbursement review process. The following were identified as key factors that could potentially impact the implementation of a recommendation for nemolizumab:

• relevant comparators

• considerations for initiation of therapy

• considerations for continuation or renewal of therapy

• considerations for prescribing of therapy

• system and economic issues.

The clinical experts provided advice on the potential implementation issues raised by the drug programs. Refer to Table 5 for more details.

Clinical Evidence

Systematic Review

Description of Studies

Three studies were included in the systematic review: ARCADIA 1, ARCADIA 2, and ARCADIA CYCLO.

ARCADIA 1 and ARCADIA 2 Studies

The study designs for the ARCADIA 1 and ARCADIA 2 studies were identical; both were phase III, double-blind, placebo-controlled, multicentre randomized controlled trials (RCTs) evaluating the efficacy and safety of nemolizumab in adults and adolescents (aged ≥ 12 years) with moderate to severe AD. The coprimary end point of the ARCADIA 1 and ARCADIA 2 studies was the proportion of patients with IGA success, and the proportion of patients with a 75% or greater improvement in the EASI score from baseline (EASI-75) at week 16. Key secondary end points also evaluated changes in PP-NRS across the study period. Patients were randomized via interactive response technology 2:1 to receive nemolizumab (ARCADIA 1: n = 620; ARCADIA 2: n = 522) or placebo (ARCADIA 1: n = 321; ARCADIA 2: n = 265) for 16 weeks. After the 16-week initial treatment period, patients in the treatment arm who met the study definition of clinical responders to treatment (i.e., an IGA of 0 [clear] or 1 [almost clear] or EASI-75 at week 16) were rerandomized 1:1:1 to receive either nemolizumab every 4 weeks (n = 169), nemolizumab every 8 weeks (n = 169), or placebo (n = 169) for the maintenance period.

ARCADIA CYCLO Study

The ARCADIA CYCLO study was a phase IIIb, double-blind, placebo-controlled, multicentre RCT evaluating the efficacy and safety of nemolizumab in adult patients aged 18 years or older with chronic AD who had a documented history of inadequate response, intolerance, or unacceptable toxicity to cyclosporine (CsA), or in whom CsA was medically inadvisable. Patients were randomized 1:1 to receive nemolizumab (n = 138) or placebo (n = 138). The primary end point of the ARCADIA CYCLO study was the proportion of patients with EASI-75, and proportion of patients with an improvement in PP-NRS of 4 or more from baseline at week 16.

Efficacy Results

ARCADIA 1 and ARCADIA 2 Studies: Initial Period

Proportion of Patients With IGA Success at Week 16

For patients in the ARCADIA 1 study, at week 16, 35.6% of patients in the nemolizumab arm and 24.6% of patients in the placebo arm attained IGA success (defined as IGA 0 or 1 and a ≥ 2-point reduction from baseline). The strata-adjusted difference between arms was 11.5% (97.5% confidence interval [CI], 4.7% to 18.3%; P = 0.0003). For patients in the ARCADIA 2 study, at week 16, 37.7% of patients in the nemolizumab arm and 26.0% of patients in the placebo arm attained IGA success. The strata-adjusted difference between arms was 12.2% (97.5% CI, 4.6% to 19.8%; P = 0.0006).

Proportion of Patients With EASI-75 at Week 16

For patients in the ARCADIA 1 study, at week 16, 43.5% of patients in the nemolizumab arm and 29.0% of patients in the placebo arm attained EASI-75. The strata-adjusted difference between arms was 14.9% (97.5% CI, 7.8% to 22.0%; P < 0.0001). For patients in the ARCADIA 2 study, at week 16, 42.1% of patients in the nemolizumab arm and 30.2% of patients in the placebo arm attained EASI-75. The strata-adjusted difference between arms was 12.5% (97.5% CI, 4.6% to 20.3%; P = 0.0006).

Change From Baseline to Week 16 in Total EASI Score

For patients in the ARCADIA 1 study, the least squares (LS) mean change from baseline in total EASI score was –12.36 (95% CI, –13.96 to –10.76) in the nemolizumab arm, and –8.68 (95% CI, –10.77 to –6.58) in the placebo arm. The LS mean difference was –3.68 (95% CI, –6.08 to –1.28; P = 0.0026). For patients in the ARCADIA 2 study, the LS mean change from baseline in total EASI score was –13.15 (95% CI, –14.57 to –11.73) in the nemolizumab arm and –9.02 (95% CI, –10.97 to –7.08) in the placebo arm. The LS mean difference between study arms was –4.13 (95% CI, –6.36 to –1.89; P = 0.0003).

Proportion of Patients With an Improvement of PP-NRS of 4 or More at Week 16

For patients in the ARCADIA 1 study, at week 16, 42.7% of patients in the nemolizumab arm and 17.8% of patients in the placebo arm reported a reduction in PP-NRS score of 4 points or greater. The strata-adjusted LS mean difference between arms was 24.9% (97.5% CI, 18.4% to 31.5%; P < 0.0001). For patients in the ARCADIA 2 study, at week 16, 41.0% of patients in the nemolizumab arm and 18.1% of patients in the placebo arm reported a reduction in PP-NRS score of 4 points or greater. The strata-adjusted LS mean difference between arms was 23.2% (97.5% CI, 16.1% to 30.3%; P < 0.0001).

Proportion of Patients With PP-NRS of Less Than 2 Points at Week 16

For patients in the ARCADIA 1 study, at week 16, 30.6% of patients in the nemolizumab arm and 11.2% of patients in the placebo arm reported a PP-NRS score of less than 2 points. The strata-adjusted LS mean difference between arms was –1.84% (97.5% CI, –2.21% to –1.47%; P < 0.0001). For patients in the ARCADIA 2 study, at week 16, 28.4% of patients in the nemolizumab arm and 11.3% of patients in the placebo arm reported a PP-NRS score of less than 2 points. The strata-adjusted LS mean difference between arms was 17.1% (97.5% CI, 10.9% to 23.3%; P < 0.0001).

Change From Baseline to Week 16 in PP-NRS Score

For patients in ARCADIA 1 study, the LS mean change from baseline in PP-NRS score was –3.77 (95% CI, –4.02 to –3.53) in the nemolizumab arm, and –1.94 (95% CI, –2.26 to –1.61) in the placebo arm. The LS mean difference was –1.84 (95% CI, –2.21 to –1.47; P < 0.0001). For patients in the ARCADIA 2 study, the LS mean change from baseline in PP-NRS score was –3.71 (95% CI, –3.94 to –3.47) in the nemolizumab arm and –1.95 (95% CI, –2.27 to –1.63) in the placebo arm. The LS mean difference between study arms was –1.76 (95% CI, –2.13 to –1.39; P < 0.0001).

Change From Baseline to Week 16 in DLQI Total Score

For adult patients in the ARCADIA 1 study, the LS mean change from baseline in DLQI total score was –7.76 (95% CI, –8.49 to –7.03) in the nemolizumab arm, and –5.26 (95% CI, –6.23 to –4.29) in the placebo arm. The LS mean difference was –2.50 (95% CI, –3.63 to –1.37; P < 0.0001). For adult patients in the ARCADIA 2 study, the LS mean change from baseline in DLQI total score was –6.96 (95% CI, –7.67 to –6.25) in the nemolizumab arm and –4.52 (95% CI, –5.48 to –3.56) in the placebo arm. The LS mean difference between study arms was –2.44 (95% CI, –3.58 to –1.30; P < 0.0001).

Change From Baseline to Week 16 in Children’s Dermatology Life Quality Index Total Score

For adolescent patients in the ARCADIA 1 study, the LS mean change from baseline in the Children’s Dermatology Life Quality Index (CDLQI) total score was –6.92 (95% CI, –9.26 to –4.58) in the nemolizumab arm, and –5.10 (95 CI, –7.71 to –2.50) in the placebo arm. The LS mean difference was –1.82 (95% CI, –4.88 to 1.25; P = 0.2454). For adolescent patients in the ARCADIA 2 study, the LS mean change from baseline in CDLQI total score was –7.40 (95% CI, –8.97 to –5.82) in the nemolizumab arm and –5.02 (95% CI, –7.48 to –2.56) in the placebo arm. The LS mean difference between study arms was –2.38 (95% CI, –5.03 to 0.27; P = 0.0788).

Incidence of Rescue Therapy Use

In the ARCADIA 1 study, during the initial period a total of 3.4% of patients in the nemolizumab arm and 4.0% of patients in the placebo arm reported using rescue therapy. In the ARCADIA 2 study, during the initial period a total of 2.1% of patients in the nemolizumab arm and 4.2% of patients in the placebo arm reported using rescue therapy.

ARCADIA 1 and ARCADIA 2 Studies: Pooled Maintenance Period

The ARCADIA 1 and ARCADIA 2 studies were noted by the sponsor to be replicate studies, having identical inclusion and exclusion criteria, study design, and analysis methods. In addition, a similar proportion of patients in each study were considered clinical responders per the study definition and continued on to the maintenance period, and study discontinuations or losses to follow-up were broadly similar. Therefore, the pooled maintenance period results were appraised in the report. Out of the intention-to-treat (ITT) population initially randomized in the initial period, 169 patients were rerandomized to the nemolizumab every 4 weeks arm, 169 to the nemolizumab every 8 weeks arm, and 169 to the placebo arm.

Proportion of Patients Maintaining IGA Success Until Week 48 From the Subset With IGA Success at Week 16

At week 48, 31.7% of patients in the nemolizumab every 4 weeks arm (n = 142), 37.3% of patients in the nemolizumab every 8 weeks arm (n = 142), and 31.3% of patients in the placebo arm (n = 131) reported maintaining IGA success at all study visits until week 48. The strata-adjusted LS mean difference between nemolizumab every 4 weeks and placebo was 0.5% (95% CI, –10.5% to 11.6%). The strata-adjusted LS mean difference between nemolizumab every 8 weeks and placebo was 6.0% (95% CI, –5.2% to 17.1%).

At week 48, 63.4% of patients in the nemolizumab every 4 weeks arm (n = 142), 64.1% of patients in the nemolizumab every 8 weeks arm (n = 142), and 55.0% of patients in the rerandomized placebo arm (n = 131) reported IGA success. The strata-adjusted LS mean difference between nemolizumab every 4 weeks and placebo was 8.5% (95% CI, –3.0% to 20.2%). The strata-adjusted LS mean difference between nemolizumab every 8 weeks and placebo was 9.1% (95% CI, –2.5% to 20.6%).

Proportion of Patients Maintaining EASI-75 Until Week 48 From the Subset With EASI-75 at Week 16

At week 48, 55.8% of patients in the nemolizumab every 4 weeks arm (n = 163), 58.9% of patients in the nemolizumab every 8 weeks arm (n = 163), and 44.6% of patients in the placebo arm (n = 163) reported maintaining EASI-75 at all study visits until week 48. The strata-adjusted LS mean difference between nemolizumab every 4 weeks and placebo was 11.2% (95% CI, 0.3% to 22.1%). The strata-adjusted LS mean difference between nemolizumab every 8 weeks and placebo was 14.1% (95% CI, 3.3% to 24.9%).

At week 48, 76.3% of patients in the nemolizumab every 4 weeks arm (n = 169), 75.7% of patients in the nemolizumab every 8 weeks arm (n = 169), and 63.9% of patients in the rerandomized placebo arm (n = 169) reported attaining EASI-75. The strata-adjusted LS mean difference between nemolizumab every 4 weeks and placebo was 12.4% (95% CI, 2.7% to 22.0%). The strata-adjusted LS mean difference between nemolizumab every 8 weeks and placebo was 11.8% (95% CI, 2.1% to 21.5%).

Change From Week 16 to Week 48 in EASI

The LS mean change from baseline in EASI score was 2.01 (95% CI, –0.16 to 4.19; n = 169) in the nemolizumab every 4 weeks arm; 2.98 (95% CI, 0.80 to 5.17; n = 169) in the nemolizumab every 8 weeks arm, and 6.79 (95% CI, 4.62 to 8.96; n = 169) in the placebo arm. The LS mean difference from placebo was –4.78 (95% CI, –7.85 to –1.70) in the nemolizumab every 4 weeks arm and –3.80 (95% CI, –6.89 to –0.72) in the nemolizumab every 8 weeks arm.

Proportion of Patients With PP-NRS Improvement of 4 or More From Initial Baseline to Week 48 From the Subset With PP-NRS of 4 or More at Week 16

At week 48, of the subset of patients who had a PP-NRS score improvement of 4 points or more at week 16, 46.9% of patients in the nemolizumab every 4 weeks arm (n = 98), 40.4% of patients in the nemolizumab every 8 weeks arm (n = 94), and 31.3% of patients in the placebo arm (n = 83) reported an improvement in PP-NRS of 4 points or more from the initial period baseline. The strata-adjusted proportion difference versus placebo was 15.4% (95% CI, 1.3% to 29.4%) in the nemolizumab every 4 weeks arm and 9.1% (95% CI, –5.0% to 23.2%) in the nemolizumab every 8 weeks arm.

At week 48 of the entire maintenance period population, 36.7% of patients in the nemolizumab every 4 weeks arm (n = 169), 29.0% of patients in the nemolizumab every 8 weeks arm (n = 169), and 21.3% of patients in the placebo arm (n = 169) reported an improvement in PP-NRS of 4 points or more from the initial period baseline. The strata-adjusted proportion difference versus placebo was 15.4% (95% CI, 5.8% to 24.9%) in the nemolizumab every 4 weeks arm and 7.7% (95% CI, –1.5% to 16.9%) in the nemolizumab every 8 weeks arm.

Proportion of Patients With PP-NRS of Less Than 2 at Week 48 From the Subset With PP-NRS of Less Than 2 at Week 16

At week 48, 39.7% of patients in the nemolizumab every 4 weeks arm (n = 169), 43.0% of patients in the nemolizumab every 8 weeks arm (n = 169), and 34.4% of patients in the placebo arm (n = 169) reported a PP-NRS score of less than 2. The strata-adjusted proportion difference versus placebo was 5.9% (95% CI, –10.1% to 21.8%) in the nemolizumab every 4 weeks arm and 8.6% (95% CI, –7.4% to 24.6%) in the nemolizumab every 8 weeks arm.

At week 48, of the entire maintenance period population, 30.2% of patients in the nemolizumab every 4 weeks arm (n = 169), 27.8% of patients in the nemolizumab every 8 weeks arm (n = 169), and 18.9% of patients in the placebo arm (n = 169) reported a PP-NRS score of 2 points or less. The strata-adjusted proportion difference versus placebo was 11.2% (95% CI, 2.1% to 20.3%) in the nemolizumab every 4 weeks arm and 8.9% (95% CI, –0.1% to 17.8%) in the nemolizumab every 8 weeks arm.

Change From Week 16 to Week 48 in PP-NRS

The LS mean change from baseline in PP-NRS score was –0.46 (95% CI, 0.88 to –0.05; n = 143) in the nemolizumab every 4 weeks arm; 0.10 (95% CI, –0.29 to 0.48; n = 155) in the nemolizumab every 8 weeks arm; and 1.35 (95% CI, 0.95 to 1.74; n = 152) in the placebo arm. The LS mean difference from placebo was –1.81 (95% CI, –2.39 to –1.23) in the nemolizumab every 4 weeks arm and –1.25 (95% CI, –1.79 to –0.71) in the nemolizumab every 8 weeks arm.

Change From Week 16 to Week 48 in DLQI Total Score

The LS mean change from baseline in DLQI score was –0.43 (95% CI, –1.11 to 0.24; n = 119) in the nemolizumab every 4 weeks arm, 0.14 (95% CI, –0.50 to 0.79; n = 130) in the nemolizumab every 8 weeks arm, and 0.70 (95% CI, 0.00 to 1.40; n = 110) in the placebo arm. The LS mean difference from placebo was –1.14 (95% CI, –2.11 to –0.16) in the nemolizumab every 4 weeks arm and –0.56 (95% CI, –1.52 to 0.40) in the nemolizumab every 8 weeks arm.

Change From Week 16 to Week 48 in CDLQI Total Score

The LS mean change from baseline in CDLQI score was –0.54 (95% CI, –1.45 to 0.36; n = 29) in the nemolizumab every 4 weeks arm, –1.27 (95% CI, –2.30 to –0.23; n = 22) in the nemolizumab every 8 weeks arm, and 0.34 (95% CI, –0.71 to 1.38; n = 22) in the placebo arm. The LS mean difference from placebo was –0.88 (95% CI, –2.27 to 0.51) in the nemolizumab every 4 weeks arm and –1.60 (95% CI, –3.08 to –0.12) in the nemolizumab every 8 weeks arm.

Incidence of Rescue Therapy Use

In the maintenance period, a total of 1.2% of patients in the nemolizumab every 4 weeks arm, 2.4% in the nemolizumab every 8 weeks arm, and 5.3% of patients in the placebo arm reported using rescue therapy.

ARCADIA CYCLO Study

Proportion of Patients With EASI-75

At week 16 in the ARCADIA CYCLO study, 47.1% of patients in the nemolizumab arm and 34.8% of patients in the placebo arm reported EASI-75. The strata-adjusted difference between the study arms was 12.2% (95% CI, 0.7% to 23.6%; P = 0.040).

Proportion of Patients With Prior CsA Use With EASI-75

██ ████ ███ █████ ██ ████████ ██ ███ ███████████ ███ ██ █ ████ ███ █████ ██ ████████ ██ ███ ███████ ███ ██ █ ████ ████████ ████████ ███ ███████████████ ██████████ ███████ ███ █████ ████ ███ █████ ████ ███ ████ ██ ██████████████.

Percent Change From Baseline in EASI

███ ██ ████ ███████ ██████ ████ ████████ ██ ████ █████ █████ ██ ████ ██ ███ ███████ ████ ███ ███████ ██ ████████ ██ ███ ███████████ ████ ███ ███████ ████ ███ ███████ ██ ████████ ██ ███ ███████ ████ ███ ██ ████ ██████████ ██████ ███████ ███ ██████ ████ ███ ███████ ██ ██████ ████████.

Proportion of Patients With an Improvement of PP-NRS of 4 or More From Baseline to Week 16

At week 16, 39.1% of patients in the nemolizumab arm and 17.4% of patients in the placebo arm reported an improvement in PP-NRS score of 4 points or greater. The strata-adjusted difference versus placebo was 21.7% (95% CI, 11.4% to 32.0%; P < 0.001).

Proportion of Patients With PP-NRS of Less Than 2 at Week 16

██ ████ ███ █████ ██ ████████ ██ ███ ███████████ ███ ███ ████ ██ ████████ ██ ███ ███████ ███ ████████ █ ██████ █████ ██ █ ██████ ██ █████ ███ ██ ████ ██████████ ██████ ███████ ███ █████ ████ ███ ████ ██ █████████████.

Percent Change From Baseline to Week 16 in PP-NRS

███ ██ ████ ███████ ██████ ██ ██████ ████ ████████ ███ ███████ ████ ███ ███████ ██ ████████ ██ ███ ███████████ ███ ███ ███████ ████ ███ ███████ ██ ████████ ██ ███ ███████ ████ ███ ██ ████ ██████████ ██████ ███████ ███ ███████ ████ ███ ███████ ██ ██████████████.

Proportion of Patients With IGA Success

██ ████ ███ █ █████ ██ █████ ██ ████████ ██ ███ ███████████ ███ ███ █████ ██ ████████ ██ ███ ███████ ███ ████████ ███ ███████ ████████ ██ ███ █ ██ █ ███ | | ███████ █████████ ████ ██████████ ███ ██ ████ ██████████ ██████ ███████ ███ ████ ████ ███ █████ ██ ██████ ████████.

Change From Baseline to Week 16 in DLQI Total Score

███ ██ ████ ██████ ████ ████████ ██ ████ █████ █████ ███ █████ ████ ███ █████ ██ ██████ ██ ███ ███████████ ███ ███ █████ ████ ███ █████ ██ ██████ ██ ███ ███████ ████ ███ ██ ████ ██████████ ██████ ███████ ███ █████ ████ ███ █████ ██ ██████ | | ███████.

Incidence of Rescue Treatment Use

██ ██ ████ ███ █ █████ ██ ████ ██ ████████ ██ ███ ███████████ ███ ███ ████ ██ ████████ ██ ███ ███████ ███ ████████ ███ ███ ██ █ ██ ████ ██████ ████████████ ███ ███████████████ ██████████ ██████ ███████ ███ ███ ████ ███ ████ ██ ████ | | ████████.

Harms Results

Adverse Events

ARCADIA 1 Study

During the initial period of the ARCADIA 1 study, 49.7% of patients in the nemolizumab arm and 45.5% of patients in the placebo arm reported at least 1 AE. The most commonly reported AEs during this period were infections and infestations (18.3% in the nemolizumab arm and 20.3% in the placebo arm); skin and SC tissue disorders (18.3% in the nemolizumab arm and 14.0% in the placebo arm); and respiratory, thoracic, or mediastinal disorders (9.4% in the nemolizumab arm and 8.7% in the placebo arm).

During the maintenance period of the ARCADIA 1 study, 58.2% of patients in the nemolizumab every 4 weeks arm, 55.6% in the nemolizumab every 8 weeks arm, 58.2% in the rerandomized placebo arm, and 55.0% in the carryover placebo arm reported AEs. The most common AEs were infections and infestations (34.1% in the nemolizumab every 4 weeks arm, 34.4% in the nemolizumab every 8 weeks arm, 34.1% in the rerandomized placebo arm, and 34.0% in the nonrandomized placebo arm); skin and SC issues (12.1% in the nemolizumab every 4 weeks arm, 11.1% in the nemolizumab every 8 weeks arm, 14.3% in the rerandomized placebo arm, and 14.0% in the nonrandomized placebo arm); and respiratory, thoracic, and mediastinal disorders (12.1% in the nemolizumab every 4 weeks arm, 11.1% in the nemolizumab every 8 weeks arm, 13.2% in the rerandomized placebo arm, and 7.0% in the nonrandomized placebo arm). There was a numerically higher proportion of patients reporting infections and infestations in the maintenance period relative to the initial period, and a numerically higher number of patients in the nemolizumab every 4 weeks arm who reported gastrointestinal disorders (12.1%) relative to the other arms in the maintenance period (ranging from 5.6% to 7.7% of patients).

ARCADIA 2 Study

During the initial period of the ARCADIA 2 study, 41.4% of patients in the nemolizumab arm and 44.5% of patients in the placebo arm reported at least 1 AE. The most commonly reported AEs during this period were infections and infestations (17.0% in the nemolizumab arm and 20.2% in the placebo arm); skin and SC tissue disorders (12.1% in the nemolizumab arm and 9.9% in the placebo arm); and respiratory, thoracic, or mediastinal disorders (5.6% in the nemolizumab arm and 5.3% in the placebo arm).

During the maintenance period of the ARCADIA 1 study, 48.1% of patients in the nemolizumab every 4 weeks arm, 51.9% in the nemolizumab every 8 weeks arm, 58.4% in the rerandomized placebo arm, and 44.0% in the carryover placebo arm reported AEs. The most common AEs were infections and infestations (27.8% in the nemolizumab every 4 weeks arm, 22.1% in the nemolizumab every 8 weeks arm, 31.2% in the rerandomized placebo arm, and 28.6% in the nonrandomized placebo arm); skin and SC issues (13.9% in the nemolizumab every 4 weeks arm, 10.4% in the nemolizumab every 8 weeks arm, 13.0% in the rerandomized placebo arm, and 9.5% in the nonrandomized placebo arm); and respiratory, thoracic, and mediastinal disorders (10.1% in the nemolizumab every 4 weeks arm, 6.5% in the nemolizumab every 8 weeks arm, 11.7% in the rerandomized placebo arm, and 2.4% in the nonrandomized placebo arm). A numerically higher proportion of patients in the rerandomized placebo arm reported COVID-19 infection (16.9%) relative to other study arms (ranging from 3.8% to 7.1%).

ARCADIA CYCLO Study

██ ███████ ██████ █████ ██ ████████ ██ ███ ███████████ ███ ███ █████ ██ ████████ ██ ███ ███████ ███ ████████ ████ ███ ████ ████████ ████████ ███ ████ ██████████ ███ ████████████ ██████ ████████████ █████ █████████ ███ ████ ███ ████████████ █████████ █████ ████████████ █████ █████████ █ ███████████ ██████ ██████████ ██ ████████ ██ ███ ███████████ ███ ████████ ████████████ ████████ ███ ███████████ ███ ██████ ████████ ██ ███ ███████ ███ ███████ ██ ████ ██ ███████████████ ███ ██████████ ██████ █████████ █████ ████████████ ████ █████████.

Serious Adverse Events

ARCADIA 1 Study

During the initial period of the ARCADIA 1 study, 1.0% of patients in the nemolizumab arm and 1.2% of patients in the placebo arm reported any serious AE (SAE). AD SAEs were reported by 2 patients in the nemolizumab arm, and 3 patients in the placebo arm. The remaining SAEs by preferred term were reported in fewer than 2 patients.

During the maintenance phase of the ARCADIA 1 study, 4.4% of patients in the nemolizumab every 4 weeks arm, 3.3% in the nemolizumab every 8 weeks arm, 2.2% in the rerandomized placebo arm, and 1.0% in the nonrandomized placebo arm reported any SAEs. The SAEs by preferred term were reported in fewer than 2 patients.

ARCADIA 2 Study

During the initial period of the ARCADIA 2 study, 2.5% of patients in the nemolizumab arm and 1.1% of patients in the placebo arm reported any SAE. SAEs reported in more than 2 patients in the nemolizumab arm included 3 patients reporting infections and infestations, 3 patients reporting musculoskeletal and connective tissue disorders, 3 patients reporting skin and SC tissue disorders, and 2 patients reporting gastrointestinal disorders.

During the maintenance phase of the ARCADIA 2 study, 7.6% of patients in the nemolizumab every 4 weeks arm, 0% in the nemolizumab every 8 weeks arm, 2.6% in the rerandomized placebo arm, and 1.2% in the nonrandomized placebo arm reported any SAEs. There were no SAEs reported in more than 2 patients.

ARCADIA CYCLO Study

During the ARCADIA CYCLO study, 2.2% of patients in the nemolizumab arm and 1.5% of patients in the placebo arm reported SAEs. In the nemolizumab arm, SAEs were reported in 2 patients in the category of infections and infestations. In the placebo arm, SAEs by preferred term were reported in fewer than 2 patients.

Withdrawals Due to AEs

ARCADIA 1 Study

During the initial period of the ARCADIA 1 study, 1.8% of patients in the nemolizumab arm and 4.0% of patients in the placebo arm discontinued study drug treatment due to AEs. In the maintenance period, 1.1% of patients in the nemolizumab every 4 weeks arm, 3.3% in the nemolizumab every 8 weeks arm, 2.2% in the rerandomized placebo arm, and 2.0% in the carryover placebo arm discontinued due to AEs. In both the initial and maintenance periods, AD was the most common reason for discontinuation (initial period: 1.6% of patients in the nemolizumab arm and 4.0% of patients in the placebo arm; maintenance period: 0% in the nemolizumab every 4 weeks arm, 2.2% in the nemolizumab every 8 weeks arm, 1.1% in the rerandomized placebo arm, and 2.0% in the carryover placebo arm).

ARCADIA 2 Study

During the initial period of the ARCADIA 2 study, 3.5% of patients in the nemolizumab arm and 1.1% of patients in the placebo arm reported discontinuing due to AEs. In the maintenance period, 3.8% of patients in the nemolizumab every 4 weeks arm, 2.6% in the nemolizumab every 8 weeks arm, 3.9% in the rerandomized placebo arm, and 2.4% in the nonrandomized placebo arm withdrew due to AEs. In both the initial and maintenance periods, the most common reason for withdrawal of the study drug was AD (initial period: 1.7% of patients in the nemolizumab arm and 0.8% in the placebo arm; maintenance period: 2.5% in the nemolizumab every 4 weeks arm, 2.6% in the nemolizumab every 8 weeks arm, 2.6% in the rerandomized placebo arm, and 2.4% in the carryover placebo arm).

ARCADIA CYCLO Study

████ ██ ████ ██ ████████ ██ ███ ███████████ ███ ███ ████ ██ ████████ ██ ███ ███████ ███ ███████ ███ █████ █████ ███ ███████ ███ ████████ ██ ███ ███████████ ███ ████ █ ███████ ██████ ████ ██ ██████████ ██████ ████ █████████████████ ███ ███ ███ ███████ ███ ████████ ██ ███ ███████ ███ ████ █ ███████ ██████ ████ ██ ███████████████ █████████ █████████ ███ ███.

Mortality

There were no deaths reported in the ARCADIA 1, ARCADIA 2, or ARCADIA CYCLO studies.

Notable Harms

The AEs of special interest (AESIs) in the ARCADIA 1 and ARCADIA 2 studies identified in the submission were injection site reactions, newly diagnosed or worsening asthma, infections, peripheral edema, and elevated ALT or AST (> 3 × the upper limit of normal [ULN]) in combination with elevated bilirubin (> 2 × ULN). The AESIs identified in the submission for the ARCADIA CYCLO study were injection-related reactions, drug hypersensitivity, infections, COVID-19 infection, asthma, peak expiratory flow rate decrease, and peripheral edema. The submission did not contain details on whether these AESIs were prespecified or identified during the conduct of the studies.

ARCADIA 1 Study

During the initial period, injection-related reactions were reported in 0.2% of patients in the nemolizumab arm and 0 patients in the placebo arm; 0 patients reported injection-related reactions in the maintenance period. Newly diagnosed or worsening asthma was reported in 5.2% of patients in the nemolizumab arm and 3.4% of patients in the placebo arm during the initial period; during the maintenance period, 3.3% of patients in the nemolizumab every 4 weeks arm, 6.7% in the nemolizumab every 8 weeks arm, 2.2% in the rerandomized placebo arm, and 5.0% in the carryover placebo arm reported this AE. Infections were reported in 3.2% of patients in the nemolizumab arm and 3.1% in the placebo arm during the initial period; during the maintenance period, 13.2% of patients in the nemolizumab every 4 weeks arm, 12.2% in the nemolizumab every 8 weeks arm, 13.2% in the rerandomized placebo arm, and 12.0% in the carryover placebo arm reported this AE. Peripheral edema was reported in 1.1% of patients in the nemolizumab arm and 0.3% in the placebo arm during the initial period; in the maintenance period, 1.1% of patients in the nemolizumab every 4 weeks arm, 2.2% in the nemolizumab every 8 weeks arm, 1.1% in the rerandomized placebo arm, and 0 in the carryover placebo arm reported this AE. Elevated ALT or AST in combination with elevated bilirubin was not reported during the trial.

ARCADIA 2 Study

During the initial period and maintenance periods of the ARCADIA 2 study, no injection-related reactions were reported. Newly diagnosed or worsening asthma was reported in 2.9% of patients in the nemolizumab arm and 3.0% of patients in the placebo arm during the initial period; in the maintenance period, 5.1% of patients in the nemolizumab every 4 weeks arm, 0% in the nemolizumab every 8 weeks arm, 2.6% in the rerandomized placebo arm, and 0% in the carryover placebo arm reported this AE. Infections were reported in 3.9% of patients in the nemolizumab arm and 4.6% in the placebo arm during the initial period; in the maintenance period, 5.1% of patients in the nemolizumab every 4 weeks arm, 7.8% in the nemolizumab every 8 weeks arm, 16.9% in the rerandomized placebo arm, and 10.7% in the carryover placebo arm reported this AE. Peripheral edema was reported in 2.3% of patients in the nemolizumab arm and 0.4% in the placebo arm during the initial period; in the maintenance period, 0% of patients in the nemolizumab every 4 weeks arm, 0% in the nemolizumab every 8 weeks arm, 2.6% in the rerandomized placebo arm, and 0% in the carryover placebo arm reported this AE. Elevated ALT or AST in combination with elevated bilirubin was not reported during the trial.

ARCADIA CYCLO Study

███ ████ ████████ ████████ ████ ██ ███████ █████ ████ ████████ ██████████ █████ ██ ████████ ██ ███ ███████████ ████ ████ ██ ████████ ██ ███ ███████ █████ ████████ ██ ██████ █████ ████████████ ████ █████████ ███ ████ ██████████ ████ ████ █████████ █████ ████████████ ████ █████████ ████ ████████████████ ███ ████████ ██ ████ ██ ████████ ██ ███ ███████████ ███ ████ ██ ██ ███████ █████ ███ ███ ███████ ██████ ██ ███ ███████ ███ ████████ ██ █████████████████ ████████ ██ █████████.

Critical Appraisal

Internal Validity

In all ARCADIA studies (ARCADIA 1, ARCADIA 2, and ARCADIA CYCLO), the procedures for randomization, treatment allocation, blinding, and study drug administration were all described and likely at low risk of bias; however, the study design is subject to some limitations. In the ARCADIA 1 and ARCADIA 2 studies, only patients who met the study definition of clinical responder (an IGA score of 0 [clear] or 1 [almost clear] or EASI-75) continued from the initial period on to the maintenance period. This imparts a response bias for the maintenance phase as the patients included in the maintenance phase are an enriched population of patients who have already responded to treatment by 16 weeks. It is not known whether the patients who met the study definition of nonresponder had specific common characteristics which would have impacted their likelihood of responding to treatment. The duration of the initial treatment period, while generally consistent with studies for other treatments reimbursed for moderate to severe AD, may not be long enough to obtain complete treatment response, as the clinical experts consulted for this review noted that a trial of up to 6 months is used in clinical practice, which is a relevant consideration for all ARCADIA trials. There was no run-in period for the ARCADIA CYCLO study, while patients’ background treatment was standardized in the ARCADIA 1 and ARCADIA 2 studies. Because patients in the ARCADIA CYCLO study have more severe disease, this may impact the certainty in the results as patients’ background therapies in the ARCADIA CYCLO study could have differed in efficacy and type of product.

With regards to the statistical analysis across trials, the assessment of outcomes for the coprimary and key secondary outcomes in the initial period is likely at lower risk of bias due to being controlled for multiplicity; however, the order of hypothesis testing for the outcomes is unknown and therefore the alpha spending among the list of outcomes, or when testing might have ended, is also unknown. In the ARCADIA 1 and ARCADIA 2 studies, the results of the sensitivity analyses testing different methods of classifying missing clinical responders and nonresponders were also consistent with the main analysis, which suggests the results are reasonably robust. However the statistical analysis methods are subject to some limitations that impact internal validity, and which increase the uncertainty in the results. In the ARCADIA 1 and ARCADIA 2 studies, there were several protocol amendments, although it is unclear how many patients were impacted by the modification, or what impact this may have had on the results. Per the individual statistical analysis plan for the ARCADIA 1 and ARCADIA 2 studies, no hypothesis testing was conducted on outcomes in the maintenance phase; however, the pooled Clinical Study Report described statistical tests comparing nemolizumab arms to the rerandomized placebo group. In addition, there was no control of multiplicity for the secondary outcomes in the initial period (change from baseline in PP-NRS, EASI, DLQI, and CDLQI) and maintenance period (all outcomes); therefore, there is a risk of increased type I error for these outcomes. The study populations of the ARCADIA 1 and ARCADIA 2 studies appeared broadly similar in terms of disposition, and the studies employed identical designs, inclusion and exclusion criteria, as well as analysis methods. However, the statistical limitations mean that results from the maintenance period should only be considered supportive evidence of the impact of nemolizumab.

External Validity

Patients enrolled in the ARCADIA 1 and ARCADIA 2 studies were required to have an EASI score of 16 or more (≥ 20 in the ARCADIA CYCLO study), and an IGA score of 3 or more in both studies. An EASI score of 7.1 to 21.0 is considered moderate AD, and severe AD ranges from 21.1 to 50.0.⁹ An IGA score of 3 is considered moderate AD. As such, patients enrolled in the ARCADIA trials may have more severe disease, and therefore the results may not be generalizable to all patients with moderate AD, depending on the disease severity criteria used. However, it is worth noting that an EASI score of 16 and an IGA score of 3 are consistent with other drugs evaluated for moderate to severe AD.

All the ARCADIA studies excluded patients with asthma, chronic obstructive pulmonary disease (COPD), and certain medications which cause sedative effects. Patients who had not improved after 16 weeks of treatment with dupilumab were also excluded from the ARCADIA 1 and ARCADIA 2 trials. The reimbursement request includes patients who are refractory to systemic immunosuppressant therapies, which could include dupilumab; therefore, results from the trials will not be generalizable to patients with a history of dupilumab exposure. The clinical experts consulted for this review noted that these patients could be considered candidates for treatment with nemolizumab in clinical practice. In addition, the fact that only patients who met the study definition of clinical responders were kept in the maintenance phase of the ARCADIA 1 and ARCADIA 2 studies means any results after 16 weeks may not include all the patients represented in the study enrolment criteria.

The ARCADIA trials were all placebo-controlled trials, which allows for adequate evaluation of the treatment effect of nemolizumab; however, it may overestimate the treatment effects. Patients in the ARCADIA 1 and ARCADIA 2 studies had their background therapy stabilized during the run-in period before randomization. There was a high placebo response observed in the ARCADIA trials, although it is unclear whether this is due to the background therapy received by all patients. If so, this also impacts patients receiving nemolizumab. Nemolizumab monotherapy was not evaluated in the evidence submitted, and therefore delineating the true impact of nemolizumab on the disease remains unknown and may overestimate the results of the ARCADIA studies, although, it is worth noting that patients in the real world would continue to receive their background therapies, as was done in the ARCADIA studies. Across all 3 trials, treatment adherence was reported to be high ██ █████; the levels of adherence observed may not be representative of clinical practice settings.

Most patients in the ARCADIA 1 and ARCADIA 2 studies were aged older than 18 years, while only 13.7% to 17.4% of patients were between the ages of 12 and 17 years. Outcomes specific to this population (i.e., the CDLQI) were highly uncertain because they come from a small proportion of patients in the study. Additionally, other results may not be generalized to this population; however, subgroup analyses by age group were generally consistent with the primary analysis.

In addition to the study design, the treatment history described in the inclusion and exclusion criteria does not exactly match the requested reimbursement criteria; the ARCADIA 1 and ARCADIA 2 trials did not require patients to have previous exposure to immunomodulatory therapies, and the ARCADIA CYCLO study only required exposure or inadvisability for CsA. In the ARCADIA 1 and ARCADIA 2 studies, less than 40% of patients had prior exposure to immunomodulatory treatment, less than 15% had exposure to CsA, and approximately 5% had had exposure to dupilumab. Therefore, it is uncertain how applicable the overall results will be to patients with these treatment histories versus patients who are not previously treated with systemic therapies, or who have been treated with other advanced systemic therapies. Subgroup analyses conducted in the ARCADIA 1 and ARCADIA 2 studies on prior use of systemic therapy, biologic therapy, or dupilumab did not show consistent trends between the 2 studies and some subgroups had small sample sizes, thus, firm conclusions could not be drawn. In addition, according to the clinical experts, the criteria for refractoriness to CsA in the ARCADIA CYCLO study were subjective and therefore might be representative of all patients who might not respond to CsA.

GRADE Summary of Findings and Certainty of the Evidence

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

ARCADIA 1 and ARCADIA 2 studies:

• Proportion of patients with IGA success at week 16

• Proportion of patients with EASI-75 at week 16

• Proportion of patients with PP-NRS score improvement of 4 or more at week 16

• Proportion of patients with PP-NRS score less than 2 at week 16

• Change from baseline to week 16 in DLQI

• Change from baseline to week 16 in CDLQI

• Harms (injection site reactions, newly diagnosed or worsening asthma)

ARCADIA CYCLO study:

• Proportion of patients with EASI-75 at week 16

• Proportion of patients with PP-NRS score improvement of 4 points or more at week 16

• Proportion of patients with PP-NRS score of 2 or more at week 16

• Proportion of patients with IGA success at week 16

• Change from baseline in DLQI score at week 16

• Harms (injection site reactions, newly diagnosed or worsening asthma)

Table 2: Summary of Findings for Nemolizumab vs. Placebo for Patients Aged 12 Years and Older With Moderate to Severe Atopic Dermatitis

CDLQI = Children’s Dermatology Life Quality Index; CI = confidence interval; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; EASI-75 = ≥ 75% in Eczema Area and Severity Index from baseline; HRQoL = health-related quality of life; IGA = Investigator Global Assessment; LS = least squares; NR = not reported; PP-NRS = Peak Pruritus Numeric Rating Scale; RCT = randomized controlled trial; SAE = serious adverse event; vs. = versus.

Note: Study limitations (which refers 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.

^aThe IGA measures the investigator’s global assessment of the patient’s overall severity of atopic dermatitis at that visit, based on a static, numeric 5-point scale ranging from 0 (clear) to 4 (severe). IGA success was defined as an IGA score of 0 or 1 and at least a 2-point reduction from baseline.

^bRated down 1 level for serious imprecision. The CI for differences between groups included the potential for little to no difference (based on the threshold for a clinically important between-group difference of 100 per 1,000 provided by the clinical experts consulted for this review).

^cThe EASI is a composite index, based on the physician’s assessment of 4 clinical signs of the disease (erythema, infiltration or papulation, excoriation, and lichenification) and the extent of body surface area involved at that visit. It is scored from 0 to 72, with higher scores indicating greater disease severity and/or extent of disease.

^dThe PP-NRS is a patient-reported, single-item, daily, 11-point scale. The scale is used by patients to rate their worst itch severity over the previous 24 hours, with 0 indicating no itch and 10 indicating the worst itch imaginable.

^eThe DLQI (for patients aged ≥ 16 years) and CDLQI (for those aged < 16 years) are patient-reported, 10-item, HRQoL questionnaires that cover 6 domains (symptoms and feelings, daily activities, leisure, work and school, personal relationships, and treatment) over the previous week. The total score ranges from 0 (no impact of skin disease on quality of life) to 30 (maximum impact on quality of life).

^fRated down 1 level for serious imprecision. The CI for differences between groups includes the potential for little to no difference (based on the threshold for clinically important between-group difference of 5 points provided by the clinical experts consulted for this review).

^gRated down 1 level for serious imprecision due to the low number of events and lack of 95% CI. The lack of CIs means it cannot be determined whether there is the possibility of benefit, no benefit, or harm in the assessment of this outcome.

^hRated down 1 level for very serious indirectness. The duration of follow-up was limited to 16 weeks, which is an insufficient duration of time to detect all occurrences of this harm.

Sources: Details included in the table are from the sponsor’s summary of clinical evidence,¹⁰ the ARCADIA 1 Clinical Study Report,¹¹ and the ARCADIA 2 Clinical Study Report.¹²

Table 3: Summary of Findings for Nemolizumab vs. Placebo for Patients Aged 12 Years and Older With Moderate to Severe Atopic Dermatitis Previously Exposed to Cyclosporine or for Whom Cyclosporine is Medically Inadvisable

CDLQI = Children’s Dermatology Life Quality Index; CI = confidence interval; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; EASI-75 = ≥ 75% improvement in Eczema Area and Severity Index from baseline; HRQoL = health-related quality of life; IGA = Investigator Global Assessment; LS = least squares; NR = not reported; PP-NRS = Peak Pruritus Numeric Rating Scale; RCT = randomized controlled trial; SAE = serious adverse event; vs. = versus.

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.

^aThe EASI is a composite index, based on the physician’s assessment of 4 clinical signs of the disease (erythema, infiltration or papulation, excoriation, and lichenification) and the extent of body surface area involved at that visit. It is scored from 0 to 72, with higher scores indicating greater disease severity and/or extent of disease.

^bRated down 1 level for serious imprecision. The CI for differences between groups included the potential for little to no difference (based on the threshold for a clinically important between-group difference of 100 per 1,000 provided by the clinical experts consulted for this review).

^cThe PP-NRS is a patient-reported, single-item, daily, 11-point scale. The scale is used by patients to rate their worst itch severity over the previous 24 hours, with 0 indicating no itch and 10 indicating the worst itch imaginable.

^dThe IGA measures the investigator’s global assessment of the patient’s overall severity of atopic dermatitis at that visit, based on a static, numeric 5-point scale ranging from 0 (clear) to 4 (severe). IGA success was defined as an IGA score of 0 or 1 and at least a 2-point reduction from baseline.

^eThe DLQI (for patients aged ≥ 16 years) and CDLQI (for those aged < 16 years) are patient-reported, 10-item, HRQoL questionnaires that cover 6 domains (symptoms and feelings, daily activities, leisure, work and school, personal relationships, and treatment) over the previous week. The total score ranges from 0 (no impact of skin disease on quality of life) to 30 (maximum impact on quality of life).

^fRated down 1 level for serious imprecision. The CI for differences between groups included the potential for little to no difference (based on the threshold for a clinically important between-group difference of 5 points provided by the clinical experts consulted for this review).

^gRated down 1 level for serious imprecision due to the low number of events and lack of 95% CI. The lack of CIs means it cannot be determined whether there is the possibility of benefit, no benefit, or harm in the assessment of this outcome.

^hRated down 1 level for serious indirectness. The duration of follow-up was limited to 16 weeks, which is insufficient evidence to detect all occurrences of this harm.

Sources: Details included in the table are from the sponsor’s summary of clinical evidence¹⁰ and the ARCADIA CYCLO Clinical Study Report.¹³

Long-Term Extension Studies

Description of Studies

The ARCADIA long-term extension (LTE) study (NCT03989206; N = 1,751) is an ongoing phase III, single-arm, multicentre, prospective study to evaluate the long-term safety and efficacy of nemolizumab in adult and adolescent patients with moderate to severe AD, when administered with background topical corticosteroid (TCS), with or without topical calcineurin inhibitor (TCI). The ARCADIA LTE study design consisted of a 4-week screening period, a 200-week treatment period, and an 8-week follow-up period. Participants received nemolizumab (30 mg) every 4 weeks via SC injection in the treatment period. Results were from 1 interim analysis (cut-off date of September 30, 2022) up to 56 weeks of treatment, as the sponsor noted that sample sizes beyond this time point were too small to draw meaningful conclusions. The study is estimated to be completed in August 2026.

The primary outcomes in the ARCADIA LTE study were the incidence and severity of AEs throughout the study, the incidence of SAEs throughout the study, and the incidence and severity of AESIs throughout the study. Secondary outcomes aligned with outcomes assessed in the ARCADIA 1 and ARCADIA 2 studies that were identified as important to this review, and included the IGA, EASI, and DLQI and CDLQI.

The study population included patients from prior nemolizumab AD studies, including the ARCADIA 1, ARCADIA 2, and ARCADIA CYCLO studies, and adolescents from selected international sites who had not previously participated in a nemolizumab study. Inclusion criteria for the ARCADIA LTE study were consistent with that of the lead-in studies. Patients were excluded if they had a history of COPD and/or chronic bronchitis, body weight of less than 30 kg, uncontrolled asthma in the preceding 3 months, asthma exacerbation requiring hospitalization in the preceding 12 months, had experienced an AE in a previous nemolizumab study, or had received restricted prior treatments.

Efficacy Results

Proportion of Patients With IGA Success

At LTE study baseline, 23.9% of all patients experienced IGA success (defined as an IGA score between 0 to 1). With treatment before the LTE study, 28.5% of patients previously treated with nemolizumab and 17.7% of patients not previously treated with nemolizumab experienced an IGA score of 0 to 1. At week 20, 42.5% of all patients in the LTE study experienced IGA success. With regards to prior treatment, 45.6% of patients previously treated with nemolizumab and 36.7% of patients not previously treated with nemolizumab experienced an IGA score of 0 to 1. At week 56, 48.9% of all patients in the LTE study experienced IGA success. With prior treatment, 47.1% of patients previously treated with nemolizumab and 49.0% of patients not previously treated with nemolizumab experienced an IGA score of 0 to 1.

Proportion of Patients With EASI-75

At LTE study baseline, 33.7% of all patients achieved EASI-75 improvement from lead-in study baseline. With prior treatment, 38.1% of patients previously treated with nemolizumab and 24.0% of patients not previously treated with nemolizumab achieved EASI-75 improvement from lead-in baseline. At week 20, 66.3% of patients achieved EASI-75 improvement from lead-in baseline and 34% achieved EASI-75 improvement from LTE study baseline. With regards to prior treatment, the proportion of patients achieving EASI-75 from lead-in and LTE study baseline were 69.3% and 32.6% for those previously treated with nemolizumab and 61.5% and 36.2% for patients not previously treated with nemolizumab, respectively. At week 56, 75.6% of patients achieved EASI-75 improvement from lead-in baseline and 57.4% achieved EASI-75 improvement from LTE study baseline. With regards to prior treatment, the proportion of patients achieving EASI-75 from lead-in and LTE study baseline were 73.0% and 52.5% for those previously treated with nemolizumab and 78.7% and 62.3% for patients not previously treated with nemolizumab, respectively.

Change From Baseline in DLQI Total Score

At LTE study baseline, the mean DLQI was 7.6 (standard deviation [SD] = 6.92) across all patients in the LTE study, 6.6 (SD = 6.57) for patients previously treated with nemolizumab, and 9.1 (SD = 7.25) in the group not previously treated with nemolizumab.

At week 20, the mean DLQI score was 4.7 (SD = 5.77) across all patients in the LTE study. Patients previously treated with nemolizumab had a mean DLQI score of 4.6 (SD = 5.84), with a mean change from lead-in baseline and LTE study baseline of −10.4 (SD = 7.52) and −2.0 (SD = 5.54), respectively. Patients not previously treated with nemolizumab had a mean DLQI score of 5.0 (SD = 5.55), with a mean change from lead-in baseline and LTE study baseline of −10.2 (SD = 7.06) and −3.9 (SD = 6.02), respectively.

At week 56, the mean DLQI score was 3.9 (SD = 4.72) across all patients in the LTE study. Patients previously treated with nemolizumab had a mean DLQI score of 4.2 (SD = 4.93), with a mean change from lead-in baseline and LTE study baseline of −11.1 (SD = 7.32) and −3.6 (SD = 6.36), respectively. Patients not previously treated with nemolizumab had a mean DLQI score of 3.7 (SD = 4.58), with a mean change from lead-in baseline and LTE study baseline of −12.2 (SD = 6.92) and −6.4 (SD = 6.71), respectively.

Change From Baseline in CDLQI Total Score

At LTE study baseline, the mean CDLQI score was 6.9 (SD = 6.17) across all patients in the LTE study, 5.3 (SD = 5.83) for patients previously treated with nemolizumab, and 9.1 (SD = 6.00) in the group not previously treated with nemolizumab.

At week 20, the mean CDLQI score was 4.3 (SD = 4.64) across all patients in the LTE study. Patients previously treated with nemolizumab had a mean CDLQI score of 4.1 (SD = 4.62), with a mean change from lead-in baseline and LTE study baseline of −8.3 (SD = 6.15) and −1.9 (SD = 4.41), respectively. Patients not previously treated with nemolizumab had a mean CDLQI score of 4.4 (SD = 4.18), with a mean change from lead-in baseline and LTE study baseline of −8.3 (SD = 6.43) and −5.1 (SD = 5.65), respectively.

At week 56, the mean CDLQI score was 3.4 (SD = 4.48) across all patients in the LTE study. Patients previously treated with nemolizumab had a mean CDLQI score of 2.1 (SD = 2.02), with a mean change from lead-in baseline and LTE study baseline of −9.2 (SD = 6.09) and −4.5 (SD = 5.11), respectively. Patients not previously treated with nemolizumab had a mean CDLQI score of 4.4 (SD = 5.28), with a mean change from lead-in baseline and LTE study baseline of −8.2 (SD = 6.80) and −6.3 (SD = 6.77), respectively.

Harms Results

During the 52-week LTE study treatment period, 63.8% of patients experienced at least 1 AE, most being mild or moderate in severity, and 16.7% experienced a study drug-related AE. A treatment-emergent SAE was experienced by 4.1% of patients. A total of 4.8% of patients experienced a severe AE, with the most common being AD (0.9%), COVID-19 infection (0.3%), increased blood creatine phosphokinase (0.2%), asthma (0.2%), asthenia (0.2%), impetigo (0.2%), and headache (0.2%). A treatment-emergent AESI (by investigator) was experienced by 23.7% of patients, most commonly infections (20.3%) and asthma (newly diagnosed or worsening; 4.3%). Treatment-emergent AEs leading to study drug withdrawal were experienced by 3.3% of patients; 3.1% of patients experienced an AE leading to study discontinuation.

During the LTE study follow-up period, 18.4% of patients experienced at least 1 AE, most being mild or moderate in severity, and 2.7% experienced a study drug-related AE. A treatment-emergent SAE was experienced by 2.7% of patients. A total of 2.0% of patients experienced a severe AE, and the only severe AE occurring in more than 1 patient was AD (0.7%). A treatment-emergent AESI (by investigator) was experienced by 4.8% of patients, most commonly infections (3.1%) and asthma (newly diagnosed or worsening; 1.7%). No treatment-emergent AEs led to study drug withdrawal in the follow-up period; however, 4.4% of patients experienced an AE leading to study discontinuation.

There were no deaths during the treatment or follow-up period of the LTE study.

Critical Appraisal

The ARCADIA LTE study was designed as a single-arm, open-label extension to assess long-term safety and efficacy of nemolizumab in the treatment of adult and adolescent patients with moderate to severe AD. This open-label design could bias the reporting for subjective efficacy outcomes and harms. The direction and magnitude of these potential biases remain unclear. In addition, the absence of statistical hypothesis testing and a control group (i.e., no active comparator or placebo arm) limits the ability to draw definitive conclusions regarding the treatment effect. The study used a dosing regimen of every 4 weeks, which does not align with the dosing schedule of every 8 weeks outlined in the product monograph and may not reflect real-world use.

The extension study partly consisted of patients who took part in the lead-in studies, ARCADIA 1, ARCADIA 2, and ARCADIA CYCLO, and therefore it is reasonable to expect that the same strengths and limitations related to generalizability apply to the extension period. Given that patients needed to complete the lead-in studies before enrolling, the treatment extension population is inherently enriched with patients who met the study definition of clinical responder. Attrition rates were high, particularly for later time points such as week 56, which may limit generalizability and interpretability of long-term outcomes. Although the study follow-up was limited to 56 weeks, the consulted clinical experts indicated that this duration was likely sufficient to assess safety and efficacy. The study is ongoing and is expected to be completed in August 2026.

Indirect Comparisons

One network meta-analysis (NMA) was submitted by the sponsor to fill gaps in the comparative evidence for nemolizumab versus relevant comparators that included abrocitinib, baricitinib, dupilumab, lebrikizumab, tralokinumab, and upadacitinib for the treatment of moderate to severe AD in patients aged 12 years or older who are candidates for systemic therapy. The focus of the NMA was in patients who were experienced with CsA. Because baricitinib, lebrikizumab, and tralokinumab were not relevant to public payers in Canada, they were excluded from this report.

Description of Studies

A feasibility assessment was undertaken to ascertain the extent of clinical heterogeneity across 26 RCTs identified in the systematic literature review. Study design characteristics, baseline characteristics, treatment regimens, outcome definitions, time points, and data availability were considered potential sources of clinical heterogeneity and explored in the feasibility assessment. Of the 26 RCTs, 22 were included in at least 1 network, which included 7 trials of interest (AD Up, ARCADIA 1, ARCADIA 2, ARCADIA CYCLO, JADE DARE, LIBERTY AD CAFÉ, and LIBERTY AD CHRONOS) comparing nemolizumab, dupilumab, abrocitinib, and upadacitinib, connected via placebo comparator arms. The feasibility assessment suggested that the included trials were similar in study design, patient characteristics, treatment regimens, and outcome measures. The authors noted that the main source of heterogeneity was attributed to trial data that included populations experienced with CsA and not experienced with CsA; no other sources of heterogeneity were described. An NMA was considered feasible for patients who were experienced with CsA for the following outcomes of interest at 16 weeks: EASI-75 response, improvement in PP-NRS of 4 points or more (PP-NRS response), IGA score of 0 or 1 (IGA response), DLQI change from baseline, AEs, and discontinuation due to AEs (DAEs). It was assumed that trials with outcome data reported for mixed-age populations, but with a majority (≥ 75%) adult population, could be assessed in adult networks, and that outcome data could be pooled across time points to maintain network integrity. There were no data available for adolescents who were experienced with CsA. Bayesian random-effect NMAs were conducted using generalized linear models to estimate odds ratios and 95% credible intervals (CrIs), except for DLQI change from baseline, which was assessed with mean difference and 95% CrI.

Efficacy and Harm Results

The results for EASI-75 response, PP-NRS response, and IGA response favoured upadacitinib 30 mg versus nemolizumab. The results for IGA response also favoured abrocitinib versus nemolizumab. The results for DLQI mean change from baseline did not favour any treatments. Across all outcomes, there was no difference between dupilumab versus nemolizumab. The results for AEs and DAEs did not favour any treatments.

Critical Appraisal

The protocol of the systematic review and NMA was not a priori registered. The methods used to conduct the systematic literature review used appropriate criteria to search databases, select studies, extract data, and assess risk of bias of the included studies. The NMA included relevant outcomes identified by the CDA-AMC team, which included EASI-75 response, PP-NRS response, IGA response, DLQI change from baseline, AEs, and DAEs. Analysis for all outcomes was performed at 16 weeks and did not include the maintenance treatment phase of 48 weeks for nemolizumab. This was attributed to insufficient data beyond 16 weeks for the comparators of interest, which represents a gap in the available indirect evidence.

The authors noted that the included trials were considered similar in study design, patient eligibility criteria, baseline patient characteristics, and outcome characteristics (i.e., definitions and methods of reporting outcomes). However, there were notable baseline differences with respect to mean body surface area (BSA) (ranging from 42.5% to 59.5% across trials and arms), and proportion of patients with baseline IGA scores of 3 or 4 (IGA = 3 ranging from 46% to 73.9%; IGA = 4 ranging from 26.1% to 54%). The included studies enrolled a combination of patients with inadequate response to, or intolerance to CsA. Efforts were made in the NMA to reduce heterogeneity by stratifying the adult and adolescent patient populations experienced with CsA into subgroups. The results across outcomes showed low heterogeneity via the I² statistic, which ranged from ██ ██ ██ (indicating low heterogeneity). The results of sensitivity analyses that removed trials with mixed patient populations (with and without experience with CsA), were consistent with the base-case analysis. There were no data available for adolescents who were experienced with CsA, which represents a gap in the available indirect evidence.

Overall, the network was sparse (i.e., many comparisons but few studies). Because most nodes were informed by only 1 trial and had small sample sizes, comparisons were underpowered, which contributed to wide CrIs in the analyses. Most 95% CrIs included both better and worse performance, except for EASI-75, PP-NRS, and IGA response, where upadacitinib 30 mg was favoured over nemolizumab 30 mg (i.e., 95% CrIs did not cross the null). Due to the absence of closed loops in the network, it was not possible to assess for inconsistency across direct and indirect evidence in the NMA.

Conclusions

Results from 2 identical phase III, double-blind, multicentre, placebo-controlled RCTs (ARCADIA 1 and ARCADIA 2) in patients with moderate to severe AD demonstrated that nemolizumab led to statistically significant improvements in markers of disease activity (measured by IGA success and EASI-75) and itch (measured by PP-NRS) at 16 weeks, when compared with placebo. Among the subset of patients who had demonstrated a clinical response by 16 weeks, results suggested that improvements in disease activity and itch were sustained, regardless of whether nemolizumab was given every 4 weeks or every 8 weeks. The clinical experts and patient groups highlighted the debilitating aspect of itch in this population. Results from subgroup analyses at 16 weeks suggested a more pronounced effect on disease activity in patients with severe itch (PP-NRS ≥ 7), although these results were only considered supportive of the overall effect of nemolizumab.

Results from a similar phase IIIb, double-blind, placebo-controlled RCT (ARCADIA CYCLO) in patients refractory to CsA or where CsA was inadvisable, suggested similar trends in EASI-75 and PP-NRS, but not IGA, and the results were more subject to uncertainty due to imprecision. Results from all 3 trials were subject to some uncertainty due to limitations primarily impacting external validity, as patients with certain comorbidities such as uncontrolled or worsening asthma and certain treatment histories were excluded from the ARCADIA 1 and ARCADIA 2 studies; and the ARCADIA CYCLO study enrolled patients with more mixed patient histories. HRQoL results demonstrated a reduction in scores (nominal P value < 0.05 for DLQI) across all trials, although the results were not considered clinically meaningful and did not attain the threshold for a clinically meaningful difference provided by the clinical experts consulted for this review.

In the ARCADIA LTE study, patients previously treated with nemolizumab showed gradual initial improvement in disease activity and symptoms, which appeared to be maintained through the treatment period. Patients who were not previously treated with nemolizumab demonstrated continued improvement throughout the LTE study. However, limitations such as the enriched population and lack of comparator preclude definitive conclusions.

The clinical experts highlighted the safety profile of nemolizumab was considered tolerable and manageable. The long-term safety profile of nemolizumab in the ARCADIA LTE study was consistent with that of the ARCADIA 1 and ARCADIA 2 studies, and no new safety signals were observed.

There was no direct evidence comparing nemolizumab to other systemic therapies used to treat moderate to severe AD in Canada. The results of the sponsor-submitted indirect treatment comparison suggested that upadacitinib 30 mg was favoured over nemolizumab for all efficacy outcomes, and abrocitinib was favoured for IGA response. The results comparing nemolizumab and dupilumab were inconclusive for all outcomes of interest due to the wide 95% CrIs that contained the possibility that either treatment could be favoured or that there could be no difference. The indirect treatment comparison safety results for overall occurrence of any AE and DAE at week 16 did not support a difference between nemolizumab and any relevant comparators. However, the NMA could not support comparisons of specific AEs, which may be of interest to patients and clinicians. Overall, the NMA estimates were subject to uncertainty due to serious imprecision, and a lack of data for adolescents and treatment effect beyond 16 weeks.

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 nemolizumab 30 mg/0.49 mL SC injection in the treatment of moderate to severe AD in patients aged 12 years or older who are candidates for systemic therapy.

Disease Background

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

AD is a chronic, relapsing, inflammatory, and noncontagious skin disease that is thought to involve a complex interplay of genetic, immune, and neuroimmune dysregulation, skin barrier dysfunction, microbial imbalance, and environmental factors.^1-4 AD is a debilitating disease that can negatively impact patients’ quality of life, and has the highest disease burden among skin diseases.¹⁴ Patients can experience a range of symptoms; however, the disease is characterized by severe and persistent itch with a chronic duration and a fluctuating course of the eczematous lesions.² Acute worsening of AD, commonly referred to as a flare, presents as dry, red, itchy skin that can lead to lesions that blister, ooze, and crust. While clinical presentation can vary depending on age, disease stage, geographic location, and race or ethnicity, itch is the leading symptom that characterizes AD.^2,4 As the most common and most burdensome symptom of AD, itch has a direct impact on patients’ daily activities and quality of life.^15,16 Chronic itch can result in sleep disruption and poor sleep quality, observed in approximately two-thirds of patients with AD.^17-21 Sleep disruption can considerably reduce daytime attention levels, in turn negatively impacting school and work performance and relationships. AD is also linked to mental and psychological distress, depression, and a higher risk of suicidal ideation, with itch being a major risk factor and predictor for these psychological comorbidities.²² Patients with AD are also more likely to have or develop other comorbidities, such as allergies or infections.^4,23

The prevalence and incidence of AD varies widely between different geographic regions due to genetic, behavioural, cultural, socioeconomic, and climate-related factors.^24-26 It is estimated that the prevalence of eczema in Canada is 15% in children and adolescents (aged 0.5 to 17 years) and 3.5% in adults.^5,6

The diagnosis of AD and assessment of disease severity is based on clinical judgment and patient-reported history and symptoms.^2,7,15,27-30 Diagnostic criteria include the Hanifin-Rajka criteria (with itch and eczema as the essential features), or the UK Working Party diagnostic criteria.^3,17 While the Hanifin-Rajka criteria are the gold standard, the UK Working Party criteria are considered easier to employ in clinical practice, although both have been validated and tested in various studies and populations.^2,3,17,31 To receive an AD diagnosis, other conditions must be excluded, particularly psoriasis, scabies, contact dermatitis, and seborrheic dermatitis.^3,17 Diagnosis of AD is generally more straightforward in infants and young children, as it can be more complicated in more severe cases and adults due to heterogeneous presentation.^3,17 Depending on the extent and the severity of symptoms, AD can be classified as mild, moderate, or severe.⁷ Patients are considered to have moderate to severe disease when they have at least 1 of the following: BSA involvement of 10% or more, moderate to severe lesions, involvement of highly visible areas or those important for function (e.g., neck, face, genitals, palm, and/or soles), or substantially impaired quality of life.⁷ AD severity can be classified using EASI and intensity of patient burden can be quantified using patient-reported outcomes.⁹ No biomarkers have been identified for the diagnosis of AD.^{3,7,15,17,27,29}

Standards of Therapy

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

The management of AD involves a multipronged approach including patient education, skin hydration, restoration of the skin, elimination of exacerbating factors, and pharmacologic treatment of skin inflammation.⁸ The clinical experts consulted for this review noted that the goals of treatment are reduction of signs and symptoms (e.g., red scaly plaques, itch, scratching, lichenification, dyspigmentation, scarring), prevention of secondary infections, improvement of quality of life (e.g., sleep, daily activities, hobbies), maintenance of independence, reduction of caregiver burdens, and enhancement of work and or school productivity.

The typical treatment paradigm for AD is to initiate treatment with a topical drug or combination of drugs, such as a TCS and/or TCI, ruxolitinib, crisaborole, or roflumilast for at least 4 weeks, pending referral to a dermatologist. If this treatment is unsuccessful, patients may be trialled on another topical drug for 4 to 8 weeks. Topical drugs may be used concurrently with narrow band phototherapy (administered in office 2 to 3 times per week) or systemic drugs for the maintenance of response, rescue treatment, or control of flares.³² The clinical experts noted that phototherapy can be limited by factors such as accessibility and convenience.

Patients with persistent, moderate to severe AD despite optimized topical therapy require systemic treatment to achieve disease control, with re-evaluation to rule out concurrent diseases and confirm the diagnosis of AD.⁸ Dupilumab, tralokinumab, abrocitinib, and upadacitinib are approved in Canada and have recommendations for use in moderate to severe AD. The American Academy of Allergy, Asthma and Immunology and American College of Allergy, Asthma and Immunology Joint Task Force recommends the addition of dupilumab or tralokinumab over continuing standard topical treatment, then switching to a JAK inhibitor if the patient’s disease is not responsive.³³ Guidelines on the conventional immunosuppressing systemic therapies is more mixed: some guidelines conditionally suggest the use of methotrexate, mycophenolate mofetil, and CsA with appropriate monitoring,³² while others suggest only CsA (recommending against methotrexate, azathioprine, or mycophenolate mofetil). Systemic corticosteroids are only recommended for acute, severe exacerbations and as a short-term bridge to other therapies.³²

The clinical experts noted that a typical progression would be for patients to trial conventional systemic therapies for a period of 3 months and if well controlled, they would remain on that therapy, but if they do not respond, then they may switch to another conventional systemic therapy or the advanced systemic therapies (dupilumab, tralokinumab, abrocitinib, and upadacitinib). They noted a trial of 3 to 6 months is usually used to determine response. If the patient’s skin is not completely clear, they may add a topical therapy. Occasionally, the experts noted that a combination of systemic therapies may be used.

Dupilumab, abrocitinib, and upadacitinib are recommended by CDA-AMC for reimbursement for moderate to severe AD. Lebrikizumab and tralokinumab are approved for use in Canada, but are not recommended to be reimbursed for moderate to severe AD.

Drug Under Review

Key characteristics of nemolizumab and other treatments available for moderate to severe AD are summarized in Table 4.

Nemolizumab is a humanized monoclonal antibody of the IgG2 subclass that inhibits IL-31 signalling by selectively binding to IL-31RA to ameliorate pruritus, inhibit inflammatory responses, and restore barrier integrity in AD.³⁴

Nemolizumab is indicated for the treatment of patients aged 12 years and older with moderate to severe AD whose disease is not adequately controlled with topical prescription therapies or when those therapies are not advisable. The reimbursement request is for the treatment of moderate to severe AD in patients aged 12 years and older whose disease is not adequately controlled with topical prescription therapies or when those therapies are not advisable, and who are refractory to or ineligible for systemic immunosuppressant therapies. Nemolizumab is currently approved for the treatment of AD in the European Union and US.^35,36

Nemolizumab is administered as an SC injection and can be used with or without TCSs and TCIs.³⁴ The Health Canada–recommended nemolizumab dose for the treatment of AD is an initial dose of 60 mg (two 30 mg injections), following by 30 mg every 4 weeks.³⁴ After 16 weeks of treatment, the recommended dose for patients who achieve a clinical response is 30 mg every 8 weeks.³⁴

Table 4: Key Characteristics of Treatments for AD

AD = atopic dermatitis; ATP = adenosine triphosphate; GI = gastrointestinal; IL = interleukin; MACE = major adverse cardiovascular events; NA = not applicable; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; RA = receptor antagonist; SC = subcutaneous; STAT = signal transducer and activator of transcription.

^aHealth Canada–approved indication.

Sources: Nemolizumab product monograph,³⁴ abrocitinib product monograph,³⁷ dupilumab product monograph,³⁸ and upadacitinib product monograph.³⁹

Perspectives of Patients, Clinicians, and Drug Programs

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

Patient Group Input

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

One patient group submitted input for this review, the ESC, a registered charity in Canada aiming to improve the lives of people living in Canada with eczema through support, education, awareness, and research. Information was gathered for this submission through questionnaires, patient interviews, review of published literature, and data from multiple surveys with a combined total of more than 3,000 people living in Canada.

When asked about their disease experience, patients with AD highlighted the profound physical and emotional challenges associated with the condition. Physical symptoms include painful and itchy skin lesions that crack, ooze, and bleed; visible skin flares; and persistent itching that disrupts sleep, school and work, and quality of life. Patients reported experiencing skin damage, bleeding, scarring, and pigment changes due to rashes and scratching, with open wounds posing an increased risk of skin infection. Some patients reported living with chronic pain and discomfort due to their AD, never experiencing relief from their symptoms. In addition to physical symptoms, patients also face a substantial emotional toll of AD, including feelings of isolation, hopelessness, frustration, and despair over limited treatment success, as well as anxiety, depression, and low self-esteem. Some patients also have trouble maintaining social and intimate relationships, due to the impact on confidence and comfort. The patient group noted that, until recently, systemic treatments for AD have been limited, including phototherapy, off-label immune-suppressing medication (e.g., CsA, methotrexate), and oral corticosteroids for treatment of flares. However, many patients fear side effects associated with long-term use of treatments such as steroids. Patients have had substantial recent success with new systemic therapies becoming available; however, many are not accessible to all patients, and some patients do not respond. Patients report feeling frustrated over the trial-and-error process of cycling through currently available treatments, and many have exhausted all treatment options. The patient group input noted that because AD is a heterogenous disease, no single treatment option can meet the needs of all patients.

ESC highlighted the urgency for innovative therapies that can offer long-term relief and improve quality of life. Outcomes of treatment identified as most important to patients with AD include improved disease control, symptom and pain relief (e.g., itch, dryness, inflammation, blistering), convenience, and improved quality of life.

ESC described general patient experiences with nemolizumab, although the source of these experiences (e.g., patient interviews, survey responses, general trial observations) was not specified. Patients treated with nemolizumab reported substantial reductions in itch, with noticeable improvements in skin clearance from rashes, lesions, sores, and open wounds, which provided further benefit in improving sleep, productivity, self-esteem, and mental health. ESC noted that many patients who were previously unsuccessful with approved and off-label medications reported success with nemolizumab.

Clinician Input

Input From Clinical Experts Consulted for This Review

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

Unmet Needs

The clinical experts noted that pruritus (itch) is a very debilitating and unpredictable aspect of AD, and better treatments are needed to target this symptom. In addition, approximately 30% to 40% of patients will either not respond to, or will become refractory to, currently available nonbiologic treatments (1 expert noted this would be approximately 20% if biologics were also considered); therefore, new treatments are needed. Moderate to severe AD often comes with a high burden of disease, which also impacts patients’ HRQoL and their ability to carry out activities due to symptoms, or due to a feeling of self-consciousness. The currently available treatments (e.g., JAK inhibitors) are not always well tolerated and have contraindications or black box warnings, further reinforcing the need for new treatment options.

Place in Therapy

The clinical experts noted that nemolizumab is the only treatment with a mechanism of action targeting IL-31, which is involved in the sensation of pruritus. The experts stated that it could be used as monotherapy or in combination with topical therapies, phototherapy, and other conventional immunosuppressing systemic drugs. They noted that nemolizumab should be considered a first-line systemic therapy after topical therapies, and considered alongside the other biologics such as dupilumab, tralokinumab, or lebrikizumab. The clinical experts noted that the reimbursement criteria for dupilumab could also be considered appropriate for nemolizumab (e.g., having been unsuccessful with 1 conventional systemic therapy trial); however, the choice to use a conventional treatment or advanced systemic drug would be partially determined by the drug coverage.

Patient Population

The experts noted that nemolizumab would be suitable for all patients with moderate to severe AD, and 1 expert noted it would also be appropriate for patients who do not tolerate, respond to, or cannot access phototherapy, as well as those who have a contraindication to alternate systemic therapies. The experts further noted that patients who are substantially impacted by pruritus, who have substantial flares of AD, whose current therapy has stopped working, or whose AD has a substantial impact on their lives would be those most in need of additional interventions and most likely to benefit.

There are no specific criteria or laboratory tests required for diagnosis, as AD is typically diagnosed clinically and is a common disease. The experts noted that there are no biomarkers or tests that would indicate a patient’s likelihood to respond to nemolizumab.

Assessing the Response to Treatment

The clinical experts noted that the measures used in the trial are also used in clinical practice to assess response, including patient global assessment of improvement, EASI reduction, numeric rating scale for itch, and DLQI. The clinical experts also noted that other subjective changes reported by patients in clinical practice may include changes to sleep, daily activities, work productivity, improvement in mental health, and reduction of scratching or itching.

The experts commented that the measures used in clinical practice vary and may be chosen according to the reimbursement criteria set by payers to obtain coverage for a treatment (e.g., change in EASI score at 6 months). Patients are often assessed every 3 months initially and when their condition has improved and stabilized, they are typically assessed every 6 months.

Discontinuing Treatment

The clinical experts stated that treatment should be discontinued if there has been no meaningful response after 6 months of therapy, if the patient experiences an allergy or hypersensitivity, or if they experience intolerable AEs. The definition of a meaningful response could vary and the decision to discontinue is often shared between the clinician and the patient. One expert shared an example that patients with minimal disease response according to a tool such as EASI may still wish to continue treatment if their pruritus is controlled. If there is a partial response, the experts noted that the addition of a new therapy should be considered first, rather than discontinuing the current treatment.

Prescribing Considerations

According to the experts, patients should receive a proper diagnosis of AD by a dermatologist before nemolizumab is prescribed. Patients can be managed and monitored by dermatologists in any outpatient clinic (community or hospital). After 6 months of treatment and confirmation of clinical efficacy, primary care providers could also manage the prescribing until reinvolvement of the dermatologist is necessary.

Clinician Group Input

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

One joint input was submitted for this review by DAO and the Atlantic Dermatology Group. DAO provides broad representation for more than one-half of registered dermatologists in Canada. The Atlantic Dermatology Group is a group of hospital-based and academic-based dermatologists practising in the Atlantic provinces. Information for this submission was gathered through a review of published literature, clinical trial data, and experience with nemolizumab as clinical trialists.

The input from clinician groups was generally consistent with the clinical experts consulted for this review, emphasizing the ongoing unmet need for treatments that provide rapid and sustained relief from pruritus, as well as improvements in sleep and quality of life for patients with moderate to severe AD. Like the experts, the clinician groups highlighted the limitations of current therapies, including slow onset of action with biologics and safety concerns or monitoring requirements with JAK inhibitors.

Aligning with expert input, the clinician groups considered nemolizumab a suitable first-line systemic option, to be positioned alongside existing biologics, particularly for patients with an inadequate response to topicals or systemic therapies, or who require more rapid symptom relief. Both also highlighted the novel mechanism and monthly dosing schedule as factors that could improve adherence and convenience. The clinician groups provided additional details on patient selection, noting that nemolizumab may be particularly appropriate for patients with contraindications to JAK inhibitors (e.g., cardiovascular disease, immunosuppression concerns, thromboembolism). They also noted that patients with mild AD that can be managed effectively by topicals alone are less appropriate candidates.

Consistent with expert input, the clinician groups noted a range of clinical and patient-reported tools (e.g., IGA, EASI, PP-NRS, sleep scores) that could be used to assess treatment response, and support reassessment intervals from 4 to 12 months. However, they provided more specific discontinuation criteria, such as not achieving a reduction in PP-NRS or sleep numeric rating scale of 4 points or more or less than 50% improvement in EASI by week 16, while the experts emphasized a flexible, shared decision-making approach. Both clinical experts and clinician groups agreed that initiation should be done by dermatologists in an outpatient setting, but the clinician groups added that allergists could also initiate therapy, and that self-injection could be possible after training. The clinician group noted that because misdiagnosis of AD can occur, particularly in adult-onset cases, clinical expertise in differentiating AD from other eczematous conditions (e.g., contact dermatitis, seborrheic dermatitis, cutaneous T-cell lymphoma) is essential to ensure appropriate patient selection.

Drug Program Input

The drug programs provide input on each drug being reviewed through the reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by for this review are summarized in Table 5.

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

AD = atopic dermatitis; CDEC = Canadian Drug Expert Committee; EASI = Eczema Area and Severity Index; EASI-75 = ≥ 75% improvement in the Eczema Area and Severity Index from baseline; IGA = Investigator Global Assessment; PP-NRS = Peak Pruritus Numeric Rating Scale; vs. = versus.

Clinical Evidence

The objective of this Clinical Review report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of nemolizumab 30 mg/0.49 mL SC injection in the treatment of moderate to severe AD in patients aged 12 years and older whose disease is not adequately controlled with topical prescription therapies or when those therapies are not advisable, or who are refractory to or ineligible for systemic immunosuppressant therapies. The focus will be placed on comparing nemolizumab to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of nemolizumab is presented in 3 sections with our 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. Our assessment of the certainty of the evidence in this first section using the GRADE approach follows the critical appraisal of the evidence. The second section includes sponsor-submitted LTE studies. The third section includes indirect evidence from the sponsor. No studies addressing gaps were included in the review.

Included Studies

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

• 3 pivotal studies or RCTs identified in systematic review

• 1 LTE study

• 1 indirect treatment comparison consisting of NMAs.

Systematic Review

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

Description of Studies

Three pivotal studies were included in this review (ARCADIA 1, ARCADIA 2, and ARCADIA CYCLO). Characteristics of the included studies are summarized in Table 6.

ARCADIA 1 and ARCADIA 2 Studies

The study designs for the ARCADIA 1 and ARCADIA 2 studies were identical; both were phase III, double-blind, placebo-controlled, multicentre RCTs evaluating the efficacy and safety of nemolizumab in adults and adolescents (aged ≥ 12 years) with moderate to severe AD. The coprimary end point of the ARCADIA 1 and ARCADIA 2 studies was the proportion of patients with IGA success, and the proportion of patients with EASI-75 at week 16. Key secondary end points also evaluated changes in PP-NRS across the study period. Patients were randomized via interactive response technology 2:1 to receive nemolizumab (ARCADIA 1: n = 620; ARCADIA 2: n = 522) or placebo (ARCADIA 1: n = 321; ARCADIA 2: n = 265). Randomization was stratified by disease severity (IGA = 3 or 4), and pruritus severity (PP-NRS < 7 or ≥ 7).

Before randomization for the initial treatment period, there was a 2-week run-in period where patients were started on standardized background topical therapy; background topical therapy included the use of authorized TCS with or without TCI. A medium-potency TCS was to be used on the body, where considered safe (e.g., trunk and extremities). A low-potency TCS or a TCI was to be used in sensitive areas considered unsafe for medium-potency TCSs (e.g., face, neck, intertriginous areas) or where medium-potency TCS was not tolerated. The prescribed background therapy use was within daily limits according to the product labelling. The protocol included guidelines to adjust background therapy use according to disease activity and tolerability.

After the 16-week initial treatment period, patients in the treatment arm who met the study definition of clinical responders (defined as an IGA of 0 [clear] or 1 [almost clear] or EASI-75 at week 16) were rerandomized 1:1:1 via interactive response technology to either continue receiving nemolizumab 30 mg every 4 weeks (n = 169 in both studies), to receive nemolizumab every 8 weeks (n = 169 in both studies), or to receive placebo (n = 169 in both studies). The patients who had been randomized to placebo who met the study definition of clinical responders continued to receive placebo every 4 weeks in the maintenance period (n = 185 in both studies).

Patients who required rescue therapy for clinical worsening of AD before week 16 were considered to be unsuccessful with treatment and may be considered for the LTE study but were required to continue with study visits until the week 16 visit; patients who completed the maintenance period were also eligible to participate in the LTE study. Patients who declined or were not eligible to enter into the LTE study were to complete a follow-up visit 12 weeks after their last study drug injection. The follow-up visit was not required for patients who participated in the LTE study. Patients who discontinued the initial or maintenance periods prematurely were to complete an early termination visit and a follow-up visit 12 weeks after the last study drug injection. Full details of the study design for the ARCADIA 1 and ARCADIA 2 studies are in Figure 1.

ARCADIA CYCLO Study

The ARCADIA CYCLO study was a phase IIIb, double-blind, placebo-controlled, multicentre RCT which included a 16-week treatment period and an 8-week follow-up period for patients who declined or were not eligible to enter the LTE study. There was no run-in period with topical therapy; patients were randomized and began treatment right after the screening period. Patients were randomized 1:1 to receive nemolizumab (n = 138) or placebo (n = 138). Randomization was stratified based on baseline IGA severity (IGA = 3, IGA = 4), and prior CsA exposure (yes, no). The primary end point of the ARCADIA CYCLO study was the proportion of patients with EASI-75, and proportion of patients with an improvement of PP-NRS score of 4 or more from baseline at week 16.

Patients who completed the 16-week treatment period may have been eligible to enrol into the LTE study; the follow-up period, including the follow-up visit, was not required for patients who participated in the LTE study. Patients who declined or were not eligible to enter the LTE study completed a follow-up visit at week 24 (12 weeks after their last study drug injection). Patients who prematurely discontinued the study completed an early termination visit at the time of discontinuation and a follow-up visit 12 weeks after the last study drug injection. Full details of the study design for the ARCADIA CYCLO study are in Figure 2.

Table 6: Details of Studies Included in the Systematic Review

AAD = American Academy of Dermatology; ACT = Asthma Control Test; AD = atopic dermatitis; AP-NRS = Average Pruritus Numeric Rating Scale; CDLQI = Children’s Dermatology Life Quality Index; COPD = chronic obstructive pulmonary disease; CsA = cyclosporine; DB = double blind; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; EASI-50 = ≥ 50% improvement in Eczema Area and Severity Index from baseline; EASI-75 = ≥ 75% improvement in Eczema Area and Severity Index from baseline; EASI-90 = ≥ 90% improvement in Eczema Area and Severity Index from baseline; EUDRACT = European Union Drug Regulating Authorities Clinical Trials Database; HADS = Hospital Anxiety and Depression Scale; HBcAb = hepatitis B core antibody; HBsAg = hepatitis B surface antigen; IGA = Investigator Global Assessment; MC = multicentre; NCT = National Clinical Trial; PBO = placebo; PDE-4 = phosphodiesterase 4; PEF = peak expiratory flow; PGADS = Patient Global Assessment of Disease Status; PGAT = Patient Global Assessment of Treatment effect; POEM = Patient-Oriented Eczema Measure; PP-NRS = Peak Pruritus Numeric Rating Scale; PROMIS = Patient-Reported Outcomes Measurement Information System; RCT = randomized control trial; SCORAD = SCORing Atopic Dermatitis; SD-NRS = Sleep Disturbance Numeric Rating Scale; TCI = topical calcineurin inhibitor; TCS = topical corticosteroid; ULN = upper limit of normal; WASO = wakefulness after sleep onset; WPAI:AD = Work Productivity and Activity Impairment Questionnaire: Atopic Dermatitis.

Note: The language used in the table is as per the sponsor and does not reflect language guidelines used at Canada’s Drug Agency.

^aInadequate response was defined as 1 of: failure to achieve or maintain remission or low disease activity (equivalent to IGA ≤ 2) despite treatment with a regimen of a medium-potency, high-potency, or very high-potency TCS (with or without TCI), applied for at least 4 weeks or for the maximum duration per prescribing information; or requirement of a long-term treatment (> 4 weeks) with a high-potency or very high-potency TCS to achieve or maintain remission or low disease activity (equivalent to IGA ≤ 2); or if documentation of inadequate response to topical treatments was not available — patients with a documented recent course of systemic treatment or phototherapy for AD (within 6 months before the visit) were also considered to have had an inadequate response to topical treatments.

^bSilverberg JI, Wollenberg A, Reich A, et al. Nemolizumab with concomitant topical therapy in adolescents and adults with moderate-to-severe atopic dermatitis (ARCADIA 1 and ARCADIA 2): results from two replicate, double-blind, randomised controlled phase 3 trials. The Lancet. 2024.

Sources: Details included in the table are from the sponsor’s summary of clinical evidence¹⁰ and the ARCADIA 1 Clinical Study Report,¹¹ ARCADIA 2 Clinical Study Report,¹² and the ARCADIA CYCLO Clinical Study Report.¹³

Figure 1: ARCADIA 1 and ARCADIA 2 Study Design

BL = baseline; q4wk = every 4 weeks; q8wk = every 8 weeks; W = week.

Source: Details in the figure are from the sponsor’s ARCADIA 1 Clinical Study Report.¹¹

Figure 2: ARCADIA CYCLO Study Design

BL = baseline; W = week.

* Applicable for patients who did not participate in the long-term extension study.

Source: Details included in the figure are from the ARCADIA CYCLO Clinical Study Report.¹³

Populations

Inclusion and Exclusion Criteria: ARCADIA 1 AND ARCADIA 2 Studies

The ARCADIA 1 and ARCADIA 2 studies had the same patient inclusion and exclusion criteria (Table 6). Briefly, patients were required to be aged 12 years or older, meet specific thresholds of IGA (≥ 3), EASI (≥ 16), BSA (≥ 10%), and PP-NRS (≥ 4), and have a minimum disease duration of 2 years. In addition, patients had to have a documented history of inadequate response to topical medications (TCS with or without TCI) within 6 months of study enrolment. Notable exclusion criteria included patients with COPD or chronic bronchitis, uncontrolled asthma, specific medications in the weeks before enrolment, or who either did not improve or worsened after a 16-week trial with dupilumab.

Inclusion and Exclusion Criteria: ARCADIA CYCLO Study

Inclusion and exclusion criteria for the ARCADIA CYCLO study are described in Table 6. Notable inclusion criteria for the ARCADIA CYCLO study were that patients had to be aged 18 years or older, meet specific thresholds of IGA (≥ 3), EASI (≥ 20), BSA (≥ 10%), and PP-NRS (≥ 4), and have a minimum disease duration of 2 years. Of note, patients also had to have a documented history of inadequate response (including at maximum dose per the prescribing information), intolerance or unacceptable toxicity to CsA, or have CsA be medically inadvisable for reasons prespecified in the protocol. Of note, the minimum EASI score required was 20. In addition, patients had to have a documented history of inadequate response to topical medications (TCS with or without TCI) within 6 months of study enrolment. Notable exclusion criteria included patients with COPD or chronic bronchitis or uncontrolled asthma.

Interventions

Interventions and Comparators: ARCADIA 1 and ARCADIA 2 Studies

The interventions were as follows.

• Initial treatment phase (week 0 to week 16)

  • Arm A: 2 injections (2 × nemolizumab 30 mg) SC at baseline, then nemolizumab 30 mg SC every 4 weeks plus topical background therapy

  • Arm B: 2 injections (2 × placebo) SC at baseline, then placebo SC every 4 weeks plus topical background therapy

• Maintenance treatment phase (week 16 to week 48)

  • Arm A (only patients in the initial phase who met the study definition of clinical responder): nemolizumab 30 mg SC every 4 weeks plus topical background therapy

  • Arm B (only patients in the initial phase who met the study definition of clinical responder): nemolizumab 30 mg SC every 8 weeks plus topical background therapy

  • Arm C (only patients in the initial phase who met the study definition of clinical responder): placebo SC every 4 weeks plus topical background therapy

  • Arm D (placebo arm from the initial phase): continued placebo SC every 4 weeks plus topical background therapy

Dose modification of nemolizumab was not permitted in either study. During the maintenance period, patients who received nemolizumab 30 mg every 8 weeks also had a placebo at weeks 20, 28, 36, and 44 to maintain the blind.

Study drug was prepared by a pharmacist or other qualified study personnel. Members of the study site staff, including those responsible for preparation, did not have access to the randomized treatment assignment. As there may have been detectable differences between active treatment and placebo during the reconstitution process, the treatment was to be delivered for injection after the reconstitution was complete. The pharmacist (or other qualified personnel) preparing study drug was not to be involved with any study assessments and should not have discussed any aspects of study drug reconstitution with the patient or caregiver or study staff involved in patient interviews or study assessments. Injections occurred at the study site and were administered by the investigator or other qualified personnel into the patient’s abdomen or alternative injection location. A different injection location was to be selected for each injection and recorded in the treatment record. For patients willing and able to self-inject or to have a caregiver inject the study drug, study site staff provided training and the patients or caregivers were allowed to inject the medication on subsequent visits under the supervision of the site staff. Patients remained at the study site for at least 30 minutes after the first 2 injections during the study.

Dose modification was not permitted. The dosing frequency was scheduled based on the baseline or day 1 visit date. If a study visit occurred outside the visit window, the study drug could have still been administered, provided there was a minimum 3-week interval between 2 injections. Future visits were to be scheduled within the required windows based on the baseline or day 1 visit date, while maintaining the minimum 3-week interval between 2 injections.

Interventions and Comparators: ARCADIA CYCLO Study

The interventions were as follows.

• Arm A: 2 injections (2 × nemolizumab 30 mg) SC at baseline, then nemolizumab 30 mg SC every 4 weeks, plus topical background therapy

• Arm B: 2 injections (2 × placebo) SC at baseline, then placebo SC every 4 weeks, plus topical background therapy

Concomitant Therapies: All ARCADIA Trials

Unless specified as prohibited therapies, all therapies were to be authorized, including basic skin care (cleansing and bathing), moisturizers, bleach baths, topical anesthetics, and antihistamines without a sedative effect.

Patients applied a moisturizer at least once daily, and liberally as needed, to dry skin and AD lesions beginning at screening, and throughout the study. The patient’s current moisturizer or a moisturizer recommended by the investigator could be used. Use should not have occurred within 8 hours before each clinic or office visit. Whenever possible, patients were to use the same moisturizer throughout the study.

Patients applied an authorized background topical therapy to all AD lesions beginning within the screening period and at least 14 days before day 1 (i.e., run-in period). A thin layer of treatment was applied at a frequency that was necessary to ensure disease stability and prevent AD flare but that did not exceed the daily frequency recommended in the product labelling — “as needed” treatments were not permitted. Patients applied a medium-potency TCS to areas of the body where use of medium-potency TCS was considered safe (e.g., trunk and extremities). A low-potency TCS or a TCI was to be used in sensitive areas considered unsafe for medium-potency TCS (e.g., face, neck, intertriginous areas) or in cases where medium-potency TCS was not tolerated. The investigator could select either low-potency TCS or TCI for each patient. The patient was allowed to apply only 1 medication to each affected area; concomitant use of low-potency TCS and TCI on the same lesion was not permitted. The investigator adjusted background therapy use during the study according to the disease activity and tolerability, including tapering when signs and symptoms improved, discontinuation when lesions cleared, and restarting if signs and symptoms recurred.

For the ARCADIA 1 and ARCADIA 2 studies only, in transitioning to the maintenance period, patients continued the same background topical therapy used in the initial treatment period leading up to the week 16 visit, including tapering or complete cessation, if applicable. Throughout the maintenance period, adjustments to background topical therapy, as determined by the investigator, were permitted based on clinical response.

Rescue Therapies: All ARCADIA Trials

Rescue therapies included:

• higher potency TCS (equivalent to class I to II according to the US classification)

• systemic corticosteroids

• biologics (including their biosimilars)

• systemic nonsteroidal immunosuppressants or immunomodulators

• phototherapy.

Whenever possible, investigators were to use topical medication as rescue therapy before escalating to other systemic therapies. If patients received topical treatments or phototherapy as rescue therapy, study drug administration was to continue unless there was a safety concern according to the investigator's judgment. If patients received systemic rescue therapy, study drug administration must have been permanently discontinued. For efficacy analysis, patients receiving any rescue therapies were considered unsuccessful with treatment.

Outcomes

A list of efficacy end points assessed in this Clinical Review report is provided in Table 7 for the ARCADIA 1 and ARCADIA 2 studies, and Table 8 for the ARCADIA CYCLO study. Summarized end points are based on outcomes included in the sponsor’s summary of clinical evidence as well as any outcomes identified as important to this review according to the clinical experts consulted for this review and input from patient and clinician groups and public drug plans. Using the same considerations, we selected end points that were considered to be most relevant to inform expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. The following efficacy outcomes at week 16 (from the initial period) were assessed using GRADE:

• proportion of patients with IGA success

• proportion of patients with EASI-75

• proportion of patients with PP-NRS improvement of 4 or more points

• proportion of patients with PP-NRS of less than 2

• change from baseline in DLQI or CDLQI.

Select notable harms outcomes considered important for informing expert committee deliberations (injection-related reactions, newly diagnosed or worsening asthma, infections, peripheral edema, elevated AST or ALT plus elevated bilirubin) were also assessed using GRADE. To provide additional context on efficacy, the change from baseline in EASI, change from baseline in PP-NRS, and incidence of topical rescue therapy use was also included in the report but not in the GRADE assessment. The corresponding efficacy outcomes in the maintenance period (from week 16 to week 48) were also included in the report but not in GRADE.

Though not considered to be a critical outcome for this review, the results for the Patient-Oriented Eczema Measure were included at the request of the sponsor to provide additional context on eczema-specific symptoms and quality of life improvements. Results for the Patient-Oriented Eczema Measure can be found in Table 38, Table 39, and Table 40 of Appendix 1.

Table 7: Outcomes Summarized From the ARCADIA 1 and ARCADIA 2 Studies

CDLQI = Children’s Dermatology Life Quality Index; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; EASI-75 = ≥ 75% improvement in Eczema Area and Severity Index from baseline; IGA = Investigator Global Assessment; PP-NRS = Peak Pruritus Numeric Rating Scale.

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

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁰

Table 8: Outcomes Summarized From the ARCADIA CYCLO Study

CsA = cyclosporine; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; EASI-75 = ≥ 75% improvement in Eczema Area and Severity Index from baseline; IGA = Investigator Global Assessment; PP-NRS = Peak Pruritus Numeric Rating Scale.

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

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁰

Outcome definitions and details of the measurement properties of the efficacy outcomes are presented in Table 9.

Disease Severity Outcomes

Investigator Global Assessment

IGA is a 5-point scale used by an investigator to evaluate the global severity of AD; scores are associated with specific clinical definitions and range from 0 (clear; defined as minor, residual hypopigmentation or hyperpigmentation, no erythema or induration or papulation, no oozing or crusting) to 4 (severe; deep or bright red erythema with severe induration or papulation with oozing or crusting).⁴⁰

In all 3 ARCADIA studies, IGA was assessed at every study visit. Change from baseline in IGA score and proportion of patients with IGA success, defined as an IGA score of 0 or 1 and a 2-point or more reduction from baseline, were appraised in the review.

No minimal important difference (MID) for the IGA was identified by the sponsor.

Eczema Area and Severity Index

EASI is a composite score commonly used to assess the severity and extent of AD signs. It ranges from 0 to 72, with higher scores indicating greater severity or extent of AD.⁴¹ The severity of erythema, induration or papulation, excoriation, and lichenification was assessed by the investigator or trained designee on a scale of 0 (absent) to 3 (severe) for each of the 4 body areas: head and neck, trunk, upper limbs, and lower limbs, with half points allowed.⁴¹ In addition, the extent of AD involvement in each of the 4 body areas was assessed as a percentage by body area of head, trunk, upper limbs, and lower limbs, and converted to a score of 0 to 6.⁴¹

In all 3 ARCADIA studies, EASI was assessed at every study visit. Change from baseline in EASI, and the proportion of patients with EASI-75 were appraised in the review.

The sponsor provided a MID of 6.6 points for adult patients, but no MID for adolescent patients was reported.

Peak Pruritus Numerical Rating Scale

The PP-NRS is a scale used by the patients to report the intensity of their pruritus (itch) during the last 24 hours; peak pruritus is an assessment of the maximum itch intensity in that period.⁴² Patients are asked for maximum itch intensity (PP-NRS): “on a scale of 0 to 10, with 0 being ‘no itch’ and 10 being the ‘worst itch imaginable,’ how would you rate your itch at the worst moment during the previous 24 hours?”

In the initial period of the ARCADIA 1 and ARCADIA 2 studies, as well as for all of the ARCADIA CYCLO study, the PP-NRS was recorded by patients once daily in the evening, beginning after the screening visit. During the maintenance period of the ARCADIA 1 and ARCADIA 2 studies, PP-NRS was assessed for 7 consecutive days before the visits at week 24, 32, 40, and 48.

The sponsor provided a MID of a change from baseline of 2 or more points to 4 points for within-person change; no MID for between study arms or in adolescents was provided.

HRQoL Outcomes

DLQI and CDLQI Questionnaires

The DLQI is a validated 10-item questionnaire for patients aged older than 16 years, covering domains including symptoms or feelings, daily activities, leisure, work or school, personal relationships, and treatment.⁴³ The CDLQI is a comparable validated 10-item questionnaire designed for pediatric patients aged 16 years or younger. Patients rate each question ranging from 0 (not at all) to 3 (very much). The DLQI or CDLQI total score is calculated by summing the individual question scores. The total score ranges from 0 to 30, with higher scores indicating greater quality of life impairment.

In the ARCADIA 1 and ARCADIA 2 studies, the DLQI or CDLQI, as applicable, was assessed at baseline, week 8, and week 16 in the initial period, and week 32 and 48 of the maintenance period. In the ARCADIA CYCLO study, the DLQI was assessed at baseline, week 8, and week 16.

No MID for the DLQI or CDLQI was identified by the sponsor.

Table 9: Summary of Outcome Measures and Their Measurement Properties

AD = atopic dermatitis; AUC = area under curve; CDLQI = Children’s Dermatology Life Quality Index; CI = confidence interval; CV = coefficient of variation; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; HRQoL = health-related quality of life; ICC = intraclass correlation coefficient; IGA = Investigator Global Assessment; MID = minimal important difference; NRS = numeric rating scale; PCS = Pruritus Categorical Scale; PGADS = Patient Global Assessment of Disease Status; POEM = Patient-Oriented Eczema Measure; PP-NRS = Peak Pruritus Numeric Rating Scale; SCORAD = SCORing Atopic Dermatitis; SD = standard deviation; VAS = visual analogue scale.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁰

Statistical Analysis

Sample Size and Power Calculations

ARCADIA 1 and ARCADIA 2 Studies

To achieve 90% power for both coprimary end points of the ARCADIA 1 and ARCADIA 2 studies, 270 patients (180 nemolizumab, 90 placebo) were randomized to detect the following differences.

• IGA success: To detect a difference of 18%, it was expected that the IGA success at week 16 would be 30% for nemolizumab and 12% for placebo.

• EASI-75 response: To detect a difference of 30%, it was expected that the EASI-75 response at week 16 would be 49% for nemolizumab and 19% for placebo.

To ensure sufficient exposure with nemolizumab and sufficient size of the safety database, the sample size was 750 patients in total with a 2:1 randomization ratio. This sample size provided more than 99% power to detect the treatment difference for both coprimary end points at 2.5% significance level.

ARCADIA CYCLO Study

In the ARCADIA CYCLO study, it was estimated that to achieve 90% power for both coprimary end points, 270 patients (135 nemolizumab, 135 placebo) were to be randomized to detect the following differences.

• EASI-75 at week 16: To detect a difference of 20%, it was expected that the EASI-75 response at week 16 would be 42% for nemolizumab and 22% for placebo.

• PP-NRS at week 16: To detect a difference of 40%, it was expected that the PP-NRS response at week 16 would be 60% for nemolizumab and 20% for placebo.

An attrition rate of 15% was assumed.

Statistical and Analytical Plans

ARCADIA 1 and ARCADIA 2 Studies

Initial Period Analyses

Primary analyses were performed in the ITT population and compared nemolizumab 30 mg every 4 weeks to placebo every 4 weeks. In the initial period of the ARCADIA 1 and ARCADIA 2 studies, the coprimary end points and key secondary end points were analyzed using a Cochran-Mantel-Haenszel test adjusted for the randomized stratification variables (IGA severity [3 = moderate, 4 = severe] and PP-NRS [≥ 7, < 7] for the full population; IGA severity only for the baseline PP-NRS ≥ 7 population), at a 2-sided significance level of 2.5%. The estimate of treatment unadjusted difference with the corresponding 2-sided 95% CIs, as well as the strata-adjusted differences with corresponding 97.5% CIs, were presented.

Binary secondary end points in the initial period were also analyzed using a Cochran-Mantel-Haenszel test, adjusted for randomized stratification variables (IGA severity and PP-NRS for full population; IGA severity only for baseline PP-NRS ≥ 7 population). Analyses were performed for observed cases. In addition, for binary end points of EASI (global score), IGA, and PP-NRS, supplemental analyses, where data collected after rescue and missing data were considered nonresponders, were performed. Continuous secondary end points in the initial period were analyzed using analysis of covariance adjusted for treatment group, the randomized stratification variables, and appropriate baseline values. The missing information under the missing at random assumption approach and the mixed model for repeated measures approach were used to handle the missing data of continuous variables. Missing HRQoL data were imputed using last observation carried forward (LOCF), where applicable.

Maintenance Period Pooled Analysis

Given the similarities in design and conduct of the ARCADIA 1 and ARCADIA 2 studies, a pooled analysis of the maintenance phase results had been conducted. The submission did not provide a detailed rationale for pooling the studies and did not specify whether the decision to pool was preplanned.

In the maintenance period, statistical tests compared the nemolizumab 30 mg every 4 weeks to every 4 weeks and nemolizumab 30 mg every 4 weeks to every 8 weeks treatment groups with the nemolizumab 30 mg every 4 weeks to placebo treatment group. Patients randomized to placebo in the initial treatment period and reassigned to placebo in the maintenance period were not included in the pooled analysis. Week 16 measurements served as baseline measurements (i.e., maintenance baseline). If week 16 measurements were missing for rerandomized participants, the last valid nonmissing measurement in the initial treatment period (i.e., before dosing for the maintenance period at week 16 visit) was taken as the maintenance baseline measurement for the maintenance period.

For the analysis of change in EASI from week 16 up to week 48 and change in PP-NRS from week 16 up to week 48, data collected after the use of rescue medication were imputed to worst possible value and missing values were imputed using a missing information, missing at random assumption. Continuous end points (change from maintenance baseline to each visit in EASI and PP-NRS) and HRQoL (change from maintenance baseline to each visit in CDLQI or DLQI) were analyzed using an analysis of covariance including treatment group and stratification variables study as factors and appropriate maintenance baseline values as covariates, if applicable; the submission did not specify which baseline variables were included in the analyses. The LS mean, estimated standard error, and 95% CIs for each end point were presented for each treatment group and analysis visit. The estimated treatment difference for each end point at each analysis visit was summarized by presenting the difference in LS mean between treatment groups, the 2-sided 95% CIs, and P value.

ARCADIA CYCLO Study

In the ARCADIA CYCLO study, the coprimary end points and binary secondary end points were analyzed using the Cochran-Mantel-Haenszel test, adjusting for the randomized stratification variables (baseline IGA severity [3 = moderate, 4 = severe] and prior CsA exposure). The estimate of treatment difference, the corresponding 2-sided 95% CI, and P values were presented. Missing data were not imputed in this trial; however, for the coprimary end points, a tipping point analysis and analysis based on MI were conducted to assess the robustness of the analyses. Analysis of primary end points was repeated on the per-protocol population. Secondary continuous end points were analyzed using both analysis of covariance adjusted for treatment group, baseline IGA severity, CsA exposure, and appropriate baseline values, and a mixed model for repeated measures adjusting for treatment group, baseline IGA severity, prior CsA exposure, visit, an interaction term between treatment and visit, and appropriate baseline values. The submission did not specify which baseline variables were included in the analyses.

Details of the statistical analysis, adjustments, and handling of missing data, by outcome, are presented for the ARCADIA 1, ARCADIA 2, and ARCADIA CYCLO studies in Table 10.

Table 10: Statistical Analysis of Efficacy End Points

ANCOVA = analysis of covariance; CDLQI = Children’s Dermatology Life Quality Index; CsA = cyclosporine; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; EASI-75 = ≥ 75% improvement in Eczema Area and Severity Index from baseline; IGA = Investigator Global Assessment; LOCF = last observation carried forward; MAR = missing at random; MI = multiple imputation; MMRM = mixed effects model for repeated measures; NA = not applicable; NR = not reported; OC = observation case; PP = per protocol; PP-NRS = Peak Pruritus Numeric Rating Scale; QoL = quality of life.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁰

Subgroup Analyses

In the ARCADIA 1 and ARCADIA 2 studies, subgroup analyses were performed on the ITT population (full population; baseline PP-NRS ≥ 7) for primary and key secondary end points. The following subgroups were descriptively evaluated in the ARCADIA 1 and ARCADIA 2 studies:

• region (Europe, North America, Asia Pacific)

• age groups (12 to 17 years, 18 to 65 years, and > 65 years)

• sex

• race

• baseline IGA (3 or 4)

• baseline PP-NRS (≥ 7 and < 7)

• previous use of any systemic therapy for AD (yes, no)

• previous use of any biologic therapy for AD (yes, no)

• previous use of any immunosuppressive or immunomodulatory drugs for AD (yes, no)

• previous use of dupilumab for AD (yes, no)

• previous use of CsA for AD (yes, no)

• country.

Subgroup results by age group, sex, baseline IGA score, baseline PP-NRS, previous use of any systemic therapy for AD, previous use of any biologic therapy for AD, previous use of any immunosuppressive or immunomodulatory drugs for AD, previous use of dupilumab for AD, and previous use of CsA for AD were of interest to this review.

There were no subgroup analyses in the ARCADIA CYCLO study.

Multiple Testing Procedure

ARCADIA 1 and ARCADIA 2 Studies

Multiplicity control was applied to the coprimary and key secondary end points. To control the type I error at 5%, a serial gatekeeping approach was implemented. The coprimary end points were tested at 2.5% significance level for each population. If both coprimary end points were statistically significant at a 2.5% significance level, the key secondary end points were tested sequentially following a hierarchical testing procedure using the prespecified order of end points for each population; the submission did not list the end points but noted that they were tested in a prespecified order. The comparisons for key secondary end points were made sequentially, and the subsequent testing was stopped when no statistical difference was found. Secondary end points were not controlled for multiplicity.

In the pooled analysis of the maintenance period of the ARCADIA 1 and ARCADIA 2 studies, none of the efficacy analyses were adjusted for multiplicity and they were tested at a 5% significance level.

ARCADIA CYCLO Study

To control the type I error at a 5% significance level, the primary end points were tested sequentially following a hierarchical testing procedure. The proportion of patients with EASI-75 at week 16 was tested first at the 2-sided 0.05 significant level. To preserve the overall alpha level at 0.05 across the 2 primary end points, formal statistical inference for the proportion of patients with an improvement of PP-NRS of 4 or more at week 16 could only be made if superiority of nemolizumab was demonstrated for the first primary efficacy end point, at the 2-sided 0.05 significance level. Apart from the primary end points, no other adjustments for multiplicity were made.

Analysis Populations

The analysis populations used in the ARCADIA 1, ARCADIA 2, and ARCADIA CYCLO studies are provided in Table 11.

Table 11: Analysis Populations of the ARCADIA 1, ARCADIA 2, and ARCADIA CYCLO Studies

ITT = intention to treat; PP = per protocol.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁰

Protocol Amendments and Deviations

The original protocol for the ARCADIA 1 study (dated January 17, 2019) and the ARCADIA 2 study (dated January 29, 2019) were amended 7 times each. Notable amendments for the ARCADIA 1 and ARCADIA 2 studies included allowing the investigator to prescribe background topical therapy and adding the incidence of rescue therapy use as an outcome, correcting and clarifying randomization methods (not specified how), updating statistical analysis and analysis populations (October 2019), clarifying the exclusion criteria around asthma, updating exclusion criteria (not specified how), and updating statistical methodology (not specified how) (June 2020). The first patient was randomized on August 9, 2019, in the ARCADIA 1 study and August 13, 2019, in the ARCADIA 2 study. It is unknown how many patients were impacted by changes to the protocol after enrolment. There was 1 country-specific amendment made to the original protocol of the ARCADIA CYCLO study (dated August 9, 2021), although it was not considered to impact study conduct.

In the ARCADIA 1 study, a total of 172 patients (27.7%) in the nemolizumab arm and 87 patients (27.1%) in the placebo arm had major protocol deviations during the initial treatment period. In the ARCADIA 2 study, a total of 139 patients (26.6%) in the nemolizumab arm and 63 patients (23.8%) in the placebo arm had major protocol deviations during the initial treatment period. In both studies, the most common major protocol deviations pertained to enrolment criteria (ARCADIA 1: 10.8% nemolizumab, 7.8% placebo; ARCADIA 2: 10.0% nemolizumab, 9.4% placebo) and other (ARCADIA 1: 12.3% nemolizumab, 13.1% placebo; ARCADIA 2: 9.0% nemolizumab, 8.7% placebo).

In the maintenance period of the ARCADIA 1 study, 6.7%, 12.1%, and 4.4% of patients in the nemolizumab every 4 weeks to every 4 weeks, nemolizumab every 4 weeks to every 8 weeks, and nemolizumab every 4 weeks to placebo groups had major protocol deviations, respectively. In the maintenance period of the ARCADIA 2 study, 12.7%, 10.3%, and 19.2% of patients in the nemolizumab every 4 weeks to every 4 weeks, nemolizumab every 4 weeks to every 8 weeks, and nemolizumab every 4 weeks to placebo groups had major protocol deviations, respectively.

██ ███████ ██████ █ █████ ██ ██ ███████ ████████ ██ ███ ███████████ ███ ███ ██ ███████ ████████ ██ ███ ███████ ███ ███ █████ ████████ ███████████ ████ ████████ ███ ██ ███████ ████████ ███████████ █████ ████████████ █████ ████████ ███ █████ ██████████████████████ █████ ████████████ ████ █████████.

Results

Patient Disposition

ARCADIA 1 and ARCADIA 2 Studies

Details of the patient disposition for the initial treatment period in the ARCADIA 1 and ARCADIA 2 studies are in Table 12. The submission did not report reasons for unsuccessful screening but reported that 29.4% of patients did not meet screening criteria in the ARCADIA 1 study and 26.5% did not meet screening criteria in the ARCADIA 2 study. A total of 10% or fewer of patients discontinued from the ARCADIA 1 or ARCADIA 2 studies, across treatment arms during the initial treatment period. The reasons for study discontinuation and the proportion of patients who discontinued the study were broadly similar between the nemolizumab and placebo arms of the ARCADIA 1 and ARCADIA 2 studies, with the exception that a numerically higher proportion of patients (3.4%) in the nemolizumab arm discontinued due to AEs than in the placebo arm (1.5%) in the ARCADIA 2 study. A total of 272 patients met the study definition for clinical responders after the initial period of the ARCADIA 1 and ARCADIA 2 studies were rerandomized for follow-up in the maintenance period.

Table 12: Summary of Patient Disposition From the ARCADIA 1 and ARCADIA 2 Studies — Initial Period

ITT = intention to treat; PP = per protocol.

^aUnsuccessful screening percentages were out of the total number screened; the remaining percentages are out of the number of patients randomized.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁰

Details of the patient disposition during the maintenance phase for the ARCADIA 1 and ARCADIA 2 studies are presented in Table 13. In the ARCADIA 1 study, of the 620 patients initially randomized to nemolizumab, a total of 272 patients (43.9%) were rerandomized, and a total of 100 patients (31.2%) initially randomized to placebo were eligible to continue in the maintenance period. A numerically higher proportion of patients in the rerandomized placebo arm discontinued the maintenance treatment in general (24.2% of patients) relative to the other study arms (15.6% nemolizumab every 4 weeks, 13.2% nemolizumab every 8 weeks, and 18.0% placebo). A numerically higher proportion of patients who were rerandomized to placebo discontinued the treatment due to lack of efficacy (13.2% of patients) than in the ARCADIA 2 study (3.8% of patients). Otherwise, the patient disposition was broadly balanced between the study arms in the ARCADIA 1 study.

In the ARCADIA 2 study, of the 522 patients initially rerandomized to nemolizumab, 235 patients (45.0%) were rerandomized, and 85 patients (32.0%) of the original 265 randomized to the placebo arm were eligible to continue in the maintenance period. The proportions of patients discontinuing maintenance treatment and the reasons for discontinuation were broadly similar across treatment arms. With the exception of the proportion of patients in the rerandomized placebo arm discontinuing maintenance treatment (described previously), the proportions of patients discontinuing treatment in the other study arms was generally consistent with the ARCADIA 2 study.

Table 13: Summary of Patient Disposition From the ARCADIA 1 and ARCADIA 2 Studies — Maintenance Period

LTE = long-term extension; Nemo = nemolizumab; NR = not reported; PBO = placebo; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks.

Sources: Details included in the table are from the sponsor’s ARCADIA 1 Clinical Study Report¹¹ and ARCADIA 2 Clinical Study Report.¹²

ARCADIA CYCLO Study

Details of the patient disposition for the ARCADIA CYCLO study are in Table 14. ███ ██████████ ███ ███ ██████ ███████ ███ █████████ ███████ ███ █████ ████ █████ ██ ████████ ██████ ██████████ █ █████ ██ ████ ██ ████████ ████████████ ████ ███ █████ ██ ███ ███████ ████ ███ ████ ████████████ ████ ███ ███████████ ████ ███ ████ ██████ ██████ ███ █████████████ ██ ███ ███████ ███ ███ ███████ ███████ ██████ ███ ██ ███ ███████████ ███ ██ ███ ████ ███████ ███████ ██████ ███ ████ ██ █████████ ███████.

Table 14: Summary of Patient Disposition From the ARCADIA CYCLO Study

ITT = intention to treat; LTE = long-term extension; PP = per protocol.

^aUnsuccessful screening percentages were out of the total number screened; the remaining percentages are out of the number of patients randomized.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁰

Baseline Characteristics

ARCADIA 1 and ARCADIA 2 Studies

The mean age, proportion of female patients, height, weight, and smoking status were similar between the ARCADIA 1 and ARCADIA 2 studies and were also generally balanced between the 2 arms within each study. The mean age in the ARCADIA 1 study was approximately 33 years in both study arms and was 35 years in the ARCADIA 2 study, in both study arms. Of note, the ARCADIA 1 study enrolled a higher proportion of patients who self-identified as Asian (15.8% in the nemolizumab arm, 16.2% in the placebo arm) than the ARCADIA 2 study (1.9% in the nemolizumab arm, 1.5% in the placebo arm), and a lower proportion of white patients (72.7% in the nemolizumab arm, 76.0% in the placebo arm), than the ARCADIA 2 study (87.7% in the nemolizumab arm, 85.7% in the placebo arm).

In terms of AD and prior treatment characteristics, the 2 studies were broadly similar and within each study, and the 2 treatment arms were generally balanced. The mean EASI score was approximately 27 in both treatment arms in both studies, and the majority of patients reported a mean IGA score of 3 (ARCADIA 1: 70.6% in the nemolizumab arm, 73.5% in the placebo arm; ARCADIA 2: 67.4% in the nemolizumab arm, 69.8% in the placebo arm). The weekly average PP-NRS score was approximately 7 in both treatment arms in both studies. The average BSA involvement was also consistent across treatment arms within the studies (ARCADIA 1: 44.9% in the nemolizumab arm, 43.8% in the placebo arm; ARCADIA 2: 44.6% in the nemolizumab arm, 45.0% in the placebo arm). The total mean DLQI score was approximately 15 in the ARCADIA 1 study and 14 in the ARCADIA 2 study; the mean total CDLQI score was approximately 12 in the ARCADIA 1 and ARCADIA 2 studies, in all study arms. Prior treatment characteristics were reasonably balanced across treatment arms; the majority of patients (> 97% in all arms in the ARCADIA 1 and ARCADIA 2 studies) reported the use of moderately potent corticosteroids before enrolment, and approximately one-half (48% to 55% across all arms in the ARCADIA 1 and ARCADIA 2 studies) reported the use of other drugs. Immunomodulating drugs were the most commonly reported systemic drug in both trials (ARCADIA 1: 38.1% in the nemolizumab arm, 36.4% in the placebo arm; ARCADIA 2: 29.1% in the nemolizumab arm, 31.7% in the placebo arm); less than 15% of patients in both arms of both trials reported the use of CsA, and less than 8% reported the use of biologics.

Details of the baseline characteristics from ARCADIA 1 and ARCADIA 2 studies are outlined in Table 15; these limited to those that are most relevant to this review or were felt to affect the outcomes or interpretation of the study results.

Table 15: Summary of Baseline Characteristics From the ARCADIA 1 and ARCADIA 2 Studies (ITT) — Initial Treatment Period

AD = atopic dermatitis; AP-NRS = Average Pruritus Numeric Rating Scale; BSA = body surface area; CDLQI = Children’s Dermatology Life Quality Index; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; IGA = Investigator Global Assessment; ITT = intention to treat; PP-NRS = Peak Pruritus Numeric Rating Scale; SCORAD = SCORing Atopic Dermatitis; SD = standard deviation; SD-NRS = Sleep Disturbance Numeric Rating Scale.

Note: Racial categories used in the table are as reported in the source and may not align with Canada's Drug Agency inclusive language guidelines.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁰

In the maintenance period of the ARCADIA 1 study, 90 patients were rerandomized from nemolizumab every 4 weeks to every 4 weeks, 91 patients were rerandomized from nemolizumab every 4 weeks to every 8 weeks, 91 patients were rerandomized from nemolizumab every 4 weeks to placebo, and 100 patients from the initial placebo group were reassigned to placebo. Across treatment arms, the demographic and AD characteristics postrandomization in the maintenance population from the ARCADIA 1 study were broadly balanced. Relative to the patients who responded to placebo, those who responded to treatment in the intervention arm had a numerically higher proportion of female patients (58.9% to 61.5% in the nemolizumab arms, 47.0% in the placebo arm). The majority of the maintenance population was white (77.8% to 86.0% across all 4 treatment arms), with an EASI score ranging from 2.74 to 3.1 across study arms, and a PP-NRS score ranging from 2.13 to 3.34 across study arms. Detailed baseline characteristics of the maintenance population are in Table 16.

Table 16: Summary of Baseline Characteristics From the ARCADIA 1 Study (ITT) — Maintenance Treatment Period

AD = atopic dermatitis; AP-NRS = Average Pruritus Numeric Rating Scale; BSA = body surface area; CDLQI = Children’s Dermatology Life Quality Index; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; IGA = Investigator Global Assessment; ITT = intention to treat; PP-NRS = Peak Pruritus Numeric Rating Scale; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; SCORAD = SCORing Atopic Dermatitis; SD = standard deviation; SD-NRS = Sleep Disturbance Numeric Rating Scale.

Note: Racial categories used in the table are as reported in the source and may not align with Canada's Drug Agency inclusive language guidelines.

Source: Details included in the table are from the sponsor’s ARCADIA 1 Clinical Study Report.¹¹

In the maintenance phase of the ARCADIA 2 study, 79 patients were rerandomized from nemolizumab every 4 weeks to every 4 weeks, 78 patients were rerandomized from nemolizumab every 4 weeks to every 8 weeks, 78 patients were rerandomized from nemolizumab every 4 weeks to placebo, and 85 patients from the initial placebo group were reassigned to placebo. The maintenance population had mean EASI and PP-NRS scores similar to the maintenance population of the ARCADIA 1 study (EASI ranged from 1.85 to 3.39 across all 4 study arms; PP-NRS ranged from 2.03 to 3.60 across all 4 study arms). Of note, there were some numeric imbalances between the rerandomized study arms in the ARCADIA 2 study. A total of 11.5% of patients in the nemolizumab every 4 weeks to every 8 weeks arm reported an IGA score of 2, relative to 19.2% in the nemolizumab every 4 weeks to placebo arm; 33.3% of patients in the nemolizumab every 4 weeks to placebo arm were male, relative to 47.4% in the nemolizumab every 4 weeks to every 8 weeks arm; and patients in the nemolizumab every 4 weeks to every 8 weeks arm had 5.9% mean BSA involvement, relative to nemolizumab every 4 weeks to placebo with 11.5% mean BSA involvement. Detailed baseline characteristics for the ARCADIA 2 study maintenance period are in Table 17.

Table 17: Summary of Baseline Characteristics From the ARCADIA 2 Study (ITT) — Maintenance Treatment Period

AD = atopic dermatitis; AP-NRS = Average Pruritus Numeric Rating Scale; BSA = body surface area; CDLQI = Children’s Dermatology Life Quality Index; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; IGA = Investigator Global Assessment; ITT = intention to treat; PP-NRS = Peak Pruritus Numeric Rating Scale; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; SCORAD = SCORing Atopic Dermatitis; SD = standard deviation; SD-NRS = Sleep Disturbance Numeric Rating Scale.

Note: Racial categories used in the table are as reported in the source and may not align with Canada's Drug Agency inclusive language guidelines.

Source: Details included in the table are from the sponsor’s ARCADIA 2 Clinical Study Report.¹²

ARCADIA CYCLO Study

Details of the baseline characteristics from the ARCADIA CYCLO study are outlined in Table 18. ███████ ███ ████████ ███████████ ████████████████ ██ ███████ ████████████████ ███ █████ █████████ ███████████████ ████ ███████ ████████ ███████ █████ █████ ███ ████ ███ ██ ████████ ███ ████ ██ ███ ███████████ ████ ████ ██ ███ ███████ ████ ███ ███ ████████ ██ ████████ ██ ████ █████████ ████ ████ █████ ██████ ███████████ ████ █████ ███████ █████ ███ ███████ ████ █████ ███ █████ ██ ███ ███████████ ████ █████ ██ ███ ███████ ████ ███ ███ ████████ ██ ████████ ██ ████ ████ ████████ ██ ███ █████ ██ █ ██████████████ ██ ██ ████ ██████ ███ ████ ██████ ██████ █████ ███ █████████████ ███ ██ ████ █████ ███ ███ ████ ███ ███ █████████████ ███ ██ ████ █████ ███ ████████ ██ ████████ ██████ ██ ████ █████ █████ ████████ ████████ █████████████████ ████ ████ ███ ████ ██████ █████ ████████ ██████████████████ ██████ ███████████ ████ █████ ███████ ████ ███ ████████ ███████████████ ██████ ███████████ ████ █████ ███████ █████ █ █████ ██ █████ ██ ████████ ██ ███ ███████████ ███ ███ █████ ██ ████████ ██ ███ ███████ ███ ████████ ████████ ███ █████████ ███ ████ ████████ ████████ ██████ ███ ████████ ███ ████ ██████ █████ ███ ███ ███ ███████████ ██████ ████████ ████ ████████ ██████ ██ ████████ ██ ███ ███████████ ████ █████ ██ ████████ ██ ███ ███████ █████.

Table 18: Summary of Baseline Characteristics From the ARCADIA CYCLO Study (ITT)

AD = atopic dermatitis; AP-NRS = Average Pruritus Numeric Rating Scale; BSA = body surface area; CsA = cyclosporine; DLQI = Dermatology Life Quality Index; EASI = Eczema Area and Severity Index; IGA = Investigator Global Assessment; ITT = intention to treat; PP-NRS = Peak Pruritus Numeric Rating Scale; SCORAD = SCORing Atopic Dermatitis; SD = standard deviation; SD-NRS = Sleep Disturbance Numeric Rating Scale.

Note: Racial categories used in the table are as reported in the source and may not align with Canada's Drug Agency inclusive language guidelines.

Sources: Details included in the table are from the sponsor’s summary of clinical evidence¹⁰ and ARCADIA CYCLO Clinical Study Report.¹³

Exposure to Study Treatments

Details of the adherence and exposure to study treatments are in Table 19 (the initial period of the ARCADIA 1 and ARCADIA 2 studies), Table 20 (the maintenance period of the ARCADIA 1 and ARCADIA 2 studies), and Table 21 (the ARCADIA CYCLO study). Briefly, patient adherence was greater than ███ in the initial period of the ARCADIA 1 and ARCADIA 2 studies, greater than ███ in the maintenance period of the ARCADIA 1 and ARCADIA 2 studies, and greater than ███ in the ARCADIA CYCLO studies. Treatment duration during the initial 16 weeks was consistent between the ARCADIA studies, ranging from 79.9 days to 83.4 days, and was consistent across the ARCADIA 1 and ARCADIA 2 studies during the maintenance phase.

Table 19: Summary of Patient Adherence From the ARCADIA 1 and ARCADIA 2 Studies (SAS) — Initial Period

SAS = safety analysis set; SD = standard deviation.

Sources: Details included in the table are from the sponsor’s ARCADIA 1 Clinical Study Report¹¹ and ARCADIA 2 Clinical Study Report.¹²

Table 20: Summary of Patient Adherence From the ARCADIA 1 and ARCADIA 2 Studies (SAS) — Maintenance Period

Nemo = nemolizumab; PBO = placebo; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; SAS = safety analysis set; SD = standard deviation.

Sources: Details included in the table are from the sponsor’s ARCADIA 1 Clinical Study Report¹¹ and ARCADIA 2 Clinical Study Report.¹²

Table 21: Summary of Patient Adherence From the ARCADIA CYCLO Study

SD = standard deviation.

Source: Details included in the table are from the sponsor’s ARCADIA CYCLO Clinical Study Report.¹³

Concomitant Medications

Details of the most common (i.e., ≥ 10% of patients) concomitant medications for the initial period of the ARCADIA 1 and ARCADIA 2 studies are in Table 22. Briefly, all patients in the ARCADIA 1 and ARCADIA 2 studies reported concomitant medication use. The most commonly reported type of concomitant medication in both the ARCADIA 1 and ARCADIA 2 studies was emollients and protectives (ARCADIA 1: 69.8% of patients in the nemolizumab arm, 70.4% of patients in the placebo arm; ARCADIA 2: 76.8% of patients in the nemolizumab arm, 74.7% of patients in the placebo arm), as well as group II (potent) corticosteroids (ARCADIA 1: 68.1% of patients in the nemolizumab arm, 70.4% of patients in the placebo arm; ARCADIA 2: 75.4% of patients in the nemolizumab arm, 78.5% of patients in the placebo arm). The use of concomitant medications was broadly balanced between study arms within each study, and also between the ARCADIA 1 and ARCADIA 2 studies. The submission did not specify whether the corticosteroids were used as authorized background therapy.

Table 22: Summary of Most Common (10% or More of Patients) Concomitant Medications in the ARCADIA 1 and ARCADIA 2 Studies (ITT) — Initial Period

ATC = anatomic therapeutic class; ITT = intention to treat.

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

Details of the concomitant medications for the ARCADIA CYCLO study are in Table 23. All patients, except for 1 patient in both arms, reported concomitant medication use. Similar to the ARCADIA 1 and ARCADIA 2 studies, the most commonly reported medication class was potent corticosteroids (█████ ████████████ █████ ███████). The second most commonly reported type of medication was emollients and protectives (█████ ███████████ █████████ █████ ███████ patients) potent corticosteroids (█████ nemolizumab, █████ placebo).

Table 23: Summary of Most Common (10% or More of Patients) Concomitant Medications in the ARCADIA CYCLO Study (ITT)

ATC = anatomic therapeutic category; ITT = intention to treat.

Source: Details included in the table are from the sponsor’s ARCADIA CYCLO Clinical Study Report.

Efficacy

ARCADIA 1 and ARCADIA 2 Studies: Initial Period Results

Detailed results from the initial period in the ARCADIA 1 and ARCADIA 2 studies are presented in Table 24.

Proportion of Patients With IGA Success at Week 16

For patients in the ARCADIA 1 study, at week 16, 35.6% of patients in the nemolizumab arm and 24.6% of patients in the placebo arm attained IGA success (defined as IGA of 0 or 1 and a ≥ 2-point reduction from baseline). The strata-adjusted difference between arms was 11.5% (97.5% CI, 4.7% to 18.3%; P = 0.0003). For patients in the ARCADIA 2 study, at week 16, 37.7% of patients in the nemolizumab arm and 26.0% of patients in the placebo arm attained IGA success. The strata-adjusted difference between arms was 12.2% (97.5% CI, 4.6% to 19.8%; P = 0.0006).

Subgroup analyses for the proportion of patients with IGA success at week 16 in the ARCADIA 1 and ARCADIA 2 studies are summarized in Figure 10, Figure 11, Figure 12, Figure 13, Figure 14, Figure 15, Figure 16, and Figure 17, with pooled results summarized in Figure 18, Figure 19, Figure 20, and Figure 21.

Proportion of Patients With EASI-75 at Week 16

For patients in the ARCADIA 1 study, at week 16, 43.5% of patients in the nemolizumab arm and 29.0% of patients in the placebo arm attained EASI-75. The strata-adjusted difference between arms was 14.9% (97.5% CI, 7.8% to 22.0%; P  < 0.0001). For patients in the ARCADIA 2 study, at week 16, 42.1% of patients in the nemolizumab arm and 30.2% of patients in the placebo arm attained EASI-75. The strata-adjusted difference between arms was 12.5% (97.5% CI, 4.6% to 20.3%; P = 0.0006).

Subgroup analyses for the proportion of patients with EASI-75 at week 16 in the ARCADIA 1 and ARCADIA 2 studies are summarized in Figure 10, Figure 11, Figure 12, Figure 13, Figure 14, Figure 15, Figure 16, and Figure 17, with pooled results summarized in Figure 18, Figure 19, Figure 20, and Figure 21.

Change From Baseline to Week 16 in Total EASI Score

For patients in the ARCADIA 1 study, the LS mean change from baseline in total EASI score was –12.36 (95% CI, –13.96 to –10.76) in the nemolizumab arm, and –8.68 (95% CI, –10.77 to –6.58) in the placebo arm. The LS mean difference was –3.68 (95% CI, –6.08 to –1.28; P = 0.0026). For patients in the ARCADIA 2 study, the LS mean change from baseline in total EASI score was –13.15 (95% CI, –14.57 to –11.73) in the nemolizumab arm and –9.02 (95% CI, –10.97 to –7.08) in the placebo arm. The LS mean difference between study arms was –4.13 (95% CI, –6.36 to –1.89; P = 0.0003).

Proportion of Patients With an Improvement of 4 or More in PP-NRS at Week 16

For patients in the ARCADIA 1 study, at week 16, 42.7% of patients in the nemolizumab arm and 17.8% of patients in the placebo arm reported a reduction in PP-NRS score of 4 points or greater. The strata-adjusted LS mean difference between arms was 24.9% (97.5% CI, 18.4% to 31.5%; P < 0.0001). For patients in the ARCADIA 2 study, at week 16, 41.0% of patients in the nemolizumab arm and 18.1% of patients in the placebo arm reported a reduction in PP-NRS score of 4 points or greater. The strata-adjusted LS mean difference between arms was 23.2% (97.5% CI, 16.1% to 30.3%; P < 0.0001).

Subgroup analyses for the proportion of patients with PP-NRS scores of 4 or more at week 16 in the ARCADIA 1 and ARCADIA 2 studies are summarized in Figure 10, Figure 11, Figure 12, Figure 13, Figure 14, Figure 15, Figure 16, and Figure 17, with pooled results summarized in Figure 18, Figure 19, Figure 20, and Figure 21.