Reimbursement Review

Baricitinib (Olumiant)

Sponsor: Eli Lilly Canada Inc.

Therapeutic area: Alopecia areata, severe

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Clinical Review

Abbreviations

Executive Summary

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

Table 1: Background Information on Application Submitted for Review

NOC = Notice of Compliance.

Introduction

Alopecia areata (AA) is a chronic autoimmune disease characterized by nonscarring hair loss at the scalp as well as the loss of eyebrow (EB), eyelash (EL), beard, pubic, or axillary hair. The onset of hair loss in AA is typically rapid and the progression is unpredictable, with the majority of patients experiencing disease onset by aged 40 years.¹ AA is associated with psychological impacts and impairment in health-related quality of life (HRQoL). It is estimated that the prevalence of AA in Canada is between 0.1% and 0.58%.^2-5 Canadian clinical practice guidelines for AA are not available to date. As per input from the clinical experts consulted by Canada’s Drug Agency (CDA-AMC), clinicians in Canada consider systemic drugs for the treatment of adults with severe AA, including off-label conventional immunosuppressants (cyclosporine, methotrexate, azathioprine, and mycophenolate mofetil) and Janus kinase (JAK) inhibitors (i.e., ritlecitinib, which has been recently approved by Health Canada for the treatment of adults and adolescents aged 12 years and older with severe AA, as well as tofacitinib, upadacitinib, and abrocitinib, which are off-label treatments for severe AA). Conventional immunosuppressants are currently reimbursed by the public drug plans in Canada. The clinical experts noted that conventional immunosuppressants are associated with poor efficacy, a risk of relapse with dose reduction and/or discontinuation, as well as potential serious adverse events (SAEs) when used long-term.

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of baricitinib 2 mg and baricitinib 4 mg oral tablets in the treatment of adult patients with severe AA.

Perspectives of Patients, Clinicians, and Drug Programs

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

Patient Input

CDA-AMC received 1 patient group submission from the Canadian Alopecia Areata Foundation (CANAAF). CANAAF was registered as a charitable organization in 2010 and is described as the voice for all patients and families affected by AA living in Canada. CANAAF collected data on the psychosocial and emotional impact of AA from peer-reviewed literature, as well as patient perspectives on AA from patient reports and support sessions.

CANAAF commented that AA is incredibly burdensome on a patient’s mental health and quality of life, and the disease causes disfiguring hair loss that occurs unexpectedly and can progress rapidly. Based on a patient report, CANAAF further stated that the anxiety, depression, and other resultant psychological conditions are not minor in nature; therefore, the loss of hair can create layers of stigma and misunderstandings. Short hair or baldness may be associated with a preference for an “edgy” look or having a certain sexuality, which may not be accurate. Those with this disease may feel less feminine or less masculine without hair. Children and teenagers may experience bullying. In addition, CANAAF revealed that there is also a significant financial burden associated with AA; this was supported by the findings of a CANAAF community alopecia patient focus group conducted in 2023. The most significant cost item was a wig purchase and maintenance, which can cost more than $2,500 a year. Some patients experienced significant impacts on their ability to work.

Based on the literature, CANAAF identified limitations of the currently available treatments for AA, including topical corticosteroids (limited effectiveness, effective only for patients with very limited AA, difficult product application, and scalp irritation), intralesional corticosteroids (painful injections and limited drug coverage by drug plans), oral corticosteroids (variable success rates, a high relapse rate, limited drug coverage, and unfavourable side effects), topical minoxidil (nondurable benefits for very mild AA, and adverse events [AEs] such as excessive hair growth on body parts other than the site of application, irritation, and allergic contact dermatitis), oral minoxidil (systemic AEs relating to its antihypertensive property and limited drug coverage), and systemic immunosuppressants (variable effectiveness; a risk of organ toxicity, infection, and malignancy; concomitant administration of oral corticosteroids required for some drugs; and limited drug coverage).

CANAAF identified a need for an effective treatment option that could result in full and sustained hair growth and alleviate anxiety and depression associated with AA. CANAAF believed that baricitinib may fulfill this need by serving as an effective treatment that has a favourable side effect profile and is easy to administer. The group noted that most patients regrew all their hair with baricitinib treatment. CANAAF also noted that the side effect profile of baricitinib is much more favourable compared to existing treatments. Baricitinib is also a much easier treatment option for patients as it only requires that they take 1 pill, once a day. This is in comparison to other treatments that must be applied topically, injected (often by a health care professional), or taken orally more than once a day.

Clinician Input

Input From Clinical Experts Consulted by CDA-AMC

The clinical experts we consulted noted that currently reimbursed off-label systemic treatments for severe AA are associated with poor efficacy, a risk of relapse with dose reduction and/or discontinuation, and potential SAEs when used long-term (especially with conventional immunosuppressants). Also, access to emerging therapies, such as ritlecitinib, is currently limited as per clinical expert input. The clinical experts noted that, because of the limited efficacy of the conventional systemic immunomodulators, it is rational to use baricitinib (and JAK inhibitors in general) as a first-line systemic therapy in severe AA rather than as the last line of treatment after the failure of conventional systemic immunomodulators. The clinical experts noted that it would be appropriate to use baricitinib in combination with topical treatments and/or intralesional corticosteroids but not in combination with other immunomodulators, except for prednisone where concomitant use with baricitinib may be appropriate.

In the clinical experts’ opinion, patients who have severe AA with scalp involvement as reflected by a Severity of Alopecia Tool (SALT) score of 50 or more and have a current episode of AA of a duration greater than 1 year but less than 10 years are potential candidates for baricitinib treatment, though the clinical experts noted that adhering to the inclusion criterion on the duration of a current episode used in the pivotal trials (i.e., more than 6 months and less than 8 years in duration) would also be reasonable. One clinical expert considered the use of baricitinib in older adults (i.e., older than 60 years for males or older than 70 years for females, which were categories of patients excluded from the pivotal trials) to be reasonable, while the other clinical expert suggested restricting the use of baricitinib as per the age restriction in the pivotal trials because of a lack of clinical trial data and unknown clinical treatment benefits.

The clinical experts felt that it is reasonable to define meaningful response to treatment as the attainment of a SALT score of 20 or less after 36 weeks of baricitinib treatment, consistent with the pivotal trials. The clinical experts noted that it would be reasonable to consider the discontinuation of baricitinib treatment in patients who do not attain cosmetically acceptable hair regrowth at 36 weeks, who have further loss of hair at 36 weeks, who experience severe AEs deemed to be related to the use of a JAK inhibitor, or who develop intercurrent conditions, making the discontinuation of a JAK inhibitor advisable (e.g., malignancy). In the clinical expert’s opinion, baricitinib treatment should be prescribed by dermatologists with experience in diagnosing, treating, and monitoring patients with severe AA.

Clinician Group Input

We received no clinician group input for the drug under review.

Drug Program Input

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

• relevant comparators

• considerations for the initiation of therapy

• considerations for the continuation or renewal of therapy

• considerations for the discontinuation of therapy

• considerations for the prescribing of therapy

• generalizability

• care provision issues

• system and economic issues.

The clinical experts we consulted 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

The sponsor-conducted systematic literature review identified 2 pivotal double-blind, randomized, placebo-controlled trials (Study BRAVE-AA1, N = 654; Study BRAVE-AA2, N = 546)⁶ that assessed the efficacy and safety of baricitinib relative to placebo in adult patients who had severe or very severe AA with at least 50% scalp involvement (i.e., a SALT score of at least 50) and had a current AA episode lasting more than 6 months and less than 8 years. In the double-blind, placebo-controlled treatment period, patients were randomized in a 2:2:3 ratio to receive placebo, baricitinib 2 mg, and baricitinib 4 mg once daily for 36 weeks, at which time the primary analysis of efficacy and safety was conducted. In the 68-week long-term extension period, patients continued the existing intervention or were reassigned a new intervention (placebo, baricitinib 2 mg, or baricitinib 4 mg), depending on response to treatment at week 36 (for patients initially assigned to placebo) or week 52 (for patients initially assigned to baricitinib 2 mg or baricitinib 4 mg) as per-protocol-defined criteria. This was followed by a 96-week bridging extension where patients continued to receive the same intervention until the end of the study. The long-term extension period is ongoing in both trials.

Efficacy end points of interest to this review included the proportion of patients attaining a SALT score of 20 or less (the primary end point), a response of at least a 50% reduction in the Severity of Alopecia Tool score from baseline (SALT₅₀), clinician-reported outcome (ClinRO) measures for EB and EL hair loss scores of 0 or 1 with at least a 2-point reduction from baseline (key secondary end points), change from baseline in Hospital Anxiety and Depression Scale (HADS) anxiety and depression domain scores, and Skindex-16 for Alopecia Areata (Skindex-16 for AA) symptoms, emotions, and functioning domain scores (secondary or exploratory outcomes). All of these were assessed at week 36.

In both trials, at baseline, there was about an equal proportion of patients with severe AA and very severe AA. The mean duration of the current AA episode was 3.6 (standard deviation [SD] = 3.9) years and 4.3 (SD = 4.9) years in Study BRAVE-AA1 and Study BRAVE-AA2, respectively. Approximately 90% of patients had received prior AA treatment, with the most common ones (reported in at least 40% of patients) being topical therapies, intralesional therapy, and systemic immunosuppressants and immunomodulators.

Efficacy Results

Proportion of Patients Attaining SALT Score of 20 or Less

The proportion of patients attaining a SALT score of 20 or less at week 36 was the primary end point in both trials. At week 36, the between-group difference comparing baricitinib 2 mg versus placebo was 16.4% (95% confidence interval [CI], 9.7% to 23.4%; P < 0.001) in Study BRAVE-AA1 and 14.7% (95% CI, 8.3% to 21.6%; P < 0.001) in Study BRAVE-AA2. The between-group difference comparing baricitinib 4 mg and placebo was 29.9% (95% CI, 23.2% to 36.2%; P < 0.001) in Study BRAVE-AA1 and 29.9% (95% CI, 23.1% to 36.3%; P < 0.001) in Study BRAVE-AA2. Results were in favour of both regimens of baricitinib treatment. In both trials, subgroup analyses by baseline disease severity and duration of the current episode of AA were consistent with the primary analysis.

The percentage change from baseline in the SALT score was assessed at week 36 in both trials (a key secondary end point). In both trials, the between-group difference comparing baricitinib and placebo was in favour of baricitinib for both the 2 mg regimen (–23.1% [95% CI, –30.6% to –15.6%; P < 0.001] in Study BRAVE-AA1 and –25.3% [95% CI, –32.8% to –17.7%] in Study BRAVE-AA2) and the 4 mg regimen (–37.7% [95% CI, –44.4% to –30.9%; P < 0.001] in Study BRAVE-AA1 and 44.5% [95% CI, –51.3% to –37.7%; P < 0.001] in Study BRAVE-AA2).

Proportion of Patients Attaining SALT₅₀ Response

The between-group difference in the proportion of patients attaining a SALT₅₀ response at week 36 (a secondary end point) comparing baricitinib 2 mg versus placebo was 17.7% (95% CI, 9.5% to 25.8%; P < 0.001) in Study BRAVE-AA1 and 23.1% (95% CI, 15.1% to 31.0%; P < 0.001) in Study BRAVE-AA2. The between-group difference comparing baricitinib 4 mg with placebo was 33.6% (95% CI, 25.6% to 40.7%; P < 0.001) in Study BRAVE-AA1 and 41.9% (95% CI, 34.0% to 48.7%; P < 0.001) in Study BRAVE-AA2. Results of the responder analysis for at least a 75% reduction in the Severity of Alopecia Tool score from baseline (SALT₇₅) were consistent with the SALT₅₀ responder analysis. Neither end point was adjusted for multiplicity in the trials.

Proportion of Patients Attaining ClinRO Measure for EB Hair Loss Score of 0 or 1 With an Improvement of 2 Points or More From Baseline Among Patients With ClinRO Measure for EB Hair Loss Score of 2 or More at Baseline

Between 66.3% and 73.9% of all randomized patients had a ClinRO measure for EB hair loss score of at least 2 at baseline in the trials and contributed to the analysis of the proportion of patients with a ClinRO measure for EB hair loss score of 0 or 1 with at least a 2-point improvement from baseline at week 36 (a key secondary end point).

The between-group difference comparing baricitinib 2 mg versus placebo was 15.9% (95% CI, 8.4% to 23.6%; P < 0.001) in favour of baricitinib 2 mg in Study BRAVE-AA1 and 7.1% (95% CI, –0.3% to 15.0%; P = 0.08) in Study BRAVE-AA2. In Study BRAVE-AA2, no formal testing was conducted for subsequent end points in the statistical hierarchy because of the failure of this end point in the study. The between-group difference was in favour of baricitinib 4 mg over placebo in both trials (28.2% [95% CI, 20.3% to 35.4%; P < 0.001] in Study BRAVE-AA1 and 30.3% [95% CI, 21.4% to 38.4%; P < 0.001] in Study BRAVE-AA2). Results based on the patient-reported outcome (PRO) measure showed consistent results.

Proportion of Patients Attaining ClinRO Measure for EL Hair Loss Score of 0 or 1 With an Improvement of 2 Points or More From Baseline Among Patients With ClinRO Measure for EL Hair Loss Score of 2 or More at Baseline

Between 51.3% and 60.3% of all randomized patients had a ClinRO measure for EL hair loss score of at least 2 at baseline in the trials and contributed to the analysis of the proportion of patients who had a ClinRO measure for EL hair loss score of 0 or 1 with at least a 2-point improvement from baseline at week 36 (a key secondary end point).

The between-group difference comparing baricitinib 2 mg and placebo was 10.4% (95% CI, 2.7% to 18.3%) in Study BRAVE-AA1 and 4.6% (95% CI, –3.7% to 13.2%) in Study BRAVE-AA2; neither difference was formally tested for statistical significance because of the prior failure of an outcome in the statistical hierarchy. The between-group difference favoured baricitinib 4 mg treatment over placebo in both trials (30.4% [95% CI, 21.6% to 38.1%; P < 0.001] in Study BRAVE-AA1 and 28.7% [95% CI, 18.7% to 37.5%; P < 0.001] in Study BRAVE-AA2). Results based on the PRO measure showed consistent results.

Change From Baseline in HADS Anxiety Domain Score

The between-group difference comparing baricitinib 2 mg and placebo with respect to change from baseline in the HADS anxiety domain score at week 36 (a secondary end point) favoured baricitinib 2 mg in Study BRAVE-AA1 at –0.8 (95% CI, –1.4 to –0.3; P ≤ 0.01) and at 0.2 (95% CI, –0.8 to 0.4; P = 0.5) in Study BRAVE-AA2. The between-group difference comparing baricitinib 4 mg and placebo favoured baricitinib 4 mg in both trials (–0.5 [95% CI, –1.1 to 0.0; P = 0.04] in Study BRAVE-AA1 and –0.7 [95% CI, –1.3 to –0.2; P = 0.01] in Study BRAVE-AA2). This end point was not adjusted for multiplicity.

Change From Baseline in HADS Depression Domain Score

The between-group difference comparing baricitinib 2 mg and placebo with respect to change from baseline in the HADS depression domain score at week 36 (a secondary end point) was –0.42 (95% CI, –0.93 to 0.109; P = 0.107) in Study BRAVE-AA1 and 0.51 (95% CI, –1.108 to 0.107; P = 0.083) in Study BRAVE-AA2. The between-group difference comparing baricitinib 4 mg and placebo favoured baricitinib 4 mg in Study BRAVE-AA2 at –0.768 (95% CI, –1.20 to –0.216; P = 0.010) and at –0.32 (95% CI, –0.78 to 0.14; P = 0.174) in Study BRAVE-AA1. This end point was not adjusted for multiplicity.

Change From Baseline in Skindex-16 for AA Symptoms Domain Score

The difference between baricitinib 2 mg and placebo with respect to change from baseline in the Skindex-16 AA symptoms domain score at week 36 favoured baricitinib 2 mg in Study BRAVE-AA1 at –4.76 (95% CI, –9.13 to –0.40; P = 0.033) and at –3.02 (95% CI, –6.91 to 0.88; P = 0.129) in Study BRAVE-AA2. The difference between baricitinib 4 mg and placebo favoured baricitinib 4 mg in Study BRAVE-AA2 at –4.21 (95% CI, –7.75 to –0.68; P = 0.020) and at –2.75 (95% CI, –6.67 to 1.17; P = 0.168) in Study BRAVE-AA1. This was an exploratory end point in Study BRAVE-AA1 and a secondary end point in Study BRAVE-AA2. It was not adjusted for multiplicity.

Change From Baseline in Skindex-16 for AA Emotions Domain Score

The between-group difference with respect to change from baseline in the Skindex-16 for AA emotions domain score at week 36 was in favour of baricitinib over placebo in both trials for both the baricitinib 2 mg regimen (–11.50 [95% CI, –17.71 to –5.30; P < 0.001] in Study BRAVE-AA1 and –6.75 [95% CI, –12.68 to –0.82; P = 0.026] in Study BRAVE-AA2) and the baricitinib 4 mg regimen (–11.01 [95% CI, –16.57 to –5.45; P < 0.001] in Study BRAVE-AA1 and –13.42 [95% CI, –18.80 to –0.84; P < 0.001] in Study BRAVE-AA2). This was an exploratory end point in Study BRAVE-AA1 and a secondary end point in Study BRAVE-AA2. It was not adjusted for multiplicity.

Change From Baseline in Skindex-16 for AA Functioning Domain Score

The difference between baricitinib 2 mg and placebo with respect to change from baseline in the Skindex-16 for AA functioning domain score at week 36 was –5.07 (95% CI, –10.94 to 0.80; P = 0.090) in Study BRAVE-AA1 and –4.38 (95% CI, –9.65 to 0.88; P = 0.103) in Study BRAVE-AA2. The difference between baricitinib 4 mg and placebo favoured baricitinib 4 mg in both trials (–7.04 [95% CI, –12.31 to –1.77; P = 0.009] in Study BRAVE-AA1 and –8.33 [95% CI, –13.10 to –3.56; P < 0.001] in Study BRAVE-AA2). This was an exploratory end point in Study BRAVE-AA1 and a secondary end point in Study BRAVE-AA2. It was not adjusted for multiplicity.

Harms Results

Treatment-Emergent Adverse Events, SAEs, Withdrawal Due to Adverse Events, and Mortality

Treatment-emergent adverse events (TEAEs) were reported in 50.8% to 68.4% of patients across the trials and occurred in similar proportions of patients across treatment groups. The most common TEAEs of baricitinib (reported in at least 5% of patients in either baricitinib group) were upper respiratory tract infection, headache, urinary tract infection, nasopharyngitis, acne, and increased blood creatine phosphokinase. SAEs (1.6% to 3.4%) and withdrawal due to adverse events (WDAEs) (1.1% to 2.6%) were uncommon in the studies. No deaths were reported in the trials.

Notable Harms (Infections, Cardiovascular and Thromboembolic Events, Gastrointestinal Perforations, Malignancies)

Treatment-emergent infections were reported in 25.1% to 37.4% of patients across treatment groups in the trials. In Study BRAVE-AA2, the frequency of infection was higher in the baricitinib 2 mg group (37.4% of patients) compared with the placebo group (29.2% of patients), but this was not observed in Study BRAVE-AA1. In Study BRAVE-AA1, none of the infections was reported to be serious or leading to treatment discontinuation. In Study BRAVE-AA2, serious infection was reported in 2 (1.3%) patients and 1 (0.4%) patient in the baricitinib 2 mg and baricitinib 4 mg groups, respectively, and infection leading to treatment discontinuation was reported in 1 (0.6%) patient in the baricitinib 2 mg group. Infection leading to treatment interruption was reported in 1.1% to 5.2% of patients across the trials.

In Study BRAVE-AA1, myocardial infarction and coronary revascularization was reported in 1 (0.5%) patient in the baricitinib 2 mg group. Serious arrhythmia was reported in 1 (0.5%) patient in the baricitinib 4 mg group. There was no report of cardiovascular events in Study BRAVE-AA2. There were no reports of venous or pulmonary thromboembolic events, gastrointestinal perforations, or nonmelanoma skin cancers in either trial. One patient in each of the placebo group (0.6% of patients) and the baricitinib 4 mg group (0.4% of patients) reported other forms of malignancies.

Critical Appraisal

The trials used adequate methods of randomization and allocation concealment. There were a few small baseline imbalances in patient characteristics that may be compatible with chance and were not believed to have a substantial impact on study results. The trials were adequately blinded; however, there was a potential for bias in the measurement of subjective outcomes (i.e., ClinRO measures, HADS, and Skindex-16 for AA). This could have led to the inflated efficacy of baricitinib based on the inferred judgment by patients and investigators regarding treatment assignment premised on response to treatment, without being unblinded. SALT₅₀ responder analysis_, HADS, and Skindex-16 for AA outcomes were not adjusted for multiplicity, so statistically significant results were at an increased risk of type I error (false-positive results). Between 31% and 42% of patients were excluded from ClinRO measures–based outcomes because of not having the specified baseline score, which could have impacted randomization, although the extent and direction of the resulting bias was unclear. There was a risk of attrition bias in favour of baricitinib with respect to change from baseline in HADS and Skindex-16 for AA domain scores, given the differential discontinuation rate between the baricitinib and placebo groups (there was a higher proportion of dropouts in the placebo group) and the use of last observation carried forward (LOCF) or modified last observation carried forward (mLOCF) as the data imputation method. There was a lack of sample size consideration and control for multiplicity for subgroup analyses, which precluded definitive conclusions on subgroup effects. Evidence for the validity and minimal important difference (MID) estimate of HADS and Skindex-16 for AA outcomes in patients with AA was not identified by the sponsor.

The clinical experts we consulted noted that the inclusion and exclusion criteria of the trials in general were reflective of the patient population eligible for baricitinib treatment in Canada, although patients with a primarily diffuse type of AA would not necessarily be excluded from treatment in clinical practice. As well, older adults (i.e., males older than 60 years and females older than 70 years) were excluded from the trials. There are differing opinions from the clinical experts suggesting that older adults may or may not be eligible for baricitinib treatment in clinical practice. In addition, the clinical experts noted that compared to clinical practice, the trials appeared to have enrolled a higher proportion of patients with very severe AA. As well, the trial populations had a lower degree of anxiety and depression at baseline as per clinical expert input, which could have impacted the generalizability of HADS outcomes. The clinical experts noted that a longer duration of follow-up beyond 36 weeks is required to adequately capture the long-term safety of baricitinib, including potential rare AEs, since baricitinib is expected to be a lifelong treatment for many patients. No head-to-head evidence comparing baricitinib with systemic treatments for severe AA that are currently reimbursed by the public drug plans (conventional immunosuppressants) was submitted. As well, the absence of evidence for baricitinib in older adults (males older than 60 years and females older than 70 years — categories that were excluded from the trials), represents another gap in evidence.

GRADE Summary of Findings and Certainty of the Evidence

For pivotal studies and randomized controlled trials (RCTs) identified in the sponsor’s systematic review, the Grading of Recommendations Assessment, Development and Evaluation (GRADE) tool was used to assess the certainty of the evidence for outcomes considered most relevant to inform our expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group. Following the GRADE approach, evidence from RCTs started as high-certainty evidence and could be rated down for concerns related to study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, the imprecision of effects, and publication bias.

The selection of outcomes for the GRADE assessment was based on the sponsor’s Summary of Clinical Evidence,⁹ consultation with clinical experts, and input received from a patient group and public drug plans. The following list of outcomes was finalized in consultation with expert committee members:

• scalp hair regrowth (the proportion of patients with a SALT score of 20 or less and a SALT₅₀ response)

• EB and EL hair regrowth (the proportion of patients attaining an EB [or EL] score of 0 or 1 with ≥ 2-point improvement from baseline, among patients with a baseline score of ≥ 2)

• anxiety and depression (the change from baseline in HADS anxiety and depression domain scores)

• HRQoL (the change from baseline in Skindex-16 for AA symptoms, emotions, and functioning domain scores)

• harms (SAEs).

The GRADE summary of findings for baricitinib versus placebo for the treatment of adults with severe or very severe AA is presented in Table 2 (baricitinib 2 mg versus placebo) and Table 3 (baricitinib 4 mg versus placebo).

Table 2: Summary of Findings for Baricitinib 2 mg Versus Placebo for Adults With Severe or Very Severe AA

AA = alopecia areata; CDA-AMC = Canada’s Drug Agency; CI = confidence interval; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; HADS = Hospital Anxiety and Depression Scale; HRQoL = health-related quality of life; NR = not reported; RCT = randomized controlled trial; SALT = Severity of Alopecia Tool; SALT₅₀ = at least a 50% improvement from baseline in the Severity of Alopecia Tool score; SE = standard error; Skindex-16 for AA = Skindex-16 for Alopecia Areata.

Notes: Details included in Table 2 are from the sponsor’s Summary of Clinical Evidence.⁹

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

^aRated down 1 level for serious imprecision. The clinical experts consulted by CDA-AMC indicated that a difference of 100 patients per 1,000 patients could be considered clinically important. The 95% CI included the possibility of benefit and no difference in both trials.

^bStatistical testing for this outcome was not adjusted for multiplicity. The results are considered as supportive evidence.

^cDid not rate down for imprecision. Although the lower boundary of the 95% CI in Study BRAVE-AA1 was 95 more patients per 1,000 patients, this was not considered to be a source of serious imprecision because of the lower boundary’s proximity to the threshold of 100 more patients per 1,000 patients as per clinical expert input.

^dRated down 1 level for serious study limitations. Randomization could potentially be impacted because of the exclusion of patients whose baseline score did not meet the specified value of at least 2, from each treatment group. The extent and direction of the resulting bias was unclear.

^eRated down 1 level for serious imprecision. The clinical experts we consulted indicated that a difference of 100 patients per 1,000 patients could be considered clinically important. In both trials, the 95% CI included the possibility of benefit and little to no difference. This was not rated down for inconsistency, though the point estimates from the trials were in different directions based on the threshold of 100 patients per 1,000 patients as per clinical expert input. This is because of overlap in the 95% CIs in the trials, including the possibility of benefit and little to no difference for both.

^fNo formal statistical testing was conducted because of a prior failure of an outcome in the statistical hierarchy. The results are considered as supportive evidence.

^gRated down 2 levels for very serious study limitations. Study treatment discontinuation was notably higher in the placebo group compared with the baricitinib 2 mg group in both trials. The differential discontinuation rate, along with the use of modified LOCF or LOCF as the data imputation method, could potentially lead to attrition bias in favour of the baricitinib 2 mg group. In addition, evidence for the validity of this outcome measure in the patient population under review (i.e., patients with AA) were not identified by the sponsor.

^hRated down 1 level for serious indirectness. The trial population had a higher mean baseline score (less severe anxiety or depression) than patients in clinical practice as per clinical expert input.

ⁱDid not rate down for imprecision using null as a threshold. Although the upper boundary of the 95% CI in Study BRAVE-AA2 was 0.88, this was not considered to be a source of serious imprecision because of the upper boundary’s proximity to the null.

^jThere were no concerns with imprecision using the null as a threshold. Although the upper boundary of the 95% CI was 0.80 and 0.88 in Study BRAVE-AA1 and Study BRAVE-AA2, respectively, this was not considered to be a source of serious imprecision because of the upper boundary’s proximity to the null.

^kRated down 1 level for serious indirectness. The duration of follow-up of 36 weeks is inadequate for capturing potential rare serious adverse events of baricitinib as per clinical expert input. We were also rated this down 1 level for serious imprecision since the results were based on a small number of events across the trials.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Reports.^7,8

Table 3: Summary of Findings for Baricitinib 4 mg Versus Placebo for Adults With Severe or Very Severe AA

AA = alopecia areata; CI = confidence interval; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; HADS = Hospital Anxiety and Depression Scale; HRQoL = health-related quality of life; NR = not reported; RCT = randomized controlled trial; SALT = Severity of Alopecia Tool; SALT₅₀ = at least a 50% improvement from baseline in the Severity of Alopecia Tool score; SE = standard error; Skindex-16 for AA = Skindex-16 for Alopecia Areata.

Notes: Details included in Table 3 are from the sponsor’s Summary of Clinical Evidence.⁹

Study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, the 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.

^aStatistical testing for this outcome was not adjusted for multiplicity. The results are considered as supportive evidence.

^bRated down 1 level for serious study limitations. Randomization could potentially be impacted because of the exclusion of a large proportion of patients from each treatment group whose baseline score did not meet the specified value of at least 2. The extent and direction of the resulting bias was unclear.

^cRated down 2 levels for serious study limitations. Study treatment discontinuation was notably higher in the placebo group compared with the baricitinib 4 mg group in both trials. The differential discontinuation rate, along with the use of modified LOCF or LOCF as the data imputation method, could potentially lead to attrition bias in favour of the baricitinib 4 mg group. In addition, evidence for the validity of this outcome measure in the patient population under review (i.e., patients with AA) were not identified by the sponsor.

^dRated down 1 level for serious indirectness. The trial population had a higher mean baseline score (less severe anxiety or depression) than patients in clinical practice as per clinical expert input.

^eRated down 1 level for serious indirectness. The duration of follow-up of 36 weeks is inadequate for capturing potential rare serious adverse events of baricitinib as per clinical expert input. We also rated this down 1 level for serious imprecision since the results were based on a small number of events across the trials.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Long-Term Extension Studies

Description of Studies

Study BRAVE-AA1

This is a long-term extension study (week 36 onward) of Study BRAVE-AA1.¹⁰ The purpose of this study is to provide the safety and efficacy analyses through week 104 to support dosing recommendations in the product labelling of baricitinib.

At week 52, patients initially randomized to baricitinib who were responders (with a SALT score of 20 or less) were rerandomized at a 3:1 ratio to stay on their current dose of baricitinib or to transition to placebo (randomized withdrawal). Responders who had been rerandomized to placebo and had experienced a loss of treatment benefit at any time after week 52 (more than a 20-point worsening in the SALT score from week 52) were re-treated with their original baricitinib dose and the efficacy of re-treatment was analyzed as part of the other secondary end points of Study BRAVE-AA1.

This extension study included week 0 to week 52 and week 52 to week 76 efficacy and safety data for patients who up-titrated at week 52. The up-titration cohort included ███ patients randomized to baricitinib 2 mg at week 0 who did not attain a SALT score of 20 or less at week 52. All ███ patients were titrated up to baricitinib 4 mg.

Study BRAVE-AA2

This is a long-term extension study (week 36 onward) of Study BRAVE-AA2.¹¹ The purpose of this study is to provide efficacy and safety analyses to support dosing recommendations in product labelling.

At week 52, patients were divided into 2 cohorts. The randomized down-titration cohort included 82 patients who were randomized at week 0 to baricitinib 4 mg, having attained a SALT score of 20 or less at week 52. Of these patients, 42 were randomly assigned to remain on baricitinib 4 mg and 40 patients were randomly assigned to down-titrate to baricitinib 2 mg. The up-titration cohort included ██ patients randomized to baricitinib 2 mg at week 0 who did not attain a SALT score of 20 or less at week 52. All ██ patients were titrated up to baricitinib 4 mg.

Efficacy Results

Proportion of Patients Attaining SALT Score of 20 or Less

In both trials, the proportion of patients attaining a SALT score of 20 or less continuously increased over the treatment period beyond 36 weeks for the baricitinib 4 mg cohort. At week 52, 40.9% and 21.2% of patients receiving baricitinib 4 mg and baricitinib 2 mg, respectively, attained a SALT score of 20 or less in Study BRAVE-AA1. Similarly, 36.8% and 24.4% of patients receiving baricitinib 4 mg and baricitinib 2 mg, respectively, attained a SALT score of 20 or less at week 52 in Study BRAVE-AA2.

Study BRAVE-AA1: Up-Titration Cohort

At week 52, ███ patients who were originally randomized to the baricitinib 2 mg group were considered nonresponders and were eligible for inclusion in the up-titration cohort titrated up to baricitinib 4 mg. At week 76, following 24 weeks of treatment on baricitinib 4 mg, ██ of ███ patients (████ ███ ███ █████ ██ █████) in the up-titration cohort attained a SALT score of 20 or less.

Study BRAVE-AA2: Randomized Down-Titration Cohort

At week 52, 82 patients who were originally randomized to the baricitinib 4 mg group were eligible for randomized down-titration to baricitinib 2 mg. At week 52, █████ ███ ██ ███ ███ ███ █████ ██████ of patients attained a SALT score of 20 or less. ███ ███████ ███ █ ███████ █████ ██ ████ ███ ███ ████████ ████ █████████ █████ ████ ██████ ███ █████████ ████ ███ ████████ ████ ██ ████ ██ ██ ███ █████

Among patients receiving baricitinib 4 mg who attained a SALT score of 20 or less at week 52, this response was retained up to week 76 in 75% (30 of 40, ███ ███ █████ ██ █████) of patients who were down-titrated to baricitinib 2 mg, and 98% (41 of 42, ███ ███ █████ ██ █████) of patients who remained on baricitinib 4 mg.

Study BRAVE-AA2: Up-Titration Cohort

At week 52, ██ patients who were originally randomized to the baricitinib 2 mg group were considered nonresponders and were eligible for inclusion in the up-titration cohort that was titrated up to baricitinib 4 mg. At week 76, after 24 weeks of titration up treatment on baricitinib 4 mg, ██ of ██ patients (██████ ███ ██ █████ ██ █████) attained a SALT score of 20 or less.

ClinRO Measures for EB and EL Hair Loss

At week 52, 39.4% and 27.9% of patients receiving baricitinib 4 mg and baricitinib 2 mg, respectively, attained a ClinRO measure for EB hair loss (0 or 1) (with ≥ 2-point improvement from baseline through week 52 among patients with a score of ≥ 2 at baseline) in Study BRAVE-AA1. Similarly, 49.7% and 16.3% of patients receiving baricitinib 4 mg and baricitinib 2 mg, respectively, attained a ClinRO measure for EB hair loss (0 or 1) (with ≥ 2-point improvement from baseline through week 52 among patients with a score of ≥ 2 at baseline) at week 52 in Study BRAVE-AA2.

At week 52, 40.7% and 21.6% of patients receiving baricitinib 4 mg and baricitinib 2 mg, respectively, attained a ClinRO measure for EL hair loss (0 or 1) (with ≥ 2-point improvement from baseline through week 52 among patients with a score of ≥ 2 at baseline) in Study BRAVE-AA1. Similarly, 50.7% and 30.3% of patients receiving baricitinib 4 mg and baricitinib 2 mg, respectively, attained a ClinRO measure for EL hair loss (0 or 1) (with ≥ 2-point improvement from baseline through week 52 among patients with a score of ≥ 2 at baseline) at week 52 in Study BRAVE-AA2.

Harms Results

Study BRAVE-AA1: Up-Titration Cohort

TEAEs were reported for ████ █ ███ ██ ████ of patients who up-titrated to baricitinib 4 mg up to week 76. Most events were mild in severity. █████ ████ █ ████████ ███ ████████ ████ ████ ████ ██████ ██████ ███ ███████ discontinued the study drug and study because of ██████ ████. There were no deaths in the up-titration cohort.

Study BRAVE-AA2: Randomized Down-Titration Cohort

For both treatment groups, most events were mild or moderate in severity. Up to week 76, of the patients who remained on baricitinib 4 mg, | ████████ ████████ ████ ███ ███ █ ███████ ████ █████ ███ █ ██████ ████. ██ ███████ ██████ ███ ████ ████████ for patients who down-titrated to baricitinib 2 mg. ██████████ ████████████ █████ ████ ██ ███ █████ ███ ██ ███ ██ ███ ████ ██████ There were no deaths in the randomized down-titration cohort.

Study BRAVE-AA2: Up-Titration Cohort

TEAEs were reported for █████ ███ ██ ███ of patients who titrated up to baricitinib 4 mg up to week 76. Most events were mild or moderate in severity; | ███████ ████████ █ ██████ █████ ██ ████████ ████████ ████ ██ ████████████ █████ ████ ██ ███ █████ ███ ██ ████ There were no deaths in the up-titration cohort.

Critical Appraisal

Both the BRAVE-AA1 and BRAVE-AA2 extension studies were limited by their noncomparative design. At time points after 36 weeks, there remained no randomized comparison to placebo, challenging causal interpretations. Although the patients and investigators remained blinded to the assigned interventions, there is still the possibility that patients may be able to infer treatment assignment because of differences in efficacy (relative to placebo during the double-blind treatment phase). As such, there may be a risk of bias in the reporting of efficacy outcomes that required some level of subjective judgment by the evaluators (e.g., ClinRO) and harms outcomes, although the extent and direction of bias cannot be predicted. It is unlikely that bias would be introduced for the SALT response, since it is measured objectively. Finally, missing information such as pooling strategies constrained a robust critical appraisal; hence, a firm conclusion cannot be drawn on the long-term efficacy and safety. Since both Study BRAVE-AA1 and Study BRAVE-AA2 included rollover patients consistent with their characteristics at entry in the core study, it is reasonable to expect similar limitations to the generalizability of study results are relevant to the long-term extension phase. Further, some outcomes that are important to patients (e.g., HRQoL, anxiety, depression) could not be evaluated against a placebo control beyond the 36-week double-blind treatment phase because of discontinuation of the placebo in nonresponders. As such, there is limited evidence for the effect of baricitinib 2 mg or baricitinib 4 mg on these outcomes for time points after 36 weeks (including for patients who up-titrated or down-titrated). Despite longer follow-up for harms, some rare harms (e.g., malignancies) may still not be fully captured.

Indirect Comparisons

No indirect comparative evidence was submitted by the sponsor. The sponsor noted that before the regulatory approval of baricitinib for severe AA in Canada, the standard of care included off-label therapies and nonpharmacological options. The sponsor further noted that the pivotal trials of baricitinib were placebo-controlled and given that no approved comparator drugs were available at the time of the phase III clinical development conduct, there is no indirect comparative efficacy evidence to present in this section.

Studies Addressing Gaps in the Evidence From the Systematic Review

Description of Studies

Additional insights into the effects of baricitinib in patients with AA were sought for males older than 60 years and females older than 70 years that were not included in the pivotal trials. A retrospective chart study (n = 14) by Tang et al. (2024)¹² describing baricitinib treatment in patients older than 65 years was included. A retrospective chart review of 36 patients conducted by Moreno-Vilchez et al. (2024)¹³ and a retrospective chart review of 95 patients in Japan by Numata et al. (2024)¹⁴ provided additional data about the effects of baricitinib.

Efficacy Results

Tang et al. (2024)

After a mean (SD) duration of 18.5 (11.9) months, a 72.0% reduction in the mean SALT score from baseline was observed. Moreover, 11 of 14 (78.6%) patients attained a SALT score of less than 10 after a mean duration of 18.6 months where SD is not reported.

Numata et al. (2024)

The percentage of patients in the entire cohort who attained a SALT score of 20 or less at week 12, week 24, and week 36 was 6.4% (6 of 94) of patients, 35.4% (28 of 79) of patients, and 46.7% (21 of 45) of patients, respectively.

The complete response rate (a SALT score of 0) at week 24 and week 36 was 1.3% (1 of 79) of patients and 6.7% (3 of 45) of patients, respectively.

Moreno-Vilchez et al. (2024)

In this study, 58.8% of patients attained a SALT score of 20 or less at week 24. The response continued for 52 weeks, with 66.6% of patients classified as responders. Additionally, the study compared the SALT scores between patients treated with monotherapy and those who received adjuvant treatment.

Harms Results

Tang et al. (2024)

Adverse effects of baricitinib were moderate and included the reactivation of herpes zoster (n = 1), elevated creatine kinase (n = 1), and grade 2 neutropenia (n = 1). Only 1 patient required a reduction in the dose of baricitinib because of grade 2 neutropenia. No cases of venous thromboembolism, MACE, or malignancy were reported.

Numata et al. (2024)

Infectious complications occurred in 6 patients during the initial 12 weeks. Herpes simplex and COVID-19 (severe acute respiratory syndrome coronavirus 2) occurred in 1 patient and 5 patients, respectively. No other severe complications occurred during the entire 36-week course.

Moreno-Vilchez et al. (2024)

Overall, AEs were mild. Three patients were discontinued because of inadequate treatment response: 2 patients at week 52 and 1 patient at week 76. Additionally, 1 patient had temporary lymphopenia with methotrexate treatment.

Critical Appraisal

Limitations of the 3 studies included their retrospective designs and small sample sizes. Moreover, most patients were treated with concomitant treatments, and without a randomized comparison group, it is not possible to attribute the observed effects to baricitinib with certainty. Furthermore, information such as treatment exposure and concomitant treatments in Numata et al. were not reported. Both Tang et al. and Numata et al. included patients with moderate-to-severe AA; however, patients with moderate AA would not be candidates for baricitinib treatment in Canada. The results of these studies may not be generalizable to patients with severe or very severe AA, which may be more difficult to treat compared with moderate AA. The study by Numata et al. included patients exclusively from Japan whereas the study by Moreno-Vilchez et al. included patients exclusively from Spain (in 2 centres). It is uncertain whether results from small samples of patients treated in these countries would be generalizable to patients living in Canada, given the potential for differences in standard of care in these countries.

Conclusions

Direct comparative evidence from 2 pivotal double-blind RCTs (Study BRAVE-AA1 and Study BRAVE-AA2) demonstrated that 36 weeks of baricitinib 4 mg treatment resulted in the clinically important regrowth of scalp hair compared with placebo in adults with severe or very severe AA. Benefits of the regrowth of EB and EL hair were shown but were less certain because of study limitations. Results also favoured baricitinib 2 mg treatment with respect to the regrowth of scalp and EB hair, although there was some uncertainty on whether the magnitude of change was clinically important because of imprecision. No definitive conclusion can be drawn regarding the direction and magnitude of the effects of baricitinib treatment on anxiety, depression, and HRQoL because of important methodological limitations (potential attrition bias and a lack of evidence supporting the validity of the instruments used in patients with AA) and concerns with indirectness (for anxiety and depression outcomes). The benefits of baricitinib in hair regrowth appeared to be sustained through week 104 in the trials, although analyses beyond week 36 were noncomparative, which precluded firm conclusions. No conclusions can be drawn regarding the clinical benefits of baricitinib in older adults with severe AA from a sponsor-submitted retrospective chart review study (Tang et al. [2024]) because of a small sample size and single-arm study design. No notable concerns with the safety profile of baricitinib were identified based on results from the pivotal trials through week 104. No direct or indirect comparative evidence for baricitinib versus systemic treatments currently reimbursed by the public drug plans (immunosuppressants) were submitted.

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 baricitinib 2 mg and baricitinib 4 mg oral tablets in the treatment of adult patients with severe AA.

Disease Background

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

AA is a chronic autoimmune disease characterized by nonscarring hair loss at the scalp as well as the loss of EB, EL, beard, pubic, or axillary hair. AA is a relapsing and remitting condition where spontaneous hair regrowth has been reported at the beginning of the disease (mild disease) progressing to extensive, chronic hair loss that does not remit without treatment.¹ Up to 50% of patients will recover within 1 year even without treatment, but as many as 85% of individuals experience multiple episodes of hair loss.¹⁵ Progression from patchy AA to more severe forms (i.e., total hair loss at the scalp or body) occurs in 14% to 25% of patients, with full recovery in less than 10% of patients.^1,16,17

The onset of hair loss in AA is typically rapid and the progression is unpredictable.¹ In a systematic review of the epidemiology and burden of AA, the mean age of onset was found to range between 25.2 years and 36.3 years across studies. A total of 83% to 88% of individuals experienced disease onset by aged 40 years and 40.2% of individuals by aged 20 years.¹⁸

Involvement of other body areas (aside from the scalp) may occur as AA progresses or may occur as an isolated incident.¹⁹ AA sometimes manifests as a partial or complete loss of EB, EL, beard, pubic, or axillary hair, all of which can present unique challenges to patients.^20,21 The loss of EBs or ELs or both may result in a dramatically altered appearance and eye irritation because of the loss of the physical barrier that helps prevent sweat and foreign particles from entering the eyes.²² Other ophthalmological issues secondary to EB or EL loss include ocular surface inflammation, dryness, and blepharitis. Since ELs and EBs also serve to define nonverbal communication and self-expression, their loss may have both a functional and psychological impact on patients’ quality of life.^21,23 The skin in AA lesions does not typically show signs of inflammation or symptoms, although slight redness in the affected area or pruritus during periods of active hair loss have been reported.^24,25

While the etiology of AA is unknown, it has been suggested that a triggering factor leads to the collapse of hair follicle immune privilege (IP), where functioning IP protects hair follicles from unwanted immune responses. In genetically predisposed individuals, the loss of IP results in characteristic hair loss.^1,19 Since hair follicles are not destroyed by the disease process, the potential for hair regrowth is in theory retained in patients with AA.¹

At present, there are no Canadian epidemiology data. When considering lifetime prevalence, approximately 2% (based on data from 1990 to 2009) of the general population will develop AA at some point in their lives, affecting men and women of all ages and races.^1,4,26,27 Data on the prevalence of AA by severity are very limited. The cross-sectional study by Benigno et al. (2020) in a representative sample of the US population showed that the overall prevalence of AA was 0.21%, with 0.12% for mild disease (a SALT score of 50 or less), 0.09% for severe disease (a SALT score greater than 50), and 0.04% for total or near-total scalp or all-body hair loss subgroup in moderate-to-severe patients.² From this data, it was estimated that among patients living with AA in Canada, 43% are living with severe disease.

AA is diagnosed based on presenting features and once other causes of hair loss have been excluded.²⁵ Typical clinical features of AA that support a diagnosis include patches of hair loss and the presence of so-called “exclamation mark” hairs that are short and broken, and taper proximally.¹⁹ In addition, the pull test (the examiner grasping approximately 40 hairs to 60 hairs between their thumb, index, and middle fingers, and gently pulling them away from the scalp) can be used to diagnose hair loss. A positive result is attained if greater than 10% of the hairs are pulled out, indicating hair shedding. It is, however, hard to standardize this test, and a negative result does not always rule out an AA diagnosis.²⁸ While further testing is often not required beyond careful evaluation of the patients’ clinical history and physical examination, additional investigations such as a dermoscopy or histopathology may sometimes be used to confirm the diagnosis.^1,29 It has been suggested that standardized and objective measures be used to describe disease severity.³⁰ However, there is no unique definition of disease severity for AA and the driver for the definition of AA severity is the extent of scalp hair loss.³¹ The main prognostic factor in patients with AA appears to be the extent of hair loss, especially at presentation.^19,31 There is a tendency for severe patterns of AA to worsen over time.^19,31 Additionally, an earlier age of disease onset corresponds to an increased lifetime risk of extensive disease (i.e., onset in the first 2 decades is often associated with severe AA).^1,18 The prognosis of patients with severe AA is poor, even on therapy.^25,31-34

Standards of Therapy

Canadian clinical practice guidelines for AA are not available to date. As per input from the clinical experts we consulted, in Canada, clinicians consider systemic drugs for the treatment of adults with severe AA. These include conventional immunosuppressants (cyclosporine, methotrexate, azathioprine, and mycophenolate mofetil) and JAK inhibitors. Ritlecitinib is a JAK inhibitor recently approved by Health Canada for the treatment of adults and adolescents aged 12 years and older with severe AA. Tofacitinib, upadacitinib, and abrocitinib are off-label JAK inhibitor treatments for severe AA; upadacitinib and abrocitinib are approved for the treatment of AD and may be used in patients with concurrent atopic dermatitis, as per clinical expert input. Conventional immunosuppressants are off-label treatments that are currently reimbursed by the public drug plans for the treatment of severe AA while ritlecitinib, tofacitinib, upadacitinib, and abrocitinib are currently not. The clinical experts noted that systemic treatments target the underlying immune mechanisms of AA, although they do not cure the condition and are associated with a high risk of relapse upon treatment discontinuation or dose reduction. In addition, oral corticosteroids (prednisone) may be used for short-term treatment in patients with AA that is rapidly progressing.

The clinical experts noted that, before the initiation of systemic therapy, patients with severe AA would typically have already received off-label topical medications (potent corticosteroids, minoxidil, a calcineurin inhibitor for EBs and beard specifically, and prostaglandin analogue for ELs specifically) and/or intralesional corticosteroids, and would continue to do so while on systemic therapy. While topical medications and intralesional corticosteroids are effective treatments in some patients with mild or localized AA, they are not effective in patients with severe AA when used alone and are reserved as an adjunct treatment to systemic drugs. Other treatments such as topical anthralin and contact immunotherapy with diphenylcyclopropenone are no longer commonly used in adults with severe AA in clinical practice because of low efficacy, difficult application, and the need for long-term treatment, according to the clinical experts. Phototherapy is not part of standard of care in Canada, as per clinical expert input. Nonpharmacological intervention with camouflage techniques (e.g., wig, hairpiece, EB tattooing) are commonly used to help patients cope with the impact of AA on altered body appearance.

According to the clinical experts, the treatment goals for severe AA are to attain a cosmetically meaningful regrowth of hair, improve HRQoL, avoid serious AEs, and if possible, reduce the risk for relapse with dose reduction or discontinuation of the systemic drug.

Drug Under Review

Key characteristics of baricitinib are summarized in Table 4 with other treatments available for severe AA in adult patients.

Baricitinib is available as 2 mg and 4 mg oral tablets and is approved by Health Canada for the treatment of adult patients with severe AA (the Notice of Compliance date was January 26, 2024). Baricitinib is a selective and reversible inhibitor of Janus kinase 1 and Janus kinase 2. JAKs are enzymes that transduce intracellular signals from a cell surface receptor for a number of growth factors involved in hematopoiesis, inflammation, and immune function. Within the intracellular signalling pathway, JAKs phosphorylate and activate signal transducers and activators of transcription, which modulate intracellular activity, including gene expression. Baricitinib modulates these signalling pathways, thereby reducing the phosphorylation and activation of signal transducers and activators of transcription thought to be involved in the pathogenesis of AA.³⁵ The sponsor’s reimbursement request aligns with the Health Canada–approved indication. Baricitinib is also indicated for the treatment of adults with rheumatoid arthritis. Baricitinib has not been previously reviewed by CDA-AMC for the treatment of AA. However, baricitinib was previously reviewed by us for the treatment of rheumatoid arthritis and received a recommendation to reimburse with conditions on August 2, 2019. The recommended dose is 2 mg once daily and, if the response to treatment is not adequate, the dose may be increased to 4 mg once daily. For patients with nearly complete or complete scalp hair loss, and/or substantial EL or EB hair loss, a starting dose of 4 mg once daily may be considered. Once a patient attains an adequate response to treatment with 4 mg, a decrease in dose to 2 mg once daily may be considered. When clinically advisable, the lowest effective dose should be used to minimize AEs. The discontinuation of therapy should be considered if patients do not show evidence of therapeutic benefit after 36 weeks of treatment. Dose adjustment for baricitinib in patients with renal impairment is recommended. A combined use with other JAK inhibitors, biologic immunomodulators, or potent immunosuppressants such as azathioprine and cyclosporine is not recommended.³⁵

Table 4: Key Characteristics of Baricitinib, Minoxidil, Corticosteroids, and Immunosuppressants

b.i.d. = twice a day; CVD = cardiovascular disease; DHFR = dihydrofolate reductase; eGFR = estimated glomerular filtration rate; IMPDH = inosine monophosphate dehydrogenase; JAK = Janus kinase; MACE = major adverse cardiac event; NA = not applicable.

^aThis is a Health Canada–approved indication that is relevant to the reimbursement request.

Sources: Sponsor’s Summary of Clinical Evidence, and Health Canada–approved product monographs for baricitinib, minoxidil, and triamcinolone acetonide.^9,35-37

Perspectives of Patients, Clinicians, and Drug Programs

Patient Group Input

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

We received 1 patient group submission from CANAAF. CANAAF was registered as a charitable organization in 2010 and is described as the voice for all patients and families affected by AA living in Canada. CANAAF collected data on the psychosocial and emotional impact of AA from peer-reviewed literature, as well as patient perspectives on AA from patient reports and support sessions.

CANAAF commented that AA is incredibly burdensome on a patient’s mental health and quality of life, and the disease causes disfiguring hair loss that occurs unexpectedly and can progress rapidly. Based on a patient report, CANAAF further stated that the anxiety, depression, and other resultant psychological conditions are not minor in nature; therefore, the loss of hair can create layers of stigma and misunderstandings. Short hair or baldness may be associated with a preference for an “edgy” look or having a certain sexuality, which may not be accurate. Those with this disease may feel less feminine or less masculine without hair. Children and teenagers may experience bullying. In addition, CANAAF revealed that there is a significant financial burden associated with AA; this was supported by the findings of a CANAAF community alopecia patient focus group conducted in 2023. The most significant cost item was wig purchase and maintenance, which can cost more than $2,500 a year. Some patients experienced significant impacts on their ability to work.

Based on the literature, CANAAF identified limitations of the currently available treatments for AA, including topical corticosteroids (limited effectiveness, only effective for patients with very limited AA, difficult product application, and scalp irritation), intralesional corticosteroids (painful injections and limited drug coverage by drug plans), oral corticosteroids (variable success rates, high relapse rate, limited drug coverage, and unfavourable side effects), topical minoxidil (nondurable benefits for very mild AA, and AEs such as excessive hair growth on body parts other than the site of application, irritation, and allergic contact dermatitis), oral minoxidil (systemic AEs relating to its antihypertensive property and limited drug coverage), and systemic immunosuppressants (variable effectiveness; a risk of organ toxicity, infection, and malignancy; concomitant administration of oral corticosteroids required for some drugs; and limited drug coverage).

CANAAF identified a need for an effective treatment option that could result in full and sustained hair growth and alleviate anxiety and depression associated with AA. CANAAF believed that baricitinib may fulfill this need by serving as an effective treatment that has a favourable side effect profile and is easy to administer. The group noted that most patients regrew all their hair with baricitinib treatment. CANAAF also noted that the side effect profile of baricitinib is much more favourable compared to existing treatments. Baricitinib is a much easier treatment option for patients as it only requires that they take 1 pill once a day. This is in comparison to other treatments that must be applied topically, injected (often by a health care professional), or taken orally more than once a day.

Clinician Input

Input From Clinical Experts Consulted by CDA-AMC

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

Unmet Needs

According to the clinical experts we consulted, currently available systemic therapies for severe AA are associated with poor efficacy and, thus, fail to attain cosmetically meaningful hair regrowth and improvement in HRQoL in patients with severe AA. As well, the clinical experts noted that the existing systemic treatments are associated with a risk of relapse with dose reduction and/or discontinuation, as well as AEs. In particular, SAEs from the long-term use of conventional systemic immunosuppressants is a notable concern. Also, patient access to emerging therapies, such as ritlecitinib, is currently limited, as per clinical expert input.

Place in Therapy

The clinical experts noted that JAK inhibitors, such as baricitinib, bring a novel approach and mechanism of action to the treatment of severe AA; although similar to other systemic immunomodulators, baricitinib is not a curative treatment. The clinical experts noted that because of the limited efficacy of older systemic immunomodulators (e.g., methotrexate, cyclosporine), it is rational to use baricitinib (and JAK inhibitors in general) as a first-line systemic therapy in severe AA, rather than as the last line of treatment after the failure of older systemic immunomodulators. This is based on their clinical experience with the effectiveness of JAK inhibitors relative to conventional immunosuppressants and the paucity of published data on immunosuppressants in patients with severe AA. There may be exceptions where an immunosuppressant is chosen ahead of a JAK inhibitor — for example, in patients with comorbid conditions such as psoriasis — but in the clinical experts’ opinion, most patients should receive baricitinib as a first-line systemic treatment.

The clinical experts noted that baricitinib is not the first JAK inhibitor used by clinicians in Canada for the treatment of severe AA. Tofacitinib is used off-label and ritlecitinib is another treatment for severe AA that has been recently approved by Health Canada. At present, the treatment choice would be determined based on patient access to and reimbursement for a JAK inhibitor rather than on patient-specific clinical characteristics, as per clinical expert input. The clinical experts anticipated that over time, as the reimbursement environment evolves and as comparative efficacy and safety data accumulate, the decision would become a clinical one.

The clinical experts noted that it would be appropriate to use baricitinib in combination with topical treatments and/or intralesional corticosteroids but not in combination with other immunomodulators, except for prednisone where concomitant use with baricitinib may be appropriate.

Patient Population

The clinical experts noted that at present, disease severity is evaluated based on a global assessment and that no specific clinical tools are used; however, this may change as new treatments are adopted in clinical practice and SALT scores become a requirement for reimbursement. In the clinical experts’ opinion, patients who have severe AA with scalp involvement as reflected by a SALT score of 50 or more and who have a current episode of AA with a duration of greater than 1 year, but less than 10 years are potential candidates for baricitinib treatment. The clinical experts noted that the diagnosis of AA is readily made by a dermatologist, typically involving, for example, the use of the hair-pull test, trichoscopy, and/or biopsy; misdiagnosis is frequent among nondermatologists. They noted that there is no uniformly accepted definition of severe AA, although they considered a SALT score of 50 or more to be reflective of severe AA and noted that this definition has been used in clinical trials of other JAK inhibitors for the treatment of AA. They also noted that adhering to the inclusion criterion on the duration of a current episode that was used in the pivotal trials (i.e., more than 6 months and less than 8 years in duration) would be reasonable when selecting treatment candidates.

In terms of eligible age group, the clinical experts recognized that baricitinib is currently approved only in the adult population (aged 18 years or older), although they noted that the off-label use of baricitinib in the adolescent population (aged 12 years to 17 years) may also be reasonable from a clinical perspective. They anticipated that most clinicians would not prescribe baricitinib to adolescents until regulatory approval is granted in this population. The clinical experts also discussed the use of baricitinib in older adults (i.e., older than 60 years for males or older than 70 years for females, categories that were excluded from the pivotal trials). One clinical expert considered the use of baricitinib in this age group to be reasonable. The other clinical expert suggested restricting the use of baricitinib as per age restriction in the pivotal trials because of a lack of clinical trial data and unknown treatment benefits. This clinical expert anticipated that the older adults would not benefit from baricitinib treatment as much as younger patients since older adult patients tend to have other concurrent causes of hair loss that are not expected to be responsive to baricitinib treatment. Also, this clinical expert noted that AA has less impact on the HRQoL of older adults since camouflage management is typically more acceptable in this age group and most older adults have learned to adapt to AA.

Assessing the Response Treatment

The clinical experts noted that the outcome measures used in clinical trials are not aligned with clinical practice. The clinical experts noted that currently, a SALT score is typically not calculated in clinical practice; instead, a clinical judgment is made based on a global assessment by the clinician and the patient to determine whether cosmetically meaningful hair regrowth has occurred following treatment. The clinical experts noted that what constitutes cosmetically meaningful regrowth of hair varies from patient to patient. As well, in patients with rapidly progressing severe AA, the stabilization or cessation of further hair loss could be a clinically meaningful response to treatment, according to the clinical experts. For drug reimbursement, the clinical experts felt that it is reasonable to define meaningful response to treatment as attainment of a SALT score of 20 or less at week 36, consistent with the pivotal trials of baricitinib.

Discontinuing Treatment

The clinical experts noted that it would be reasonable to consider the discontinuation of baricitinib treatment in patients who fail to achieve cosmetically acceptable hair regrowth at 36 weeks, have further loss of hair at 36 weeks, experience severe AEs deemed to be related to the use of a JAK inhibitor (e.g., infection, hematologic abnormality, biochemical abnormality, hypersensitivity), or develop an intercurrent condition that makes the discontinuation of a JAK inhibitor advisable (e.g., malignancy).

Prescribing Considerations

Clinical experts noted that baricitinib treatment could be prescribed by dermatologists with experience in diagnosing, treating, and monitoring patients with severe AA. The clinical experts felt that it would not be appropriate for family physicians to prescribe baricitinib treatment at this time because of the risks of misdiagnosis and the inappropriate selection of patients for treatment, and the need for routine monitoring associated with JAK inhibitor use. The starting dose would likely be determined based on clinical judgment as per clinical expert input.

Clinician Group Input

We received no clinician group input for the drug under review.

Drug Program Input

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

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

AA = alopecia areata; CBC = complete blood count; CDA-AMC = Canada’s Drug Agency; CDEC = Canadian Drug Expert Committee; LOI = letter of intent; JAK = Janus kinase; SALT = Severity of Alopecia Tool; vs. = versus.

Clinical Evidence

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of baricitinib 2 mg and baricitinib 4 mg oral tablets in the treatment of adult patients with severe AA. The focus has been placed on comparing baricitinib to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of baricitinib is typically presented in 4 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 long-term extension studies. The third section typically includes indirect evidence from the sponsor; however, such evidence was not submitted for this review. The fourth section includes additional studies that were considered by the sponsor to address important gaps in the systematic review evidence.

Included Studies

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

• 2 pivotal RCTs identified in the sponsor-conducted systematic review (Study BRAVE-AA1 and Study BRAVE-AA2)⁶

• long-term extension results of Study BRAVE-AA1 and Study BRAVE-AA2³⁸

• 3 additional studies addressing gaps in evidence (Tang et al. [2024],¹² Numata et al. [2024],¹⁴ and Moreno-Vilchez et al. [2024]¹³).

Systematic Review

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

Description of Studies

Characteristics of the included studies are summarized in Table 6.

Two pivotal, multicentre, double-blind RCTs (Study BRAVE-AA1, N = 654; Study BRAVE-AA2, N = 546)⁶ aiming to assess the efficacy and safety of baricitinib relative to placebo in adult patients with severe or very severe AA were identified by the systematic review conducted by the sponsor. Study BRAVE-AA1 was an adaptive phase II and phase III trial; the phase II portion of Study BRAVE-AA1 was designed to determine the doses of baricitinib for use in the phase III portion of the study and in the phase III BRAVE-AA2 trial. Study BRAVE-AA1 was conducted in 70 study sites in North America and Asia. Study BRAVE-AA2 was conducted in 84 sites in North America, South America, Asia, and Australia. None of the trials included sites in Canada. The studies are ongoing; patient enrolment began in September 2018 in Study BRAVE-AA1 and in July 2019 in Study BRAVE-AA2. The analyses presented in this report are based on a database lock date of February 2, 2021, for Study BRAVE-AA1 and February 19, 2021, for Study BRAVE-AA2.

Study BRAVE-AA1

As shown in Figure 1, the phase III portion of the Study BRAVE-AA1 consisted of the following 5 periods.

• A 5-week screening period (day –3 to day –35): This was for study eligibility assessment.

• A 36-week double-blind, placebo-controlled treatment period (week 0 to week 36): Enrolled patients were randomized in a 2:2:3 ratio to receive either placebo, baricitinib 2 mg, or baricitinib 4 mg once daily for 36 weeks. Randomization was conducted in a blinded manner using an interactive web response system and was stratified by geographic region (North America, Asia, and the rest of the world) and duration of the current AA episode at baseline (less than 4 years versus at least 4 years). Patients, investigators, and study personnel were blinded to treatment assignment.

• A 68-week long-term extension period (week 36 to week 104): Double-blinding was maintained throughout this treatment period.

  At week 36:

  • Patients with a SALT score greater than 20 (hereinafter referred to as nonresponders as per the Clinical Study Report submitted by the sponsor) receiving placebo at week 36 were rescued to receive a baricitinib 4 mg or baricitinib 2 mg regimen by randomized assignment (1:1) for the remainder of the trial. Patients with a SALT score of 20 or less (hereinafter referred to as responders as per the Clinical Study Report submitted by the sponsor) receiving placebo continued the regimen with placebo for the remainder of the trial, regardless of whether relapse occurred later in the trial.

  • Baricitinib-treated patients continued treatment regardless of response at week 36 and were reassessed for response to treatment at week 52.

  At week 52:

  • Responders: Patients who had stayed on the same dose of baricitinib from initial randomization and were responders at week 52 underwent randomized withdrawal (randomization not stratified), where they were randomized in a 3:1 ratio to either stay on their existing dose of baricitinib or transition to placebo. Patients in the placebo group at week 52 who were responders remained on placebo. Responders who experienced a loss of treatment benefit after week 52 (defined as greater than a 20-point absolute worsening in the total SALT score) and:

    • who were randomized to placebo at week 52 (randomized withdrawal) were re-treated with their baricitinib dose, as randomized at baseline

    • who remained on baricitinib at week 52 continued to receive the same dose of baricitinib

    • who were randomized to placebo at baseline remained on placebo.

  • Nonresponders:

    • patients who had been in the baricitinib 4 mg group from baseline, had never attained a response to treatment by week 52, and did not have an improvement of 2 points or more from baseline in a ClinRO measure for EB or EL hair loss at week 52 were transitioned to placebo

    • patients who had been in the baricitinib 4 mg group had attained a response to treatment before week 52, and had lost response remained on the baricitinib 4 mg dose

    • those who had been in the baricitinib 2 mg group from baseline were rescued to receive baricitinib 4 mg

    • those who were rescued to baricitinib at week 36 continued in their current treatment group at week 52.

  At week 76:

  • Nonresponders at week 52 and week 76 were discontinued from the study unless they had at least a 2-point improvement from baseline in a ClinRO measure for EB or EL hair loss.

• A 96-week bridging extension period (week 104 to week 200): Patients who completed week 104 and did not meet criteria for permanent discontinuation could continue to receive the same treatment they received during the long-term extension period in a double-blinded manner.

• A 4-week post-treatment follow-up period (week 200 to week 204): A post-treatment follow-up visit was scheduled approximately 4 weeks after the bridging extension period.

Figure 1: Study BRAVE-AA1 Study Design

BARI = baricitinib; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; PBO = placebo; PTFU = post-treatment follow-up; QD = every day; SALT = Severity of Alopecia Tool; W = week; WO = washout.

^a Placebo responders stayed on placebo for the remainder of the trial, even if relapse was observed later.

^b Patients with a SALT score of 20 or less who stayed on the same dose of baricitinib from week 0 were randomized to stay on the existing baricitinib dose or transitioned to placebo.

^c Responders participating in randomized withdrawal who experienced more than a 20-point absolute worsening in their total SALT score after week 52 were re-treated with the baricitinib dose to which they were originally randomized if they were randomized to placebo at week 52, or continued to receive the same dose of baricitinib if they were randomized to remain on baricitinib at week 52.

^d Nonresponders at week 52 were rescued to baricitinib 4 mg if receiving baricitinib 2 mg from baseline or remained on baricitinib 4 mg if they were in the 4 mg group and attained a SALT score of 20 or less before week 52.

^e Never-responders (those who never attained a SALT score of ≤ 20 by week 52 despite being in the baricitinib 4 mg group from baseline and had not experienced a ≥ 2-point improvement from baseline in ClinRO measures for EB or EL hair loss) were automatically transitioned to placebo.

^f Nonresponders at week 52 and week 76 were automatically discontinued at week 76 unless they had an improvement of 2 points or more from baseline in ClinRO measures for EB or EL hair loss.

Source: Study BRAVE-AA1 Clinical Study Report.⁷

Study BRAVE-AA2

As shown in Figure 2, the study design of Study BRAVE-AA2 is nearly identical to that of Study BRAVE-AA1 (as previously described), except that the 68-week long-term extension period was conducted as described as follows. Note that the double-blinding of patients, investigators, and study personnel was maintained throughout this treatment period.

At week 36:

• Placebo-treated nonresponders at week 36 were rescued to receive baricitinib 2 mg or baricitinib 4 mg by randomized assignment (1:1). Patients in the placebo group who attained a response at week 36 remained on placebo. Those patients who experienced spontaneous regrowth remained on placebo until week 52, even if a relapse was observed between week 36 and week 52.

• Baricitinib-treated patients continued treatment regardless of response at week 36 and were reassessed at week 52.

At week 52:

• Responders: Patients who had been in the baricitinib 4 mg group from baseline and attained a response at week 52 underwent randomized down-titration where they were randomized in a 1:1 ratio to either stay on their existing 4 mg dose of baricitinib or transition to a 2 mg dose of baricitinib. Patients who were rescued to baricitinib at week 36 and patients randomized to a 2 mg dose of baricitinib were not eligible for randomized down-titration and remained in their same treatment group. Patients in the placebo group at week 52 who attained a response remained on placebo. Responders who experienced a loss of treatment benefit after week 52 and:

  • who were randomized to baricitinib 2 mg at week 52 (randomized down-titration) were re-treated with the 4 mg dose of baricitinib as randomized at baseline

  • who remained on a 4 mg dose of baricitinib at week 52 (randomized down-titration) continued to receive the same dose of baricitinib

  • who were randomized to a 2 mg dose of baricitinib at baseline were rescued to a 4 mg dose of baricitinib

  • who had remained on placebo since baseline were rescued to baricitinib 2 mg.

• Nonresponders:

  • Patients who had been in the baricitinib 4 mg dose group from baseline, had never attained a response by week 52, and had not had at least a 2-point improvement from baseline in a ClinRO measure for EB or EL hair loss at week 52 were transitioned to placebo.

  • Patients who had been in the baricitinib 4 mg dose treatment group, had attained a response before week 52, and had lost response remained on the baricitinib 4 mg dose.

  • Those who had been in the baricitinib 2 mg dose treatment group from baseline were rescued to the baricitinib 4 mg dose.

  • Those who were rescued to baricitinib at week 36 continued in their current treatment group at week 52.

  • Those who were randomized to placebo at baseline and were not eligible for rescue to baricitinib at week 36 (spontaneous remission) were rescued to baricitinib 2 mg.

At week 76:

• Patients who were nonresponders at both week 52 and week 76 were discontinued from the study at week 76, unless they had at least a 2-point improvement from baseline in a ClinRO measure for EB or EL hair loss.

This section summarizes the 36-week double-blind, placebo-controlled treatment period of Study BRAVE-AA1 (the phase III portion only) and Study BRAVE-AA2. The long-term extension period and bridging extension period and their results are summarized in the long term Extension Studies section.

Figure 2: Study Design of Study BRAVE-AA2

BARI = baricitinib; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; PBO = placebo; PTFU = post-treatment follow-up; QD = every day; SALT = Severity of Alopecia Tool; W = week; WO = washout.

^a Placebo-treated patients not eligible for rescue to baricitinib at week 36 (because of spontaneous remission) were rescued to baricitinib if they were nonresponders at week 52, or if they experienced a loss of treatment benefit after week 52.

^b Patients randomized to baricitinib 2 mg at week 0 were rescued to the baricitinib 4 mg dose if they were nonresponders at week 52 or were responders at week 52 but experienced more than a 20-point worsening in their SALT score after week 52.

^c Responders in the baricitinib 4 mg group (patients with a SALT score ≤ 20 who stayed on baricitinib 4 mg from week 0) were randomized to either stay on 4 mg or transition to baricitinib 2 mg.

^d Responders participating in the randomized down-titration who experienced a loss of treatment benefit after week 52 were re-treated with baricitinib 4 mg if they were randomized to the baricitinib 2 mg dose at week 52, or continued to receive baricitinib 4 mg if they were randomized to remain on the 4 mg dose at week 52.

^e At week 52, nonresponders (patients with a SALT score > 20) in the baricitinib 4 mg group since baseline who attained a SALT score of 20 or less before week 52 remained on 4 mg.

^f Never-responders (those who never attained a SALT score of ≤ 20 by week 52 despite being in the baricitinib 4 mg group from baseline and had not experienced a ≥ 2-point improvement from baseline in ClinRO measures for EB or EL hair loss) were automatically transitioned to placebo.

^g Nonresponders at both week 52 and week 76 were automatically discontinued at week 76 unless they had an improvement of 2 points or more from baseline in ClinRO measures for EB or EL hair loss.

Source: Study BRAVE-AA2 Clinical Study Report.⁸

Table 6: Details of Studies Included in the Systematic Review

AA = alopecia areata; AA-IGA = Alopecia Areata Investigator’s Global Assessment; ClinRO = clinician-reported outcome; EB = eyebrow; EI = eye irritation; EL = eyelash; EQ VAS = EQ visual analogue scale; HADS-A = Hospital Anxiety and Depression Scale-Anxiety; HADS-D = Hospital Anxiety and Depression Scale-Depression; MCS = mental component score; PCS = physical component score; PRO = patient-reported outcome; RCT = randomized controlled trial; SALT = Severity of Alopecia Tool; SALT₃₀ = at least a 30% improvement from baseline in the Severity of Alopecia Tool score; SALT₅₀ = at least a 50% improvement from baseline in the Severity of Alopecia Tool score; SALT₇₅ = at least a 75% improvement from baseline in the Severity of Alopecia Tool score; SALT₉₀ = at least a 90% improvement from baseline in the Severity of Alopecia Tool score; SALT₁₀₀ = a 100% improvement from baseline in the Severity of Alopecia Tool score; SF-36 = Short Form (36) Health Survey; Skindex-16 for AA = Skindex-16 for Alopecia Areata.

Note: Details included in Table 6 are from the sponsor’s Summary of Clinical Evidence.⁹

^aOnly details of the phase III portion of Study BRAVE-AA1 are reported in this table.

^bThis is as assessed at screening and baseline.

^cThis includes but is not limited to androgenetic alopecia, trichotillomania, and telogen effluvium.

^dExamples are tinea capitis, psoriasis, lupus erythematosus, or secondary syphilis.

^eThis was as clinically judged by the investigator to be at risk for suicide, or have suicidal ideation or behaviour as indicated by the Columbia-Suicide Severity Rating Scale, and ideation or behaviour that occurred within 2 months of the first visit in the study.

^fThis was as assessed among patients with a score of 2 or more at baseline.

^gThis was as assessed among patients with a score of 3 or more at baseline.

^hThis was a component of the SF-36.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Populations

Inclusion and Exclusion Criteria

Key inclusion and exclusion criteria of the BRAVE-AA1 and BRAVE-AA2 trials are presented in Table 6. The trials included patients who were aged at least 18 years and no more than 60 years for males (aged no more than 70 years for females), and had a current AA episode of more than 6 months’ duration that was severe or very severe in nature (i.e., hair loss encompassing at least 50% of the scalp as measured by SALT [the Alopecia Areata Investigator’s Global Assessment scale score of 3 or 4, corresponding to a SALT score of between 50 and 94 and a SALT score of between 95 and 100, respectively]). Eligible patients had no spontaneous improvement over the past 6 months (i.e., no more than 10 points in spontaneous reduction in the SALT score) and had a current episode of severe or very severe AA of less than 8 years’ duration. Patients were excluded if they had a primarily diffuse type of AA (characterized by diffuse hair shedding) or had significant uncontrolled neuropsychiatric disorder (e.g., suicidal ideation or behaviour).

Interventions

All study interventions were given once daily by oral means: baricitinib 2 mg, baricitinib 4 mg, and placebo. The baricitinib dose for patients randomized to the 4 mg daily treatment group who had renal impairment (defined as an estimated glomerular filtration rate less than 60 mL per minute per 1.73 m²) was 2 mg daily.

Each strength of a baricitinib tablet had a distinctive shape and colour, 4 mg versus 2 mg, and each tablet strength had a matching placebo. Patients received 2 tablets per day as follows:

• baricitinib 4 mg daily regimen — baricitinib 4 mg tablet and placebo tablet (matching baricitinib 2 mg); 1 each

• baricitinib 2 mg daily regimen — baricitinib 2 mg tablet and placebo tablet (matching baricitinib 4 mg); 1 each

• placebo daily regimen — placebo tablet (matching baricitinib 2 mg) and placebo tablet (matching baricitinib 4 mg); 1 each.

Double-blinding was in place to minimize bias. All study assessments were performed by study personnel who were blinded to patients’ treatment groups. Patients, investigators, and any other personnel interacting directly with patients or investigative sites remained blinded to intervention assignment until after completion of the study. Dose adjustment for renal impairment, if required, was managed by an interactive web response system to ensure the maintenance of blinding.

The following medications were permitted during the study: topical corticosteroids and topical calcineurin inhibitors (except on the scalp, EBs, and eyelids); intranasal, ophthalmic, or inhaled steroid; intra-articular or soft tissue corticosteroid injections (a maximum of 2 injections until week 36, no limit thereafter); nonlive vaccinations; bimatoprost ophthalmic solution (if on a stable dose for 8 weeks before randomization); 5-alpha reductase inhibitors, or oral or topical minoxidil (if on a stable dose for 12 months before randomization); statins; and ezetimibe.

The temporary interruption of treatment intervention occurred when any of the following prohibited medications was received by patients: live vaccines, probenecid, systemic corticosteroids, and phototherapy. The permanent discontinuation of treatment intervention occurred when any of the following prohibited medications was taken by patients: corticosteroids (systemic, intralesional, or topical on the scalp, EBs, and/or eyelids) for the treatment of AA; topical JAK inhibitors applied to the scalp, EBs, and eyelids; other oral JAK inhibitors; systemic immunosuppressants or immunomodulators; and any other AA treatment.

Outcomes

A list of efficacy end points assessed in this Clinical Review Report is provided in Table 7, followed by descriptions of the outcome measures. 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 we consulted and input from patient and clinician groups and public drug plans. Using the same considerations, our review team selected end points that were considered to be most relevant to inform our expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. All summarized efficacy end points were assessed using the GRADE tool. SAEs were also assessed using GRADE.

Considerations that informed the selection of outcomes to be summarized and assessed using GRADE include the following.

• SALT scores are used to assess the extent of scalp hair loss, which is an important outcome according to input from a patient group and the clinical experts. A SALT score of 20 or less was considered to be a reasonable threshold for clinically meaningful improvement in patients with severe AA based on literature identified by the sponsor⁴⁰ and clinical expert input. A brief description of the percentage change from baseline in SALT score results (GRADE not applied) is also provided in the Results section of this report.

• The proportion of patients attaining a SALT₅₀ response was an outcome that informs the sponsor’s pharmacoeconomic model and was included in the GRADE assessment. A brief description of the degree of alignment between SALT₇₅ (GRADE not applied) and SALT₅₀ responses is provided in the text.

• The loss of EB hair and EL hair is distressing to patients with AA as per clinical expert input. ClinRO measure-based outcomes were included in the GRADE assessment. A brief description of the degree of alignment between PRO measure-based results (GRADE not applied) and ClinRO measure-based results is provided in the text.

• The impact of AA on anxiety and depression was identified as an important outcome by a patient group and was assessed by the change from baseline in HADS anxiety and depression domain scores in the trials.

• HRQoL was identified as an important outcome in patients with AA, as per input from a patient group and the clinical experts. Of those HRQoL measures included in the trials, Skindex-16 for AA, which is an AA-specific measure, was included in the GRADE assessment. The Short Form (36) Health Survey (SF-36) and the EQ-5D-5L are generic HRQoL measures and were not included in the GRADE assessment. Results of these generic HRQoL measures are summarized in Appendix 1 given that they inform the sponsor’s pharmacoeconomic model.

• Clinical expert input noted that 36 weeks is a reasonable time frame in which to assess response to baricitinib treatment.

• Harms were considered important outcomes according to input from a patient group and the clinical experts. SAE was selected for the GRADE assessment. The 36-week time point was selected, as it was the latest follow-up time where the randomized placebo comparison was maintained.

Table 7: Outcomes Summarized From Study BRAVE-AA1 and Study BRAVE-AA2

ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; HADS-A = Hospital Anxiety and Depression Scale-Anxiety; HADS-D = Hospital Anxiety and Depression Scale-Depression; HRQoL = health-related quality of life; SALT = Severity of Alopecia Tool; Skindex-16 for AA = Skindex-16 for Alopecia Areata.

Note: Details included in Table 7 are from the sponsor’s Summary of Clinical Evidence.⁹

^aStatistical testing for these end points was adjusted for multiple comparisons.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Severity of Alopecia Tool

The SALT is an investigator-reported assessment tool used to assess the extent of scalp hair loss in AA.^30,41 The SALT uses a visual aid that helps visualize the amount of terminal hair loss in each of the 4 quadrants of the scalp and then, upon the summing of these, generates a SALT score (i.e., the total percentage of scalp hair loss), which ranges from 0 (no scalp hair loss) to 100 (complete scalp hair loss).³⁰ Evidence for the validity, reliability, and responsiveness of this instrument in patients with AA is summarized in Table 8. Attaining 80% or more scalp hair (a SALT score of 20 or less) has been shown to be a threshold for clinically meaningful improvement in patients with at least 50% scalp hair loss.⁴⁰ SALT₅₀ corresponds to at least 50% reduction (i.e., improvement) from baseline in the SALT score. Evidence for a MID in percentage change from baseline in the SALT score was not identified by the sponsor. Based on input from the clinical experts we consulted, a difference of at least 10% between the baricitinib group and the placebo group is likely to be clinically important with respect to the proportion of patients attaining a SALT score of 20 or less and the proportion of patients attaining SALT_50.

ClinRO Measures for EB and EL Hair Loss

The ClinRO assessments are novel clinician-administered instruments developed by the sponsor and are used to assess the extent of EB hair loss (ClinRO measure for EB hair loss) and EL hair loss (ClinRO measure for EL hair loss).⁴² The instruments use a 4-point categorical scale, ranging from 0 to 3; a higher score indicates more hair loss. Evidence for validity, reliability, responsiveness, and MID in patients with AA is summarized in Table 8. Based on input from the clinical experts we consulted, a difference of at least 10% between the baricitinib group and the placebo group with respect to the proportion of patients attaining a ClinRO measure for EB (or EL) hair loss score of 0 or 1 with at least a 2-point improvement from baseline is likely to be clinically important.

Hospital Anxiety and Depression Scale

The HADS is a 14-item patient-administered instrument that is used to assess the levels of anxiety and depression that a patient has experienced over the past week.^43,44 The HADS uses a 4-point Likert scale (0 to 3) for each question and is intended for patients aged between 12 years and 65 years.^43,44 Scores for each domain (anxiety and depression) range from 0 to 21, with higher scores indicating greater anxiety or depression.^44,45 Evidence for the validity, reliability, responsiveness, and MID of this instrument in patients with AA was not identified by the sponsor. The clinical experts we consulted did not suggest a reasonable MID estimate with certainty; therefore, the null was selected as the threshold when applying the GRADE assessment for these outcomes.

Skindex-16 for AA

The Skindex-16 questionnaire has been used to assess the HRQoL in patients with skin diseases. The Skindex-16 for AA is a version of the Skindex-16 that was adapted for use among adults with AA.⁹ It examines the degree to which patients are bothered by alopecia and associated symptoms. It is composed of 16 items grouped under 3 domains: symptoms, emotions, and functioning. Domain scores range from 0 (no effect) to 100 (effect experienced all the time), with higher scores indicating greater impact on quality of life.⁹ Evidence for the validity, reliability, responsiveness, and MID of this instrument in patients with AA was not identified by the sponsor. The clinical experts we consulted did not suggest a reasonable MID estimate with certainty; therefore, the null was selected as the threshold when applying the GRADE assessment for these outcomes.

Harms Outcomes

Harms outcomes included TEAEs, SAEs, WDAEs, death, and notable harms (infections, malignancies, gastrointestinal perforations, major adverse cardiac events [MACEs], and thromboembolic events). Based on clinical expert input, a difference of at least 5% between the baricitinib group and the placebo group in the proportion of patients experiencing SAEs is likely clinically important.

Table 8: Summary of Outcome Measures and Their Measurement Properties

AA = alopecia areata; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; HADS = Hospital Anxiety and Depression Scale; ICC = intraclass correlation coefficient; MID = minimal important difference, PRO = patient-reported outcome; SALT = Severity of Alopecia Tool; Skindex-16 for AA = Skindex-16 for alopecia areata.

Statistical Analysis

Table 9 summarizes the statistical analysis methods of the efficacy end points in the BRAVE-AA1 and BRAVE-AA2 trials. Baseline was defined as the last nonmissing measurement recorded on or before the first dose of treatment.

Sample Size and Power Calculation

A sample size calculation determined that approximately 625 randomized patients were required in the BRAVE-AA1 trial to demonstrate a statistically significant difference between each of the 2 treatment groups (baricitinib 2 mg and baricitinib 4 mg) and placebo with respect to the primary end point of proportion of patients attaining a SALT score of 20 or less at week 36, at a 2-sided significance level of 0.04 for baricitinib 4 mg and 0.01 for baricitinib 2 mg, with more than 90% power. This assumes that the response rates for baricitinib 4 mg, baricitinib 2 mg, and placebo are 30%, 20%, and 5%, respectively. In the BRAVE-AA2 trial, it was determined that 476 randomized patients were required to provide approximately 90% power to detect a statistically significant difference in the primary outcome based on the same assumptions. There was no sample size consideration for patients who entered randomized withdrawal (Study BRAVE-AA1) or randomized down-titration (Study BRAVE-AA2) at week 52.

Statistical Model

All efficacy analyses in the BRAVE-AA1 and BRAVE-AA2 trials were based upon the full analysis set population (Table 10). The difference between baricitinib and placebo with respect to the primary and key secondary end points were analyzed at an overall 2-sided significance level of 0.05. Categorical end points were tested using a logistic regression model, including geographic region, duration of the current episode of AA at baseline, baseline value, and treatment group as adjustment factors, except for outcomes related to HADS where baseline value was not included. Continuous efficacy end points were analyzed using an analysis of covariance model, including geographic region, duration of the current episode of AA at baseline, baseline value, and treatment group as adjustment factors.

Harms outcomes were assessed in the safety population (Table 10) and presented descriptively.

Multiple Testing Procedure

In both trials, a graphical procedure with a gatekeeping testing strategy for the primary and key secondary end points was implemented to control for type I error at a 2-sided significance level of 0.05. An overview of the testing procedure for Study BRAVE-AA1 and Study BRAVE-AA2 is outlined in Figure 3. Figure 4 further illustrates the testing scheme used within the tier 1 group for the testing procedure for Study BRAVE-AA2. Other secondary end points and exploratory end points were not controlled for type I error.

Handling of Missing Data and Imputation

Missing data for categorical efficacy outcomes were imputed using the nonresponder imputation (NRI) method, where patients were considered nonresponders if they did not meet clinical response criteria, permanently discontinued study treatment, or discontinued from the study at any time before the time point of interest for any reason. Missing data for continuous efficacy outcomes were imputed using the mLOCF method, where the most recent nonmissing postbaseline assessment was used and data after permanent study treatment discontinuation were not carried forward.

Subgroup Analyses

In consultation with the clinical experts we consulted, the CDA-AMC review team determined that the prespecified subgroup analyses of interest to this review included the duration of the current episode of AA category (less than 4 years and at least 4 years) and baseline SALT category (the severe category was a SALT score of 50 to 94 and the very severe category was a SALT score of 95 to 100). In both studies, to assess whether the treatment effect was similar across subgroups for the primary efficacy outcome, a logistic regression model was used and included treatment, baseline value, stratification variables, the subgroup variable, and the subgroup-by-treatment interaction. If the interaction was statistically significant at a significance level of 0.10, the treatment effect was estimated within each subgroup. There were no sample size considerations and no control for multiplicity in subgroup analyses.

Figure 3: Overview of the Graphical Testing Procedure for Study BRAVE-AA1 and Study Design BRAVE-AA2

Bari = baricitinib; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; PCFB = percentage change from baseline; PRO = patient-reported outcome; SALT = Severity of Alopecia Tool; SALT₅₀ = at least a 50% improvement from baseline in SALT score; SALT₉₀ = at least a 90% improvement from baseline in SALT score; SH = Scalp Hair Assessment; Wk = week.

Note: Details included in Figure 3 are from the sponsor’s Summary of Clinical Evidence.⁹

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Figure 4: Graphical Testing Procedure Within Tier 1 Group of End Points for Study BRAVE-AA2

ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; PCFB = percentage change from baseline; PRO = patient-reported outcome; SALT = Severity of Alopecia Tool; SALT₉₀ = at least a 90% improvement from baseline in SALT score; SH = Scalp Hair Assessment; Wk = week.

Note: Details included in Figure 4 are from the sponsor’s Summary of Clinical Evidence.⁹

Source: Study BRAVE-AA2 Clinical Study Report.⁸

Sensitivity Analyses

Hybrid imputation (primary and key secondary end points): Earlier in the COVID-19 pandemic, data were considered missing when the assessments were collected at remote visits or the whole visit was missed because of the pandemic. A sensitivity analysis using a hybrid imputation method was conducted to determine the effect of missing data because of the COVID-19 pandemic. For categorical end points, data that were missing because of COVID-19 were imputed using multiple imputation, while other data that were missing for reasons other than COVID-19 were imputed by NRI. This imputation procedure assumed that the effects of treatments would have been the same had patients not experienced any intercurrent event related to COVID-19 (i.e., a remote visit or missed visit) or that the effect would disappear after any intercurrent event unrelated to COVID-19. For continuous end points, data that were missing because of COVID-19 were imputed by multiple imputation, while other data that were missing for reasons other than COVID-19 were imputed by mLOCF. This imputation procedure assumed that the effects of treatments would remain the same had patients not experienced any intercurrent event related to COVID-19 or would remain the same after the event that caused data to be missing for reasons other than COVID-19.

Analysis Populations

Analysis populations of the BRAVE-AA1 and BRAVE-AA2 trials are summarized in Table 10.

Table 9: Statistical Analysis of Efficacy End Points in Study BRAVE-AA1 and Study BRAVE-AA2

ANCOVA = analysis of covariance; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; HADS-A = Hospital Anxiety and Depression Scale-Anxiety; HADS-D = Hospital Anxiety and Depression Scale-Depression; LOCF = last observation carried forward; MI = multiple imputation; mLOCF = modified last observation carried forward; NA = not applicable; NRI = nonresponder imputation; SALT = Severity of Alopecia Tool; SALT₅₀ = at least a 50% improvement from baseline in the Severity of Alopecia Tool score; Skindex-16 for AA = Skindex-16 for Alopecia Areata.

Note: Details included in Table 9 are from the sponsor’s Summary of Clinical Evidence.⁹

^aThis was among patients with a score of 2 or more at baseline.

^bPatients were considered nonresponders if they did not meet clinical response criteria or if they permanently discontinued study treatment or discontinued from the study at any time before the time point of interest for any reason.

^cThe most recent nonmissing postbaseline assessment was used. Data after permanent study treatment discontinuation were not carried forward.

^dThe most recent nonmissing postbaseline assessment was used.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Table 10: Analysis Populations of Study BRAVE-AA1 and Study BRAVE-AA2

Notes: Details included in Table 10 are from the sponsor’s Summary of Clinical Evidence.⁹

The trials included 2 additional analysis sets (a modified full analysis set and a per-protocol set). They were not of interest to this review and not summarized.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Results

Results of the 36-week, double-blind, placebo-controlled treatment phase of Study BRAVE-AA1 (the phase III portion only) and Study BRAVE-AA2 are summarized in this section.

Patient Disposition

Patient disposition in the BRAVE-AA1 and BRAVE-AA2 trials in the double-blind treatment period are summarized in Table 11.

In Study BRAVE-AA1 and Study BRAVE-AA2, 21.1% and 24.9% of patients were excluded during the screening period, respectively, mostly because they did not meet eligibility criteria. In Study BRAVE-AA1, a total of 654 patients were randomized to placebo (n = 189), baricitinib 2 mg (n = 184), and baricitinib 4 mg (n = 281). In Study BRAVE-AA2, a total of 546 patients were randomized to placebo (n = 156), baricitinib 2 mg (n = 156), and baricitinib 4 mg (n = 234). In both trials, study treatment discontinuation in the placebo group was higher numerically (11.1% in Study BRAVE-AA1 and 13.5% in Study BRAVE-AA2) compared with the baricitinib groups (2 mg [8.7%] and 4 mg [6.8%] in Study BRAVE-AA1 and 2 mg [10.9%] and 4 mg [7.7%] in Study BRAVE-AA2), with withdrawal by patient being the most common reason for treatment discontinuation in the placebo group.

In both trials, the full analysis set population consisted of all randomized patients in the treatment groups and the safety population consisted of all or close to all randomized patients (98.7% to 100%) in the treatment groups.

Table 11: Patient Disposition in Study BRAVE-AA1 and Study BRAVE-AA2 — DB Treatment Period

BARI = baricitinib; DB = double-blind; FAS = full analysis set.

Note: Details included in Table 11 are from the sponsor’s Summary of Clinical Evidence.⁹

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Baseline Characteristics

Patient baseline demographics, disease characteristics, and the history of treatment for AA of Study BRAVE-AA1 and Study BRAVE-AA2 are summarized in Table 12. The baseline characteristics outlined in the table are limited to those that are most relevant to this review or were felt to affect the outcomes or interpretation of the study results.

The baseline characteristics of the enrolled patients in general were similar between the BRAVE-AA1 and BRAVE-AA2 trials. The mean age of patients was 37.1 (SD = 13.0) years and 38.0 (SD = 12.7) years in Study BRAVE-AA1 and Study BRAVE-AA2, respectively. Compared to Study BRAVE-AA1, Study BRAVE-AA2 had proportionally more patients who were female (58.6% in Study BRAVE-AA1 and 63.2% in Study BRAVE-AA2) and white (45.9% in Study BRAVE-AA1 and 58.8% in Study BRAVE-AA2), and fewer patients who were Asian (41.2% in Study BRAVE-AA1 and 30.6% in Study BRAVE-AA2). In Study BRAVE-AA1 and Study BRAVE-AA2, patients had a mean duration of the current AA episode of 3.6 (SD = 3.9) years and 4.3 (SD = 4.9) years, respectively. Both trials consisted of a slightly higher proportion of patients with very severe AA (53.8% in Study BRAVE-AA1 and 52.5% in Study BRAVE-AA2) compared with those with severe AA (46.2% in Study BRAVE-AA1 and 47.5% in Study BRAVE-AA2). Approximately 90% of patients in the trials received prior AA treatment, with the most common treatments (reported in at least 40% of patients) being topical therapies, intralesional therapy, and systemic immunosuppressants and immunomodulators for AA.

Several between-group imbalances in baseline characteristics were identified in the trials. In Study BRAVE-AA1, compared with the placebo group, the baricitinib 2 mg group had a higher proportion of patients who were white (44.1% in the placebo group and 50.8% in the baricitinib 2 mg group), had very severe AA (51.3% in the placebo group and 58.2% in the baricitinib 2 mg group), and had adult onset of AA (59.3% in the placebo group and 67.9% in the baricitinib 2 mg group), as well as a lower proportion of patients who had received prior systemic immunosuppressive and immunomodulating treatments (53.4% in the placebo group and 45.7% in the baricitinib 2 mg group). A higher proportion of patients in the baricitinib 2 mg and baricitinib 4 mg groups compared to the placebo group had beforepical immunotherapy (31.0% in the baricitinib 2 mg group, 29.9% in the baricitinib 4 mg group, and 23.8% in the placebo group), phototherapy (18.5% in the baricitinib 2 mg group, 19.2% in the baricitinib 4 mg group, and 12.2% in the placebo group), and procedures for treating AA (22.3% in the baricitinib 2 mg group, 23.1% in the baricitinib 4 mg group, and 15.9% in the placebo group). In Study BRAVE-AA2, the baricitinib 2 mg and baricitinib 4 mg groups had a higher proportion of patients compared to the placebo group who were white (59.0% in the baricitinib 2 mg group, 61.5% in the baricitinib 4 mg group, and 54.5% in the placebo group) and a lower proportion of patients who had received prior systemic immunosuppressants and immunomodulators (57.1% in the baricitinib 2 mg group, 53.0% in the baricitinib 4 mg group, and 62.2% in the placebo group), and prior intralesional therapy (52.6% in the baricitinib 2 mg group, 44.4% in the baricitinib 4 mg group, and 56.4% in the placebo group). Also, a lower proportion of patients in the baricitinib 2 mg group (19.9%) had received beforepical immunotherapy compared to the placebo group (26.3%). No other notable between-group imbalances were observed for other baseline characteristics in the trials.

Table 12: Patient Baseline Characteristics in Study BRAVE-AA1 and Study BRAVE-AA2 — DB Treatment Period (FAS)

AA = alopecia areata; BARI = baricitinib; DB = double-blind; FAS = full analysis set; JAK = Janus kinase; SALT = Severity of Alopecia Tool; SD = standard deviation.

Note: Details included in Table 12 are from the sponsor’s Summary of Clinical Evidence.⁹

^aThis excludes topical immunotherapy.

^bThis includes cryotherapy, microneedling, and platelet-rich plasma injections.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Exposure to Interventions

Study Treatments

Patient exposure to study treatment in the double-blind treatment period of Study BRAVE-AA1 and Study BRAVE-AA2 are summarized in Table 13. No notable between-group difference was observed in the mean duration of exposure in both trials. The median treatment adherence rate varied from 98.4% to 99.2% across the treatment groups in the trials. One patient in the baricitinib 4 mg group of the BRAVE-AA1 trial had renal impairment requiring a dose adjustment to baricitinib 2 mg.

Concomitant Medications

The use of concomitant medications for AA in the double-blind treatment period of Study BRAVE-AA1 and Study BRAVE-AA2 are summarized in Table 14. In both trials, a small proportion of patients used at least 1 concomitant medication for AA (3.2% to 5.7% for Study BRAVE-AA1 and 3.2% to 5.1% for Study BRAVE-AA2). No notable imbalance was observed between treatment groups.

Table 13: Patient Exposure to Study Treatment in Study BRAVE-AA1 and Study BRAVE-AA2 — DB Treatment Period (Safety Population)

BARI = baricitinib; DB = double-blind; IQR = interquartile range; SD = standard deviation.

Note: Details included in Table 13 are from the sponsor’s Summary of Clinical Evidence.⁹

^aTotal patient-years was calculated in the safety population as the sum of the duration of exposure in days for all patients in the dosing regimen divided by 365.25.

^bThe duration of exposure was calculated in the full analysis set population as the date of the last dose of the study drug through week 36 – the date of the first dose of the study drug + 1.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Table 14: Concomitant AA Medications in Study BRAVE-AA1 and Study BRAVE-AA2 — DB Treatment Period (FAS)

AA = alopecia areata; BARI = baricitinib; DB = double-blind; FAS = full analysis set.

Note: Details included in Table 14 are from the sponsor’s Summary of Clinical Evidence.⁹

^aThese medications were reported in 2 or more patients in any treatment group.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Efficacy

Key efficacy results from the double-blind treatment period of Study BRAVE-AA1 and Study BRAVE-AA2 are summarized in Table 15.

SALT Score

Proportion of Patients Attaining SALT Score of 20 or Less

The proportion of patients attaining a SALT score of 20 or less at week 36 was the primary end point in both trials. At week 36, the between-group difference comparing the baricitinib 2 mg group versus the placebo group was 16.4% (95% CI, 9.7% to 23.4%; P < 0.001) in Study BRAVE-AA1 and 14.7% (95% CI, 8.3% to 21.6%; P < 0.001) in Study BRAVE-AA2. The between-group difference comparing the baricitinib 4 mg group and the placebo group was 29.9% (95% CI, 23.2% to 36.2%; P < 0.001) in Study BRAVE-AA1 and 29.9% (95% CI, 23.1% to 36.3%; P < 0.001) in Study BRAVE-AA2. Results of the primary analysis were in favour of both regimens of baricitinib and the sensitivity analysis (accounting for missing data because of the COVID-19 pandemic) showed consistent results. In both trials, subgroup analyses by baseline disease severity and duration of the current episode of AA were consistent with the primary analysis (Appendix 1, Table 28).

The percentage change from baseline in the SALT score was assessed at week 36 (a key secondary end point) in both trials. In both trials, the between-group difference comparing baricitinib and placebo was in favour of baricitinib for both the 2 mg regimen (–23.1% [95% CI, –30.6% to –15.6%; P < 0.001] in Study BRAVE-AA1 and –25.3% [95% CI, –32.8% to –17.7%] in Study BRAVE-AA2) and the 4 mg regimen (–37.7% [95% CI, –44.4% to –30.9%; P < 0.001] in Study BRAVE-AA1 and 44.5% [95% CI, –51.3% to –37.7%; P < 0.001] in Study BRAVE-AA2).

Proportion of Patients Attaining SALT₅₀

The between-group difference in the proportion of patients attaining a SALT₅₀ response at week 36 (a secondary end point) comparing the baricitinib 2 mg group versus the placebo group was 17.7% (95% CI, 9.5% to 25.8%; P < 0.001) in Study BRAVE-AA1 and 23.1% (95% CI, 15.1% to 31.0%; P < 0.001) in Study BRAVE-AA2. The between-group difference comparing the baricitinib 4 mg group with the placebo group was 33.6% (95% CI, 25.6% to 40.7%; P < 0.001) in Study BRAVE-AA1 and 41.9% (95% CI, 34.0% to 48.7%; P < 0.001) in Study BRAVE-AA2. Results of the SALT₇₅ responder analysis were consistent with the SALT₅₀ responder analysis (Table 29 in Appendix 1). Neither end point was adjusted for multiplicity in the trials. Analyses using a lower threshold (e.g., at least a 30% improvement from baseline in the SALT score) were not conducted.

ClinRO Measures for EB and EL Hair Loss

Proportion of Patients Attaining ClinRO Measure for EB Hair Loss Score of 0 or 1 With Improvement of 2 Points or More From Baseline Among Patients With ClinRO Measure for EB Hair Loss Score of 2 or More at Baseline

Between 66.3% and 73.9% of all randomized patients had a ClinRO measure for EB hair loss score of at least 2 at baseline in the trials and contributed to the analysis of the proportion of patients who had a ClinRO measure for EB hair loss score of 0 or 1 with at least a 2-point improvement from baseline at week 36 (a key secondary end point).

The between-group difference comparing baricitinib 2 mg versus placebo was 15.9% (95% CI, 8.4% to 23.6%; P < 0.001) in favour of baricitinib 2 mg in Study BRAVE-AA1. In Study BRAVE-AA2, the between-group difference comparing baricitinib 2 mg versus placebo was 7.1% (95% CI, –0.3% to 15.0%; P = 0.08); because of the failure of this end point, no formal testing was conducted for subsequent end points in the statistical hierarchy in this study. The between-group difference was in favour of baricitinib 4 mg over placebo in both trials (28.2% [95% CI, 20.3% to 35.4%; P < 0.001] in Study BRAVE-AA1 and 30.3% [95% CI, 21.4% to 38.4%; P < 0.001] in Study BRAVE-AA2). Results based on the PRO measure showed consistent results.

Proportion of Patients Attaining ClinRO Measure for EL Hair Loss Score of 0 or 1 With Improvement of 2 Points or More From Baseline Among Patients With ClinRO Measure for EL Hair Loss Score of 2 or More at Baseline

Between 51.3% and 60.3% of all randomized patients had a ClinRO measure for EL hair loss score of at least 2 at baseline in the trials and contributed to the analysis of the proportion of patients who had a ClinRO measure for EL hair loss score of 0 or 1 with at least a 2-point improvement from baseline at week 36 (a key secondary end point).

The between-group difference comparing baricitinib 2 mg and placebo was 10.4% (95% CI, 2.7% to 18.3%) in Study BRAVE-AA1 and 4.6% (95% CI, –3.7% to 13.2%) in Study BRAVE-AA2, neither of which was formally tested for statistical significance because of a prior failure of an outcome in the statistical hierarchy. The between-group difference favoured baricitinib 4 mg treatment over placebo in both trials (30.4% [95% CI, 21.6% to 38.1%; P < 0.001] in Study BRAVE-AA1 and 28.7% [95% CI, 18.7% to 37.5%; P < 0.001] in Study BRAVE-AA2). Results based on the PRO measure showed consistent results.

HADS Anxiety and HADS Depression Domain Scores

The mean changes from baseline in HADS anxiety and HADS depression domain scores were secondary end points and not adjusted for multiplicity in both trials.

Change From Baseline in HADS Anxiety Domain Score

The between-group difference comparing baricitinib 2 mg and placebo with respect to change from baseline in the HADS anxiety domain score at week 36 favoured baricitinib 2 mg in Study BRAVE-AA1 at –0.8 (95% CI, –1.4 to –0.3; P ≤ 0.01) and at –0.2 (95% CI, –0.8 to 0.4; P = 0.5) in Study BRAVE-AA2. The between-group difference comparing baricitinib 4 mg and placebo favoured baricitinib 4 mg in both trials (–0.5 [95% CI, –1.1 to 0.0; P = 0.04] in Study BRAVE-AA1 and –0.7 [95% CI, –1.3 to –0.2; P = 0.01] in Study BRAVE-AA2).

Change From Baseline in HADS Depression Domain Score

The between-group difference comparing baricitinib 2 mg and placebo with respect to change from baseline in the HADS depression domain score at week 36 was –0.4 (95% CI, –0.9 to 0.1; P = 0.1) in Study BRAVE-AA1 and –0.5 (95% CI, –1.1 to 0.1; P = 0.08) in Study BRAVE-AA2. The between-group difference comparing baricitinib 4 mg and placebo favoured baricitinib 4 mg in Study BRAVE-AA2 at –0.7 (95% CI, –1.2 to –0.2; P = 0.01) and at –0.3 (95% CI, –0.8 to 0.1; P = 0.2) in Study BRAVE-AA1.

Skindex-16 for AA

The mean changes from baseline in Skindex-16 for AA domain scores (symptoms, emotions, and functioning) were exploratory end points in Study BRAVE-AA1 and secondary end points in Study BRAVE-AA2. These end points were not adjusted for multiplicity.

Change From Baseline in Skindex-16 for AA Symptoms Domain Score

The difference between the baricitinib 2 mg group and the placebo group with respect to change from baseline in the Skindex-16 for AA symptoms domain score at week 36 favoured the baricitinib 2 mg group in Study BRAVE-AA1 at –4.76 (95% CI, –9.13 to –0.40; P = 0.033) and at –3.02 (95% CI, –6.91 to 0.88; P = 0.129) in Study BRAVE-AA2. The difference between the baricitinib 4 mg group and the placebo group favoured the baricitinib 4 mg group in Study BRAVE-AA2 at –4.21 (95% CI, –7.75 to –0.68; P = 0.020) and at –2.75 (95% CI, –6.67 to 1.17; P = 0.168) in Study BRAVE-AA1.

Change From Baseline in Skindex-16 for AA Emotions Domain Score

The between-group difference with respect to change from baseline in the Skindex-16 for AA emotions domain score at week 36 was in favour of baricitinib over placebo in both trials for both the baricitinib 2 mg regimen (–11.50 [95% CI, –17.71 to –5.30; P < 0.001] in Study BRAVE-AA1 and –6.75 [95% CI, –12.68 to –0.82; P = 0.026] in Study BRAVE-AA2) and the baricitinib 4 mg regimen (–11.01 [95% CI, –16.57 to –5.45; P < 0.001] in Study BRAVE-AA1 and –13.42 [95% CI, –18.80 to –0.84; P < 0.001] in Study BRAVE-AA2).

Change From Baseline in Skindex-16 for AA Functioning Domain Score

The difference between the baricitinib 2 mg group and the placebo group with respect to change from baseline in the Skindex-16 for AA functioning domain score at week 36 was –5.07 (95% CI, –10.94 to 0.80; P = 0.090) in Study BRAVE-AA1 and –4.38 (95% CI, –9.65 to 0.88; P = 0.103) in Study BRAVE-AA2. The difference between the baricitinib 4 mg group and the placebo group favoured the baricitinib 4 mg group in both trials (–7.04 [95% CI, –12.31 to –1.77; P = 0.009] in Study BRAVE-AA1 and –8.33 [95% CI, –13.10 to –3.56; P < 0.001] in Study BRAVE-AA2).

Table 15: Key Efficacy Results From Study BRAVE-AA1 and Study BRAVE-AA2 — DB Treatment Period (FAS)

ANCOVA = analysis of covariance; BARI = baricitinib; CI = confidence interval; ClinRO = clinician-reported outcome; DB = double-blind; EB = eyebrow; EL = eyelash; FAS = full analysis set; HADS = Hospital Anxiety and Depression Scale; LSM = least squares mean; NA = not applicable; PBO = placebo; SALT = Severity of Alopecia Tool; SALT₅₀ = at least a 50% improvement from baseline in the Severity of Alopecia Tool score; SE = standard error; Skindex-16 for AA = Skindex-16 for Alopecia Areata; vs. = versus.

Note: Details included in Table 15 are from the sponsor’s Summary of Clinical Evidence.⁹

^aThese between-group differences were analyzed using a logistic regression model, including geographic region, the duration of the current episode at baseline, baseline value, and treatment group as adjustment factors.

^bThis end point was not adjusted for multiplicity and was at a higher risk of type I error (false-positive results).

^cThis end point was not formally tested for statistical significance because of a prior failure in the statistical hierarchy.

^dThese between-group differences were analyzed using an ANCOVA model, including geographic region, the duration of the current episode at baseline, baseline value, and treatment group as adjustment factors.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Harms

Harms results from the double-blind treatment period of the BRAVE-AA1 and BRAVE-AA2 trials are summarized in Table 16.

Treatment-Emergent Adverse Events

TEAEs were reported in 50.8% to 59.6% of patients across treatment groups in Study BRAVE-AA1; the frequency was slightly higher in the baricitinib 4 mg group (59.6%) compared with the placebo group (51.3%). TEAEs were reported in 63.0% to 68.4% of patients across treatment groups in Study BRAVE-AA2, with no notable between-group differences. In both trials, the most common TEAEs of baricitinib (reported in at least 5% of patients in either baricitinib group) were upper respiratory tract infection, headache, urinary tract infection, nasopharyngitis, acne, and increased blood creatine phosphokinase.

Serious Adverse Events

SAEs were reported at a low frequency, with no notable between-group differences in either trial (1.6% to 2.1% in Study BRAVE-AA1 and 1.9% to 3.4% in Study BRAVE-AA2).

Withdrawal Due to Adverse Events

WDAEs were reported at a low frequency, with no notable between-group differences in either trial (1.1% to 1.8% in Study BRAVE-AA1 and 2.6% in all groups in Study BRAVE-AA2).

Mortality

No deaths were reported in either trial.

Notable Harms

Infection

Treatment-emergent infection was reported in 25.1% to 31.4% of patients across treatment groups in Study BRAVE-AA1 and 29.2% to 37.4% of patients across treatment groups in Study BRAVE-AA2. In Study BRAVE-AA2, the frequency of infection was higher in the baricitinib 2 mg group (37.4%) compared with the placebo group (29.2%), but this was not observed in Study BRAVE-AA1. In Study BRAVE-AA1, none of the infections was reported to be serious or to have led to treatment discontinuation. In Study BRAVE-AA2, serious infection was reported in 2 (1.3%) patients and 1 (0.4%) patient in the baricitinib 2 mg and baricitinib 4 mg groups, respectively, and infection leading to treatment discontinuation was reported in 1 (0.6%) patient in the baricitinib 2 mg group. Infection leading to treatment interruption was reported in 1.1% to 5.2% of patients across the trials.

Cardiovascular and Thromboembolic Events

In Study BRAVE-AA1, myocardial infarction and coronary revascularization was reported in 1 (0.5%) patient in the baricitinib 2 mg group. Serious arrhythmia was reported in 1 (0.5%) patient in the baricitinib 4 mg group. There were no reports of cardiovascular events in Study BRAVE-AA2. There were no reports of venous or pulmonary thromboembolic events in either trial.

Other Notable Harms

No gastrointestinal perforations were reported in either trial. No nonmelanoma skin cancers were reported in either trial; 1 patient in each of the placebo group (0.6%) and the baricitinib 4 mg group (0.4%) reported other forms of malignancies in Study BRAVE-AA2.

Table 16: Harms Results from Study BRAVE-AA1 and Study BRAVE-AA2 — DB Treatment Period (Safety Population)

BARI = baricitinib; CV = cardiovascular; DB = double-blind; MACE = major adverse cardiac event; MI = myocardial infarction; NMSC = nonmelanoma skin cancer; SAE = serious adverse event; TEAE = treatment-emergent adverse event; WDAE = withdrawal due to adverse event.

Note: Details included in Table 16 are from the sponsor’s Summary of Clinical Evidence.⁹

^aThese occurred in at least 5% of patients in any group.

^bNo adverse event occurred in more than 1 patient.

^cThis was a TEAE.

Sources: Study BRAVE-AA1 Clinical Study Report and Study BRAVE-AA2 Clinical Study Report.^7,8

Critical Appraisal

Internal Validity

Study BRAVE-AA1 and Study BRAVE-AA2 were randomized, double-blind, placebo-controlled trials. The method used for randomization consisted of an interactive web response system, which enabled the concealment of the allocation sequence. The baseline patient demographics and disease characteristics were in general balanced between treatment groups, except for race, the onset of AA, the proportion of patients with severe AA, and treatment history, where between-group differences were identified. The clinical experts noted that the impact of such imbalances on study results is likely to be insignificant.

The blinding of patients and study personnel was appropriately maintained. However, given that a placebo was used as a control in the trials, there is a possibility that patients may have been able to infer treatment assignment through improvement in hair loss over the study period, which could have introduced bias to the results in favour of baricitinib for efficacy outcomes that required subjective judgment (ClinRO measure, HADS, and Skindex-16 for AA). The risk of bias in the measurement of the outcome is low for the SALT tool, which was an objective measure assessed by the investigators. There is a risk of inflated subjective harms with baricitinib treatment because of potential unblinding.

An adequate sample size was attained in both trials based on the a priori sample size calculations for the primary end point. Efficacy analyses were conducted in the full analysis set population, which included all randomized patients in the group to which they were randomized (i.e., intention-to-treat); this is the ideal approach to assess the effect of assignment to the intervention. Responder analyses of ClinRO measures (EB and EL loss) were conducted in patients with specific baseline scores. In total, 30% to 50% of patients were excluded from these analyses because of not having the specified baseline score, which could impact the maintenance of randomization (i.e., prognostic imbalances between groups could arise). The extent and direction of the resulting bias is, however, unclear.

In both trials, study treatment discontinuation in the placebo group was higher numerically (11.1% in Study BRAVE-AA1 and 13.5% in Study BRAVE-AA2) compared with the baricitinib groups (8.7% [baricitinib 2 mg group] and 6.8% [baricitinib 4 mg group] in Study BRAVE-AA1 and 10.9% [baricitinib 2 mg group] and 7.7% [baricitinib 4 mg group] in Study BRAVE-AA2), primarily because of withdrawal by patient, which could potentially lead to attrition bias in favour of baricitinib. Missing data for binary outcomes (SALT and ClinRO measure-based responder analyses) were imputed using NRI, which was a conservative approach. Missing data for continuous secondary end points (change from baseline in the HADS anxiety and depression domain scores and Skindex-16 for AA domain scores) were imputed using LOCF (where the most recent nonmissing postbaseline assessment was used) or modified mLOCF (where, in addition, data after permanent study treatment discontinuation were not carried forward). The sponsor noted that very few patients experienced waxing and waning in scalp hair during treatment from the phase II portion of Study BRAVE-AA1 and several external studies on AA; therefore, they felt that the use of mLOCF or LOCF imputation for missing data was reasonable. Nonetheless, given the differential discontinuation rate, particularly between the baricitinib 4 mg group and the placebo group where a notable difference was consistently observed in both trials, potential bias in favour of baricitinib could not be ruled out. A sensitivity analysis (hybrid imputation) assessing the impact of COVID-19 was conducted on the primary end point and showed results consistent with the primary analysis.

A hierarchical testing procedure was appropriately used to account for multiplicity in the primary and key secondary end points. Analyses of other secondary and exploratory end points were not part of the statistical hierarchy. As such, statistically significant results are at an increased risk of type I error (false-positive results). In both studies, results of the ClinRO measure for EL hair loss responder analysis comparing the baricitinib 2 mg and placebo groups were not formally tested for statistical significance because of a prior failure in the statistical hierarchy but provided supportive evidence. Subgroup analyses of interest (by baseline disease severity and duration of the current episode of AA) were specified a priori; however, there was a lack of sample size consideration and control for multiplicity for these subgroup analyses, which precluded definitive conclusions on subgroup effects.

The primary end point was assessed using the SALT tool for which evidence for validity in patients with AA is available. The responder analysis was based on a threshold of a SALT score of less than 20, which was consistent with the literature-identified MID estimate of the instrument.⁴⁰ Evidence supporting the choice of threshold in percentage reduction in the SALT score responder analysis (e.g., SALT₅₀) was not identified by the sponsor. The validity of ClinRO measures, which are novel instruments developed by the sponsor, was demonstrated in patients with AA in a study.⁴² Evidence supporting the use of a ██████████ ███████████ ████ ████████ ██ ███ █████████ ████████ of ClinRO EB and EL scores was available.^48,49 Evidence for the validity and MID estimate of HADS and Skindex-16 for AA outcomes in patients with AA was not identified by the sponsor. In the absence of an MID estimate, it is unclear if the treatment effects of baricitinib observed in the studies were clinically important to patients and clinicians.

External Validity

The clinical experts we consulted noted that the inclusion and exclusion criteria of the trials in general were reflective of the patient population eligible for baricitinib treatment in Canada, although patients with a diffuse type of AA (estimated to range from 5% to 10% of patients) would not necessarily be excluded from treatment in clinical practice. The trials excluded males older than 60 years and females older than 70 years. One clinical expert commented that older adults are reasonable candidates for baricitinib treatment while the other clinical expert noted that older adults are much less likely to benefit from baricitinib treatment compared with younger patients since older adults tend to have other concurrent causes of hair loss that are not expected to be responsive to baricitinib treatment. According to the clinical experts, the baseline characteristics of the trial population were in general aligned with the patient population expected to receive baricitinib treatment, except that the proportion of patients with very severe AA appeared to be higher than in clinical practice; the clinical experts noted that patients with very severe AA tend to be less responsive to treatment. As well, the mean baseline HADS scores suggested the trial population had a normal to low degree of anxiety and depression; according to the clinical experts, this did not align with clinical practice since severe AA is generally associated with significant psychological burden. The clinical experts commented that the exclusion of patients with significant uncontrolled neuropsychiatric disorder (e.g., suicidal ideation or behaviour) from the trials was appropriate since these patients would not be candidates for baricitinib treatment in clinical practice because of concerns with treatment adherence.

The dosing regimen of baricitinib in the trials generally aligns with the product monograph, although in the trials, the dosing of baricitinib was assigned by randomization without regard to disease severity. This is different from the recommended dose from the product monograph where the starting dose should be determined based on the severity of hair loss in patients. In addition, dose adjustment of baricitinib occurred at week 52 in the trials. According to the clinical experts, the decision to adjust the dose as per response to treatment would likely take place in clinical practice at an earlier time point than week 52 (e.g., week 36). With regard to concomitant medication use, corticosteroids (systemic, intralesional, and topical) were prohibited in the trials; however, the clinical experts expected that these treatments might be used in combination with baricitinib treatment in clinical practice.

The efficacy outcomes assessed in the study were of clinical importance to patients and clinicians, including the severity of hair loss (scalp, EBs, and ELs), psychological impacts (anxiety and depression), and HRQoL. The clinical experts noted that none of the outcome instruments used in the clinical trials are administered in clinical practice. According to the clinical experts, the duration of follow-up of 36 weeks was adequate for the assessment of the efficacy of baricitinib. However, the clinical experts felt that a longer follow-up would be required to capture the long-term safety of baricitinib, including potential rare AEs (e.g., MACE, malignancy, thrombosis), since baricitinib is expected to be a lifelong treatment for many patients with AA.

It should be noted that the BRAVE-AA1 and BRAVE-AA2 trials are the only phase III RCT evidence submitted for review and that they are placebo-controlled studies. No head-to-head evidence comparing baricitinib with systemic treatments for severe AA that are currently reimbursed by the public drug plans (conventional immunosuppressants) were submitted. In addition, the absence of evidence for baricitinib in older adults (males older than 60 years and females older than 70 years — categories that were excluded from the trials) was an evidence gap in the treatment of severe AA since some clinicians may be open to prescribing baricitinib treatment to patients in this age group. In an effort to address this treatment gap, a single-arm observational study (Tang et al. [2024]) assessing the efficacy of baricitinib treatment in older adults with AA was submitted by the sponsor. This study is summarized in the Studies Addressing Gaps in the Systematic Review Evidence section of this report.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered most relevant to inform our expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group.^50,51

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

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

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

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

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

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. For this review, the target of the certainty of evidence assessment was based on the presence or absence of an important effect as informed by thresholds identified based on clinical expert input (a SALT score of 20 or less, a SALT₅₀ response, and ClinRO measures). For other outcomes of interest to this review (HADS and Skindex-16 for AA outcomes), the target of the certainty of evidence assessment was based on the null since there was uncertainty in the thresholds suggested by the clinical experts. Findings from Study BRAVE-AA1 and Study BRAVE-AA2 were considered together and summarized narratively per outcome because these studies were similar in population, interventions, design, and outcome measures.

Results of GRADE Assessments

Table 2 and Table 3 present the GRADE summary of findings for baricitinib versus placebo in patients with severe or very severe AA.

Long-Term Extension Studies

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

Description of Studies

Study BRAVE-AA1

This is an ongoing long-term extension study (week 36 onward) of Study BRAVE-AA1 (the anticipated completion date is January 26, 2025). The purpose of this study is to provide safety and efficacy analyses through week 104 to support dosing recommendations in product labelling.

At week 52, patients initially randomized to baricitinib who were responders (with a SALT score of ≤ 20) were rerandomized at a 3:1 ratio to stay on their current dose of baricitinib or to transition to placebo (a randomized withdrawal). Responders who had been rerandomized to placebo and had experienced a loss of treatment benefit (defined as > a 20-point worsening in the SALT score from week 52) at any time after week 52 were re-treated with their original baricitinib dose and the efficacy of re-treatment was analyzed as part of the other secondary end points of Study BRAVE-AA1.

This extension study included week 0 to week 52 and week 52 to week 76 efficacy and safety data for patients whose treatment dose with baricitinib was titrated up at week 52 (the up-titration cohort). The up-titration cohort included ███ patients randomized to baricitinib 2 mg at week 0 who did not attain a SALT score of 20 or less at week 52. The baricitinib dose in all ███ patients was titrated up to 4 mg.

Study BRAVE-AA2

This is an ongoing long-term extension study (week 36 onward) of Study BRAVE-AA2 (the anticipated completion date is July 29, 2025). The purpose of this study is to provide efficacy and safety analyses to support dosing recommendations in product labelling.

At week 52, patients were divided into 2 cohorts. The randomized down-titration cohort included 82 patients randomized at week 0 to baricitinib 4 mg who attained a SALT score of 20 or less at week 52. Of these, 42 patients were randomly assigned to remain on baricitinib 4 mg and 40 patients were randomly assigned to the down-titration of baricitinib to 2 mg. The up-titration cohort included ██ patients randomized to baricitinib 2 mg at week 0 who did not attain a SALT score of 20 or less at week 52. The baricitinib dose in all ██ patients was titrated up to 4 mg.

Eligibility Criteria

The inclusion and exclusion criteria for the long-term extension studies were previously described in the Systematic Review section.

Statistical Analysis

Study BRAVE-AA1 and Study BRAVE-AA2

Efficacy

Efficacy analyses were descriptive without inferential treatment comparisons. The efficacy results were presented for 2 analysis periods — 1 from week 0 to week 52 and 1 from week 52 to week 76 — using data from the baricitinib 2 mg up-titration cohort. Baseline assessments for the study were conducted at the start of each trial.

For the efficacy analysis from week 0 through week 52, the primary censoring rule was implemented; that is, efficacy results after permanent study drug discontinuation or results that were collected during remote visits because of the COVID-19 pandemic were censored. The primary censoring rule along with NRI or mLOCF for missing data imputation was applied to the summary statistics.

For the efficacy analysis from week 52 through week 76, the primary censoring rule was also implemented. The primary censoring rule along with NRI or mLOCF for missing data imputation was applied to the summary statistics. The mLOCF method imputed missing values post–week 52 using the last nonmissing assessment after patients switched to baricitinib 4 mg. If no assessment was completed post–week 52, the mLOCF method was not applicable. Censoring rules in the long term studies are summarized in Table 17.

Table 17: Censoring Rules in the long term Extension Period of Study BRAVE-AA1 and Study BRAVE-AA2

Source: Senna et al. (2024).³⁸

Harms

Safety analyses were performed for 2 analysis periods: 1 from week 0 through week 52 and 1 from week 52 through week 76. Descriptive statistics were provided. No treatment comparison was made. These analyses were conducted for TEAEs, SAEs, and AEs leading to permanent study drug discontinuation, and AEs of special interest.

Populations

Baseline Characteristics

A summary of the baseline demographic and clinical characteristics of participants in the Study BRAVE-AA1 up-titration cohort are presented in Table 18. For the randomized down-titration cohort and up-titration cohort of Study BRAVE-AA2, baseline demographic and clinical characteristics are presented in Table 19 and Table 20, respectively.

Table 18: Summary of Baseline (Week 52) Demographic and AA Clinical Characteristics in the Up-Titration Population of Study BRAVE-AA1

AA = alopecia areata, BARI = baricitinib; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; HADS = Hospital Anxiety and Depression Scale; PRO = patient-reported outcome; SALT = Severity of Alopecia Tool; SD = standard deviation.

^aAtopic background is defined as a medical history of or ongoing atopic dermatitis, allergic rhinitis, or allergic asthma.

Source: Study BRAVE-AA1 Clinical Study Report Addendum.¹⁰

Table 19: Summary of Baseline Demographics and AA Clinical Characteristics in the Randomized Down-Titration Cohort in Study BRAVE-AA2

AA = alopecia areata, BARI = baricitinib; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; HADS = Hospital Anxiety and Depression Scale; PRO = patient-reported outcome; SALT = Severity of Alopecia Tool; SD = standard deviation.

^aAtopic background is defined as a medical history of or ongoing atopic dermatitis, allergic rhinitis, or allergic asthma.

Source: Study BRAVE-AA2 Clinical Study Report Addendum.¹¹

Table 20: Summary of Baseline Demographics and AA Clinical Characteristics in the Titration Up Cohort in Study BRAVE-AA2

AA = alopecia areata, BARI = baricitinib; ClinRO = clinician-reported outcome; EB = eyebrow; EL = eyelash; HADS = Hospital Anxiety and Depression Scale; PRO = patient-reported outcome; SALT = Severity of Alopecia Tool; SD = standard deviation.

^aAtopic background is defined as a medical history of or ongoing atopic dermatitis, allergic rhinitis, or allergic asthma.

Source: Study BRAVE-AA2 Clinical Study Report Addendum.¹¹

Outcomes

Study BRAVE-AA1 and Study BRAVE-AA2

The main efficacy end point was the proportion of patients attaining a SALT score of 20 or less at week 52, week 76, and week 104.

Results

Patient Disposition

Study BRAVE-AA1

The baricitinib 2 mg up-titration cohort included ███ patients randomized at week 0 to baricitinib 2 mg who did not attain a SALT score of 20 or less at week 52. Of the ███ patients, ███ patients █████ completed week 76, and ███ patients █████ discontinued after week 52 and before week 76 because of AE ██ ████████ █████, lack of efficacy ██ █████████ █████, withdrawal by patient ██ █████████ █████, lost to follow-up ██ █████████ █████, and ██ ████████ ███████ included in the “other” category who were nonresponders at week 76.

Study BRAVE-AA2

Randomized Down-Titration Cohort

The randomized down-titration cohort included 82 patients randomized at week 0 to baricitinib 4 mg who attained a SALT score of 20 or less at week 52. Of these 82 patients, 42 patients were randomly assigned to remain on baricitinib 4 mg and 40 patients were randomly assigned to the titration down of baricitinib to 2 mg.

Of the 42 patients who remained on the baricitinib 4 mg dose, 100% of the patients completed week 76. Of the 40 patients whose treatment dose was down-titrated to baricitinib 2 mg, ███ patients █████ completed week 76, and ███ patients ████ discontinued before week 76 because of AE ██ ████████ █████, and withdrawal by patient ██ ████████ ██████

Up-Titration Cohort

The up-titration cohort included ███ patients randomized at week 0 to receive baricitinib 2 mg who did not attain a SALT score of 20 or less at week 52. Of these ███ patients, ███ patients █████ completed week 76, and ███ patients █████ discontinued before week 76 because of lack of efficacy ██ █████████ █████, withdrawal by patient ██ █████████ █████, and ██ ████████ ███████ included in the “other” category who were nonresponders at week 76.

Exposure to Study Treatments

A summary of the duration of exposure to baricitinib in Study BRAVE-AA1 and Study BRAVE-AA2 is presented in Table 21 and Table 22, respectively.

Table 21: Patient Exposure in Study BRAVE-AA1

BARI = baricitinib; IQR = interquartile range; NR = not reported; PBO = placebo; SD = standard deviation.

Note: In Study BRAVE-AA1, patients were deemed nonadherent if their adherence was less than 80% or 120% or greater between week 0 and week 52.

Source: Study BRAVE-AA1 Clinical Study Report Addendum.¹⁰

Table 22: Patient Exposure in Study BRAVE-AA2

BARI = baricitinib; IQR = interquartile range; SD = standard deviation.

Note: In Study BRAVE-AA2, patients are deemed nonadherent if their adherence was less than 80% or 120% or greater between week 0 and week 52.

Source: Study BRAVE-AA2 Clinical Study Report Addendum.¹¹

Efficacy: Up to Week 76 (Data Cut-Off Date of February 2021)

Proportion of Patients Attaining SALT Score of 20 or Less

The proportion of patients attaining a SALT score of 20 or less continuously increased over the treatment period beyond 36 weeks for the baricitinib 4 mg cohort. At week 52, 40.9% and 21.2% of patients receiving baricitinib 4 mg (N = 281) and baricitinib 2 mg (N = 184), respectively, attained a SALT score of 20 or less in Study BRAVE-AA1. Similarly, 36.8% and 24.4% of patients receiving baricitinib 4 mg (N = 234) and baricitinib 2 mg (N = 156), respectively, attained a SALT score of 20 or less at week 52 in Study BRAVE-AA2.

Study BRAVE-AA1: Up-Titration Cohort

At week 52, ███ patients who were originally randomized to baricitinib 2 mg were considered nonresponders and were titrated up to baricitinib 4 mg. At week 76, following 24 weeks of treatment on baricitinib 4 mg, ███ of ███ patients (████ ███ ████████ ██ ██████ in the up-titration cohort attained a SALT score of 20 or less.

Study BRAVE-AA2: Randomized Down-Titration Cohort

At week 52, 82 patients who were originally randomized to baricitinib 4 mg were eligible for randomized down-titration to baricitinib 2 mg. At week 52, █████ ████ ███ █████ ██ ██████ of patients attained a SALT score of 20 or less. ███ ███████ ███ █ ███████ █████ ██ ████ ███ ███ ████████ ████ █████████ █████ ████ ██████ ███ █████████ ████ ███ ████████ ████ ██ ████ ██ ██ ███ ████.

Among patients receiving baricitinib 4 mg who attained a SALT score of 20 or less at week 52, this response was retained up to week 76 in 75% (30 of 40 | ███ ██ ██████ ██ █████) of patients who were down-titrated to baricitinib 2 mg, and 98% (41 of 42, ███ ███ █████ ██ █████) of patients who remained on baricitinib 4 mg.

Study BRAVE-AA2: Up-Titration Cohort

At week 52, ███ patients who were originally randomized to baricitinib 2 mg were considered nonresponders and their baricitinib dose was titrated up to 4 mg. At week 76, after 24 weeks of titration up treatment on baricitinib 4 mg, ███ of ███ patients ███████ ███ ██ ██████ ██████ attained a SALT score of 20 or less.

ClinRO Measures for EB and EL Hair Loss

At week 52, 39.4% and 27.9% of patients receiving baricitinib 4 mg (N = 188) and baricitinib 2 mg (N = 136), respectively, attained a ClinRO measure EB hair loss score of 0 or 1 (with ≥ 2-point improvement from baseline through week 52 among patients with a score of ≥ 2 at baseline) in Study BRAVE-AA1 (Figure 5). Similarly, 49.7% and 16.3% of patients receiving baricitinib 4 mg (N = 161) and baricitinib 2 mg (N = 104), respectively, attained a ClinRO measure EB hair loss score of 0 or 1 (with ≥ 2-point improvement from baseline through week 52 among patients with a score of ≥ 2 at baseline) at week 52 in Study BRAVE-AA2 (Figure 5).

At week 52, 40.7% and 21.6% of patients receiving baricitinib 4 mg (N = 167) and baricitinib 2 mg (N = 111), respectively, attained a ClinRO measure EL hair loss score of 0 or 1 (with ≥ 2-point improvement from baseline through week 52 among patients with a score of ≥ 2 at baseline) in Study BRAVE-AA1 (Figure 6). Similarly, 50.7% and 30.3% of patients receiving baricitinib 4 mg (N = 140) and baricitinib 2 mg (N = 89), respectively, attained a ClinRO measure EL hair loss score of 0 or 1 (with ≥ 2-point improvement from baseline through week 52 among patients with a score of ≥ 2 at baseline) at week 52 in Study BRAVE-AA2 (Figure 6).

Figure 5: Patients Attaining ClinRO Measure Eyebrow Hair Loss Score of 0 or 1 Through Week 52 in Study BRAVE-AA1 and Study BRAVE AA2

ClinRO = clinician-reported outcome; NRI = nonresponder imputation.

Note: Bars represent 95% confidence intervals. NRI was applied to missing data (prespecified analysis).

Source: Kwon et al. (2023).³⁹

Figure 6: Patients Attaining ClinRO Measure Eyelash Hair Loss Score of 0 or 1 Through Week 52 in Study BRAVE-AA1 and Study BRAVE AA2

ClinRO = clinician-reported outcome; NRI = nonresponder imputation.

Note: Bars represent 95% confidence intervals. NRI was applied to missing data (prespecified analysis).

Source: Kwon et al. (2023).³⁹

Efficacy (Updated): Up to Week 104 (Data Cut-Off Date of May 2022)

Clinically meaningful scalp hair regrowth was maintained through week 104 in nearly 90% of patients treated with baricitinib 4 mg or baricitinib 2 mg who responded at week 52 (Figure 7).

The proportion of patients attaining complete or nearly complete regrowth of EBs and ELs increased from week 52 through week 104 among week 52 responders (Figure 8).

Figure 7: Proportion of Week 52 Responders Who Attained SALT Score of 20 or Less Through Week 104

BARI = baricitinib; CI = confidence interval; mLOCF = modified LOCF; SALT = Severity of Alopecia Tool.

Note: Data were summarized with mLOCF imputation.

^a These were patients who attained a SALT score of 20 or less at week 52.

Source: Senna et al. (2024).³⁸

Figure 8: Proportion of Patients Attaining Complete or Nearly Complete Regrowth of Eyebrows and Eyelashes Through Week 104

BARI = baricitinib; CI = confidence interval; ClinRO = clinician-reported outcome; EB = eyebrow; eyelash; mLOCF = modified LOCF; SALT = Severity of Alopecia Tool.

Note: Data were summarized with mLOCF imputation.

^a These were BARI 4 mg–treated and BARI 2 mg–treated patients with a SALT score of 20 or less at week 52.

^b This was with improvements of 2 points or more from baseline.

Source: Senna et al. (2024).³⁸

Harms: Up to Week 76 (Data Cut-Off Date of February 2021)

Study BRAVE-AA1: Up-Titration Cohort

Week 0 Through Week 52

The overview of AEs between week 0 and week 52 for the ███ patients who were eligible for inclusion in the up-titration cohort are presented in Table 23. TEAEs were reported for █████ ███ ██ ████ of patients, and there ███ █ ███████ ███ ████████ ████ ████ ████ ██████ ██████

Week 52 Through Week 76

TEAEs were reported for ████ █ ███ ██ ████ of patients who up-titrated to baricitinib 4 mg. Most events were mild in severity. █████ ████ █ ████████ ███ ████████ ████ ████ ████ ██████ █████. ███ ███████ discontinued the study drug and study because of AEs of ██████ ████. There were no deaths in the up-titration cohort.

Study BRAVE-AA2: Randomized Down-Titration Cohort

Week 0 Through Week 52

The overview of AEs between week 0 and week 52 for the 82 patients who were eligible for inclusion in the down-titration cohort is presented in Table 24. TEAEs were reported for █████ ███ ██ ███ of patients, and there were | ████████ ███ ████████ ████ ████ ████ ██████ ██████

Week 52 Through Week 76

During the 24-week period, TEAEs were reported for █████ ███ ██ ███ of patients who remained on baricitinib 4 mg, and █████ ███ ██ ███ of patients who down-titrated to baricitinib 2 mg.

For both treatment groups, most events were mild or moderate in severity. Of the patients who remained on baricitinib 4 mg, |████████ ████████ ████ ███ ███ █ ███████ ████ █████ ███ █ ██████ █████

██ ████████ ████████ for patients who titrated down to baricitinib 2 mg.

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

There were no deaths in the randomized down-titration cohort for all treatment groups.

Study BRAVE-AA2: Up-Titration Cohort

Week 0 Through Week 52

The overview of AEs between week 0 and week 52 for the ██ patients who were eligible for inclusion in the up-titration cohort is presented in Table 25. TEAEs were reported for █████ ███ ██ ███ of patients, and there were | ████████ ███ ████████ ████ ████ ████ ██████ ██████

Week 52 Through Week 76

TEAEs were reported for █████ ███ ██ ██ █████████ of patients who up-titrated to baricitinib 4 mg. Most events were mild or moderate in severity; | ███████ ████████ █ ██████ █████ ██ ████████ ████████ ████ ██ ████████████ █████ ████ ██ █████ ███ ██ ████ There were no deaths in the up-titration cohort in any treatment group.

Table 23: Summary of Harms of Study BRAVE-AA1—Titration Up Cohort

AE = adverse event; BARI = baricitinib; SAE = serious adverse event; TE = treatment-emergent; TEAE = treatment-emergent adverse event.

^aPatients with multiple occurrences of the same event were counted under the highest severity category.

^bThe denominator was adjusted because the event was specific to males; N = 49 (BARI 2 mg group [Study BRAVE-AA1]). In other cases, the denominator was adjusted because the event was specific to females; N = 63 (BARI 2 mg group [Study BRAVE-AA1]).

Source: Study BRAVE-AA1 Clinical Study Report Addendum.¹⁰

Table 24: Summary of Harms of Study BRAVE-AA2— Randomized Titration Down Cohort

AE = adverse event; BARI = baricitinib; SAE = serious adverse event; TE = treatment-emergent; TEAE = treatment-emergent adverse event.

^aPatients with multiple occurrences of the same event were counted under the highest severity category.

^bThe denominator was adjusted because the event was specific to males; N = 28 (BARI 4 mg group [Study BRAVE-AA2]). In other cases, the denominator was adjusted because the event was specific to females; N = 54 (BARI 4 mg group [Study BRAVE-AA2]).

Source: Study BRAVE-AA2 Clinical Study Report Addendum.¹¹

Table 25: Summary of Harms of Study BRAVE-AA2 Extension — Titration Up Cohort

AE = adverse event; BARI = baricitinib; SAE = serious adverse event; TE = treatment-emergent; TEAE = treatment-emergent adverse event.

^aPatients with multiple occurrences of the same event were counted under the highest severity category.

^bThe denominator was adjusted because the event was specific to males; N = 27 (BARI 2 mg group [Study BRAVE-AA2]). In other cases, the denominator was adjusted because the event was specific to females; N = 57 (BARI 2 mg group [Study BRAVE-AA2]).

Source: Study BRAVE-AA2 Clinical Study Report Addendum.¹¹

Harms (Updated): Up to Week 104 (Data Cut-Off Date of May 2022)

A pooled safety analysis was conducted, including data for all patients receiving baricitinib during Study BRAVE-AA1 and Study BRAVE-AA2 up to the data cut-off date (May 24, 2022, for Study BRAVE-AA1 and May 10, 2022, for Study BRAVE-AA2); this included the 36-week, placebo-controlled periods and long-term extension periods.^38,52 Data were collected for 1,303 patients reflecting 2,218 patient-years of exposure (mean of 621.7 [SD = 303.8] days). The majority of TEAEs (93.2%) in the all-baricitinib dataset were mild to moderate in severity and the incidence rates (IRs) per 100 patient-years of SAEs (2.9) and treatment discontinuations because of AEs (1.9) were low. There was 1 opportunistic infection of multidermatomal herpes zoster (IR < 0.1), 44 cases of herpes zoster (IR = 2.0), 1 positively adjudicated MACE (a myocardial infarction [IR < 0.1]; this was in the baricitinib 2 mg group), 2 cases of pulmonary embolism (IR = 0.1; these cases were in the baricitinib 2 mg group), 4 malignancies other than nonmelanoma skin cancer (IR = 0.2; 2 malignancies were in the baricitinib 2 mg group and 2 malignancies were in the baricitinib 4 mg group), and 1 gastrointestinal perforation (IR < 0.1; this was in the baricitinib 4 mg group). The myocardial infarction was the same case that occurred in the placebo-controlled, 36-week treatment period. Meanwhile, in the 2 cases of pulmonary embolism, both patients had multiple risk factors for venous thromboembolism while 1 of these patients also had an identified deep venous thrombosis. No deaths were reported.

Critical Appraisal

Internal Validity

Both Study BRAVE-AA1 and Study BRAVE-AA2 extension studies were limited by their noncomparative design. At time points after 36 weeks, there remained no randomized comparison to placebo, challenging causal interpretations. Although the patients and investigators remained blinded to the assigned interventions, there is still the possibility that patients may be able to infer treatment assignment because of differences in efficacy (relative to placebo during the double-blind treatment phase). As such, there may be a risk of bias in the reporting of efficacy outcomes that required some level of subjective judgment by the evaluators (e.g., ClinRO), and harms outcomes, although the magnitude of bias cannot be predicted. It is unlikely that bias would be introduced for the SALT response since it is measured objectively. Finally, missing information such as pooling strategies for the 104-week results constrained a robust critical appraisal; hence, firm conclusions cannot be drawn on its long-term efficacy and safety.

External Validity

Since both Study BRAVE-AA1 and Study BRAVE-AA2 included rollover patients consistent with their characteristics at entry in the study, it is reasonable to expect similar limitations to the generalizability of the study results are relevant to the long-term extension phase. Further, some outcomes that are important to patients (e.g., HRQoL, anxiety, depression) could not be evaluated against a placebo control beyond the 36-week, double-blind treatment phase because of discontinuation of the placebo in nonresponders. As such, there is limited evidence for the effect of baricitinib 2 mg or baricitinib 4 mg on these outcomes for time points after 36 weeks (including for patients who up-titrated or down-titrated). Despite longer follow-up for harms, some rare harms (e.g., malignancies) may still not be fully captured.

Indirect Evidence

No indirect comparative evidence was submitted by the sponsor. The sponsor noted that before the regulatory approval of baricitinib for severe AA in Canada, the standard of care included off-label therapies and nonpharmacological options. The sponsor further noted that the pivotal trials of baricitinib were placebo-controlled and given that no approved comparator drugs were available at the time of the phase III clinical development conduct, there is no indirect comparative efficacy evidence to present in this section.

Studies Addressing Gaps in the Systematic Review Evidence

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

The sponsor submitted 1 study to provide evidence to address gaps in the systematic review evidence. The single-arm retrospective study by Tang et al. (2024)¹² reported on baricitinib treatment among adult patients older than 65 years, as these older patients were excluded from the pivotal trials. Studies assessing the effects of baricitinib in patients in Japan (Numata et al. [2024])¹⁴ and in Spain along with evidence for adjuvant therapy among nonresponders (Moreno-Vilchez et al. [2024])¹³ were also submitted. However, our review team did not consider these studies to have addressed any identified gaps in evidence for the purpose of informing the reimbursement recommendation of baricitinib in the Canadian context. Table 26 summarizes the gaps in pivotal trials identified by the sponsor and the sponsor-submitted studies that could potentially address these gaps.

Table 26: Summary of Gaps in the Systematic Review Evidence

AA = alopecia areata; RCT = randomized controlled trial; SALT = Severity of Alopecia Tool; SD = standard deviation.

Note: Sources: Tang et al. (2024),¹² Numata et al. (2024),¹⁴ and Moreno-Vilchez et al. (2024).¹³ Details included in Table 26 are from the sponsor’s Summary of Clinical Evidence.⁹

Description of Studies

Table 27 summarizes the Population, Intervention, Comparator, Outcome of the 3 sponsor-submitted observational studies included in this submission.

Tang et al. (2024)

The study investigators retrospectively reviewed the records of 14 patients aged 65 years or older with moderate-to-severe AA treated with baricitinib between April 2020 and September 2023. Hair loss before and after treatment was scored using the SALT tool based on standardized scalp photographs, with a SALT score of more than 20 representing moderate-to-severe AA. Patients were reviewed every 3 months to 4 months while on treatment.

The objective of this study was to evaluate the efficacy and safety of baricitinib in the treatment of moderate-to-severe AA in this older adult population (aged 65 years or older). All participants in the study received baricitinib. The mean dose of baricitinib was 3.52 (1.05) mg, with a mean (SD) duration of treatment of 18.5 (11.9) months. The study measured reduction in the mean SALT score from baseline in patients aged 65 years or older with moderate-to-severe AA.

Numata et al. (2024)

The study investigators retrospectively reviewed medical record data, including age, sex, age at AA onset, duration of the current episode of AA, AA subtype, prior treatments, nail involvement, trichoscopic findings (hair and scalp dermoscopy), and complications, among 95 patients in Japan with moderate-to-severe AA who were treated with baricitinib. Patients who had lost approximately 50% of their scalp hair for longer than 5 months were enrolled in the study, and all participants received baricitinib 4 mg between June 2022 and February 2023. The study measured the percentage of patients in the entire cohort who attained a SALT score of 20 or less at week 12, week 24, and week 36.

Moreno-Vilchez et al. (2024)

The study investigators retrospectively reviewed medical records for patients with severe or very severe AA who were treated with baricitinib 4 mg once daily for at least 24 weeks with participants on average being treated for 55.78 (range, 24 to 120) weeks. The study reported mean SALT scores as efficacy outcomes for the study.

Table 27: Details of Studies Addressing Gaps in the Systematic Review Evidence

AA = alopecia areata; SALT = Severity of Alopecia Tool.

Sources: Tang et al. (2024),¹² Numata et al. (2024),¹⁴ and Moreno-Vilchez et al. (2024).¹³

Results

Baseline Characteristics

Tang et al. (2024)

Fourteen patients (3 males and 11 females) with a mean age of 72.1 (range, 65 to 87) years were included in the study. The median baseline SALT score was 59.8 (SD = 34.4) and the mean duration of the current episode of AA at baseline was 41.1 (SD = 59.8) months. Previous treatments included topical corticosteroids (n = 5), topical calcineurin inhibitors (n = 1), intralesional corticosteroids (n = 4), systemic corticosteroids (n = 5), diphenylcyclopropenone (n = 2), azathioprine (n = 1), cyclosporine (n = 1), methotrexate, (n = 1), and oral tofacitinib (n = 5). Concurrent treatments included topical corticosteroids (n = 1), intralesional corticosteroids (n = 2), and low-dose oral minoxidil (n = 13).

Numata et al. (2024)

The mean patient age was 38.7 (range, 18 to 65) years. The mean age at AA onset was 19.3 years where SD is not reported and the mean duration of the current episode of AA at baseline was 7.8 years where SD is not reported. The median baseline SALT score was 89.2. Previous treatments included topical corticosteroid (n = 2), intralesional corticosteroid (n = 4), oral corticosteroid (n = 3), dupilumab (n = 5), contact immunotherapy (n = 44), and phototherapy (n = 10). Twenty-seven previously untreated patients were enrolled.

Moreno-Vilchez et al. (2024)

The study involved 36 patients treated with baricitinib for an average of 55.78 (range, 24 to 120) weeks. The median baseline SALT score was 94.81 (SD = 9.759). Previous treatments included oral corticosteroid (n = 29), topical corticosteroid (n = 25), intralesional corticosteroid (n = 24), cyclosporine (n = 22), methotrexate (n = 19), and diphencyprone (n = 4).

Exposure to Study Treatments

Tang et al. (2024)

Treatments given in conjunction with baricitinib included corticosteroids (topical, n = 1; intralesional, n = 2) and low-dose oral minoxidil (n = 13). One patient applied a super-potent topical corticosteroid to his scalp in conjunction with baricitinib treatment. One patient with a baseline SALT score of 26.6 who still had residual bald patches on the scalp (with a SALT score of 15) despite baricitinib therapy received intralesional triamcinolone acetonide injections, resulting in a further reduction in the SALT score to 0. One 65-year-old male received intralesional corticosteroid injections in the EBs because of persistent gaps, resulting in complete hair regrowth.

Numata et al. (2024)

Information was not provided.

Moreno-Vilchez et al. (2024)

Out of the total cohort, 9 (25%) patients received combination therapy: 7 patients were treated with a combination of baricitinib and methotrexate (10 mg to 15 mg per week) while 2 patients received a combination of baricitinib and oral minoxidil (1 mg to 2 mg per day).

Efficacy

Tang et al. (2024)

After a mean (SD) duration of 18.5 (11.9) months, a 72.0% reduction in the mean SALT score from baseline was observed. Moreover, 11 (78.6%) patients attained a SALT score of less than 10 after a mean duration of 18.6 months where SD is not reported.

Numata et al. (2024)

The percentage of patients in the entire cohort who attained a SALT score of 20 or less at week 12, week 24, and week 36 was 6.4% (6 of 94 patients), 35.4% (28 of 79 patients), and 46.7% (21 of 45 patients), respectively. The complete response rate (a SALT score of 0) at week 24 and week 36 was 1.3% (1 of 79) of patients and 6.7% (3 of 45) of patients, respectively. Among these, the percentage of patients with patchy AA, alopecia totalis, alopecia universalis, or the ophiasis subtype of AA who attained a SALT score of 20 or less at week 36 was 75.0%, 48.0%, and 75.0%, respectively. The percentage of patients with a current AA episode of less than 4 years’ duration and a SALT score of 20 or lower at week 36 was greater than that of patients with an AA episode of 4 years’ duration or longer.

Moreno-Vilchez et al. (2024)

In the study, 58.8% of patients attained a SALT score of 20 or less at week 24. The response continued for 52 weeks, with 66.6% of patients classified as responders. Additionally, the study compared the SALT scores between patients treated with monotherapy and those who received adjuvant treatment.

Harms

Tang et al. (2024)

Adverse effects of baricitinib included the reactivation of herpes zoster (n = 1), elevated creatine kinase (n = 1), and grade 2 neutropenia (n = 1). One patient required a reduction in the dose of baricitinib because of grade 2 neutropenia. No cases of venous thromboembolism, MACE, or malignancy were reported.

Numata et al. (2024)

Infectious complications occurred in 6 patients during the initial 12 weeks. Herpes simplex and COVID-19 (severe acute respiratory syndrome coronavirus 2) occurred in 1 patient and 5 patients, respectively. No other complications occurred during the entire 36-week course.

Moreno-Vilchez et al. (2024)

Three patients discontinued baricitinib because of inadequate treatment response: 2 patients at week 52 and 1 patient at week 76. Additionally, 1 patient had temporary lymphopenia with methotrexate treatment.

Critical Appraisal

Internal Validity

Limitations of the 3 studies include their single-arm, retrospective designs and small sample sizes. Moreover, many patients in the studies were treated with concomitant treatments. Without a randomized comparison group, it is not possible to attribute the observed effects to baricitinib with certainty. Information such as treatment exposure and concomitant treatments in Numata et al. was not reported. Furthermore, there is a risk that what has been reported was among multiple analyses of the data because of lacking a priori protocols for these 3 studies. Lastly, there is also a risk of selection bias since information about how patients were selected for analysis was not presented.

External Validity

Both Tang et al. and Numata et al. included patients with moderate-to-severe AA; however, patients with moderate AA would not be candidates for baricitinib treatment in Canada. The results of these studies may not be generalizable to patients with severe or very severe AA, which may be more difficult to treat compared with moderate AA. The study by Numata et al. included patients exclusively from Japan whereas the study by Moreno-Vilchez et al. included patients exclusively from Spain (in 2 centres). It is uncertain whether results from small samples of patients treated in these countries would be generalizable to patients living in Canada, given the potential for differences in standard of care in these countries. Although the studies provided additional evidence among patients not included in the pivotal trials, many outcomes that are important to patients (e.g., HRQoL, EB and EL hair loss, anxiety, depression) were not measured or reported.